PDA

View Full Version : Elusive P0172 code - diagnositic experts opinion needed


Pages : [1] 2 3

cboss
11-11-2008, 01:52 PM
First, let me say that if I can find the solution to my problem, it likely will help many, many others. I have searched the web extensively and the P0172 code (bank 1 too rich) appears to be one of the most troublesome codes to solve and I have rarely ever found someone say they found a solution to their problem.

Solving riddles is my area of expertese (I am computer programmer) and there is always a reason for a problem. I just have to find it.

I have been dealing with the P0172 code coming and going for a couple weeks now, since I purchased the car.

It is a 1996 Saturn SL2 (DOHC).

I have replaced the following:

- ECTS
- MAP sensor
- TPS
- AIC
- spark plugs
- spark plug wires
- air filters
- O2 sensor (upstream)
- Catalytic Converter

I solved a problem with the fuel pressure regulator (vacuum leak) and the fuel pressure is now 40 to 41 psi (key on) and 35 to 36 psi when running (at idle), well within the proper range.

Using a OBD II scanner and testing the live data, I am working on finding a way of bring the short term fuel trim down to normal. Initially, the SHFT was in the range of 0 to -20 and the LTFT was set at -21. Well too rich.

The weather changed this week and it got very, very cold (in the 30's at night, 50's day time). before this it was in the 50's at night and 70's daytime.

The STFT is now going as high as -40 or higher.

The strange thing is, that when I hit the gas pedal quickly (rev it), the STFT drops back to close to zero and stays around -5 to 0 for a second or two and then even at a higher RPM is starts progressing climbing to over -30 or -40.

The STFT was never this bad and it only started the last day or so.

I pulled all the plugs and they are all black as coal now. Since I did a fuel pressure leak down test (pressurize the fuel pressure at key on [40 psi], then key off, then wait ten minutes to see if it drops) and the fuel pressure doesn't drop at all, all four plugs are black, so no specific fuel injector is worse than any others, I assume the fuel injectors are reasonably OK.

I cleaned the EGR valve and then tested it, by puttinga thin piece of steel over the gasket and reinstalling the EGR valve. This would test to see if it leaked at idle and no change in the STFT or idle problems.

This rules out the fuel pressure regulator, the fuel pressure, the injectors, the EGR valve and the O2 , MAP, TPS and AIC sensors (replaced).

Today I plan on replacing the Evap Canister Solenoid. That would definitely cause a rich condition if it were stuck open.

Any other suggestions ?

Could the timing chain slip and cause this problem ?

How about a leak in the exhaust or intake manifolds ?


I will figure this out one way or another.

OldNuc
11-11-2008, 03:10 PM
Does your scanner show the timing? The other thing that causes a P0172 is a restricted exhaust. Muffler or CAT. I would check the fuel pressure if you have a tester.

cboss
11-11-2008, 03:41 PM
I did test the fuel pressure (as noted above) and it is fine. I have both a OBD II scanner which can read live data and a fuel pressure gauge.

The STFT is actually better when the engine gets warm. It is around -5 to -17 and the LTFT ranges from -14 to -21.

When the engine is cold, though the STFT is quite high (in the negative), as much as -40.

I was wondering about the resonator and the muffler ?

When I got the car a few weeks ago, there was no CAT installed (I didn't notice this when I purchased it). I immediately had a new CAT installed by a local muffler shop (had the dealer I bought it from cough up the $150 to have it installed).

Since the car obviously had been run without a CAT for some time (car originated in N.C. and I think they can do that there, or so I heard) and was running very rich (the original plugs were black as coal), it is quite possible both the resonator (silencer) and the muffler are clogged up.

When I put my hand near the tail pipe, a lot of black flakes stick to my hand.

rc1488
11-11-2008, 04:22 PM
The first thing that comes to mind is that some '96 PCM's have had timing issues. However its usually a single cylinder or just a random miss;not all 4 in your case. hmmmm....

SparkPlug type?
ECTS Connector nice and clean?
Is the system going into closed loop ?
Stupid question but always important: Was the rear O2 sensor installed and connected?

How does the power of the car feel? You can remove the front O2 sensor and drive the car around. See if that show an improved feeling of power response from the engine. If it does, then your guess on a clogged exhaust may be on target

cboss
11-11-2008, 05:32 PM
Before I came to the forums, I had already installed Platinum 2's.

I plan on swapping them out later (ie. NGK coppers) later.

Besides, as long as the engine is running rich, it is messing up the plugs, so it might as well be the platinums while I am working on the problem.

I don't think the platinums would cause it to run rich though (or am I worng about that). The previous owner had also installed platinum 2's, which is what I took out of it when I got the car.

ECTS is new and the connector appears to be a new one (wire was spliced).

Yes, the system goes into closed loop relatively quickly (with a few minutes, maybe less).

The rear O2 sensor is installed and connected (it shows up on the canner in the .700 to .900 range).

When the muffler shop installed the new CAT (a universal one), they torched out the straight pipe the previous owner had installed. He got awful close to the rear O2 sensor, so I don't know it the heat would damage it. The guy said it was welded in, but I looked and it didn't look welded. It may a spot weld on one side I can't see though.

Aside from the idle being not quite right, the engine running rich and the P0172 code, it acfually drives very well. There is lots of power, pickup and speed out of it. It may hesitate just slightly once in ahile, but all in all it is not noticable. I like how it drives. More power and pickup than my Geo prizm.

OldNuc
11-11-2008, 05:38 PM
The fuel trim numbers you are getting DO NOT indicate a P0172 condition. I am fast suspecting an other failure. P0172 will set when the averaged long term idle and cruise fuel trim cells are 101 and the current short term fuel trim cell is below 128 for 3 seconds.

Fuel trim is not your problem.


Evap canister purge is about all there is. Beforre ripping the purge system apart check the average O2 readings for a while. Try as live data on the front O2 sensor.

What brand is the front O2 sensor? BOSCH universal can cause problems....

I think I sent you this chart before but if not, here it is.

cboss
11-11-2008, 09:56 PM
The O2 sensor is a Bosch Universal.

It was only $17.

You have to cut the old O2 sensors wire off and attach it to the new sensor (crimp connector).

The O2 sensor appears to work fine according to my scanner.

What specific problems do the Bosch sensors have ?

fdryer
11-11-2008, 10:21 PM
By chance is there possibly a 'mod' wired in series to the intake air temperature (IAT) sensor? Some think that these resistors or metal boxes being sold as 'chips' are to modify the engine computer to add more power when in actuality all it does is fool the PCM into enrichening the air/fuel mixture. The right resistor added in series to the air temperature sensor will be interpreted by the PCM as very cold air temps and result in a richer mixture than called for without this 'mod'. The same can be done to the ects wiring if done there. You'll just have to look carefully for anything unusual taped up on the wiring harness, perhaps some splicing neatly taped into the harness. Just a thought...

OldNuc
11-11-2008, 11:01 PM
The O2 sensor is a Bosch Universal.

It was only $17.

You have to cut the old O2 sensors wire off and attach it to the new sensor (crimp connector).

The O2 sensor appears to work fine according to my scanner.

What specific problems do the Bosch sensors have ?

Very slow dynamic response. The switching time is very close to the DTC limit right out of the box. The Denso and NGK-NTK sensors are much faster switching time and are thus more responsive. All you need is a 3 second spike to set the code. BOSCH manufactures many very good components, O2 universal O2 sensors do not seem to be one of them. They do work for some people though. It is "interesting" that you are setting a code that is based on long and short fuel trim and you are nowhere close to the trigger point. something is broke and hiding at he same time. As fdryer suggested check the cold agreement between the ECTS and IAT sensors. Do this before you start the car, they should be within 3 degrees. No point to crawl the wiring, let the scanner do its job.

delsydsoftware
11-12-2008, 05:47 AM
Since the car obviously had been run without a CAT for some time (car originated in N.C. and I think they can do that there, or so I heard)


Actually,running without a catalytic converter in a modern car is against federal law.

http://catalyticconverter.org/law/index.htm

OldNuc
11-12-2008, 08:25 AM
The Platinum plugs can cause all kinds of problems. Change the plugs. The O2 sensor will survive 800 degrees or so. I doubt that the rear sensor is fried. The rear sensor only sets codes for CAT performance, no control function.

cboss
11-12-2008, 08:56 AM
The Evap Purge solenoid I ordered hasn't come in yet, but once I get it and install it, I'll report back on the results.

Maybe I should change out the plugs right away to see what results I get from that.

Also I read some more (in the Haynes OBDII book) about PCM monitors. I didn't realize how important the monitors are (they test the engine). I have disconnected the battery recently so the monitors data were lost, so I need to drive the car some more so they all complete their tests and see what the results are. You have drive in all sorts of conditions for all the monitor tests to complete. Maybe they will give me some clues.

OldNuc
11-12-2008, 09:29 AM
All the monitors say is pass/fail and complete yes/no.

You can skip the electrical load test part of step #2

http://www.saturnfans.com/photos/data/12963/thumbs/OBD-II_Drive_Cycle.jpg (http://www.saturnfans.com/photos/showphoto.php?photo=48942)

cboss
11-12-2008, 05:14 PM
I can get a Denso universal O2 sensor for $26 at AutoZone.

Would it be worth swapping out the Bosch universal for a Denso universal ?

I bought some Champion Copper Plus spark plugs (# 71RC12YC) today (the NGKs not in stock and they weren't sure whether they were copper).

I plan to swap out the Bosch Platinums tomorrow.

I also got the Evap Purge Solenoid today (Wells brand, # PV161). I plan on installing it tomorrow too.

OldNuc
11-12-2008, 05:46 PM
All NGK are copper. Universal O2 sensors are a one size fits all device. Saturns do not play well with universals, of anything. They are very picky cars.

Motorcycle shops carry NGK plugs. Champion plugs have a narrow heat range when compared to the NGK and tend to foul out or burn the insulator.

I would order a set from RockAuto along with the correct Denso O2 sensor.


Based on the fuel trim numbers there is no valid reason for a P0172. Normally you can get away with sticking in an oddball part but when looking for "what is broke" it is not a good idea.

cboss
11-13-2008, 12:09 PM
OK, I hadn't installed the Champion plugs yet (still in box), so O I brought them back to Advance and ordered the NGK (BKR5ESA-11) plugs (same price as Champions). They will be in tomorrow.

I read a lot about 'heat range" of plugs today and I can see why the Platinums could cause some problems.

I also jacked up the car to start replacing the Evap Purge Solenoid and guess what ? The vacuum line was disconnected. It had a tendency to slide off the connectors on the solenoid (oil tends to drip down from when adding oil to crankcase right on top of the solenoid).

I am going to replace the solenoid for good measure though. It's a tight fit to get in there. RichPin recommends removing the starter to do it (in his video), but I am going to try without doing that.

New plugs (NGK) and a new Evap Solenoid should make a difference. I will report back.

Just connecting the vacuum line improved the idle, but not completely. The solenoid may still be sticking open and the plugs are very dirty (black).

Going in the right direction though.

cboss
11-13-2008, 03:47 PM
My error.

I was thinking of a different richpin video (which mentioned removing the starter). The one about the evap solenoid, didn't remove the starter to replace it.

I got the evap purge solenoid installed. I had to clean the rubber vacuum line connector (connects the hard plastic vacuum line to the evap solenoid), since it was very greasy (oily). The evap solenoid is right underneath the area were you add oil to the engine so any overflow drips down right on it. The rubber connector wouldn't stay on very good. After I lcean ed it real good, it now stays on tight.

The engine runs a little better now, but I assume I have to run it awhile for the computer to fine tune itself to no more vacuum leak.

When I get the NGK plugs tomorrow, I'll install them and that should improve things over the platinum plugs.


I am wondering if I should I should replace the Bosch (universal) O2 sensor with a Denso O2 sensor (non-universal) ?

Would that effect how it runs at all ?

OldNuc
11-13-2008, 06:53 PM
Platinum plugs may have any heat range that can be engineered into the plug and if they have a copper core it will be a broad er range that the same plug without the copper core. the main reason the platinum and any other fine wire center electrode plug tends to fail in a waste spark system is that the resistance of the plug is dependent on the polarity of the applied voltage. Current flows at a lower resistance from the point to ground but the resistance is much higher when attempting to establish a current flow from the ground to the point. The coil is fired in the same polarity every time and the output oscillates between positive going and negative going. If you have a unidirectional plug then one spark will be delayed until the waveform swings to the proper low resistance polarity. This results in 1/2 of the cylinders having an effectively retarded timing. Its not large but its there.

I would replace the O2 sensor with a model specific one.

Before I discovered that my P0172 was being caused by high fuel pressure I replaced the entire exhaust system. This can be a difficult problem to run down and what makes yours so hard to find is that the fuel trim is not showing it and this code is only triggered from fuel trim.


Is the intake manifold full of black oily gunk?

cboss
11-13-2008, 10:06 PM
Fixing the vacuum leak to the Evap Purge Solenoid and installing the new Evap Purge Solenoid has improved things a great deal. The fuel trim is much better now. Almost there!

When I change the plugs, that should make a big difference too.

I think the P0172 code problem is solved now.

So far the major fixes that made the most difference have been:

- New fuel filter (old one very, very clogged)
- new air filter (old one also very, very clogged and dried up and hard)
- vacuum leak to fuel pressure regulator found and fixed (didn't notice it at first, but the vacuum line just broke one day at the rubber connector when I was working on it)
- vacuum leak (not connected actually) to Evap Purge Solenoid (new solenoid installed as well).

The vacuum line to the Evap Solenoid, the rubber connector was oily and slick, so it would not stay put on the solenoid. I took it off and clean it good, so the rubber grabbed when connected and won't slide of again.

The fuel trim values on the scanner are very useful, since they give me an idea of how well the engine is running. They are quite reasonable right now, but I hope for better.

What effect will the Bosch O2 sensor have on how the engine runs ?

cboss
11-13-2008, 10:14 PM
Just a note to those who are dealing with the P0172 code:

A definite thing to check which is easily overlooked:

- Jack up the front end of the car.
- crawl underneath and examine the condition of the vacuum line to the Evap Purge Solenoid and make sure it is still connected.

The Evap Purge Solenoid is right underneath (vertically) the area where you add oil to the engine. Since the Saturns burn a lot of oil, you are adding oil a lot and the valve cover tends to pool the oil when adding it so it is easy for it to overflow when filling and it will drip right down on top of the Evap Purge Solenoid.

The effect this has is that the vacuum line gets oily after time and the rubber connector can become like a greased pig (very slippery) and easily pop off. It needs to be cleaned good so the rubber grabs the connector properly. I took it off and used dish detergent and hot water to clean it.

This little oversight will wreak havoc on the engine (big vacuum leak).

OldNuc
11-13-2008, 10:40 PM
You may get away with the BOSCH sensor. If you do not get any more P0172 then you got away with it. The next item is fuel mileage. I have found that they have a slight negative impact on mileage. And how much is dependent on what kind of driving you do.

cboss
11-14-2008, 04:04 PM
Well, sadly I was too hopeful about having fixed the problem.

The P0172 code has returned after a day of driving.

I need to jack up the car again to make sure the vacuum line didn't slip off the Evap Purge Solenoid again. If it did, then that is the problem and I need to make sure it doesn't slip off.

If the vacuum line is OK, then I have a real problem and no solution.

I can't afford to just keep changing parts either.

First learnt lesson, be careful what brand parts you buy.

Now where to start ?

The suspect systems to consider:

- O2 Sensor (Bosch installed probably garbage)
- clogged exhaust (installed new CAT, but maybe the resonator and muffler clogged)
- fuel injectors (they don't leak, but that doesn't mean they don't have a problem spraying properly)

OldNuc
11-14-2008, 04:13 PM
If you decide to buy a new O2 sensor, get the NGK-NTK from either Summit Racing or sparkplugs.com and as to the injectors shipping them off for a flow balance and cleaning to witchhunter is a good idea. http://www.witchhunter.com/

cboss
11-14-2008, 06:02 PM
Some part questions:
(SL2 1996)

What brand O2 sensors are acceptable ?

Which brand O2 sensor is the best (or the exact same as original) ?

I will swap out the plugs soon and install NGK's.

I am curious though.

Are the Bosch Platinum 2's a resistor or a non-resistor type plug ?

How about the NGK's ?

cboss
11-14-2008, 06:19 PM
Please correct me if I am wrong here!

The biggest clue I have is the condition of the spark plugs.
The old plugs (Bosch Platinum 2's) were all carbon deposited.

The new Bosch Platinum 2's have less than 800 miles on them and they are all black as coal (carbon deposits).

This would indicate to me, that whatever is wrong is most likely not cylinder specific (ie. a bad injector) or one of the coils bad.

The problem effects all the cylinders equally.

I ruled out the fuel pressure, since I tested this with a fuel pressure gauge.

I fixed the vacuum leaks.

Would a lousy O2 sensor (ie. Bosch) be enough to cause this problem ?

Would the Platinum plugs be enough to cause this problem ?

The CAT is new, so how does one test the resonator and muffler to see if they are plugged up ?

cboss
11-14-2008, 07:04 PM
OldNuc you have been very helpful.

I am new to Saturns so this is all a learning curve for me (my last car was a 1990 Geo Prizm and I have never changed a single sensor on it and it now has 330,000 miles on it).

I saved the old O2 sensor and just checked it to see what brand it is. There were no markings, but I wen to Rock Auto, Advance and AutoZone web sites and looked at the photos of all the O2 sensors. The old sensor is a Bosch too ! No doubt. Not a universal, but definitely a Bosch.

The car obviously had fuel richness problems when I got it, since all the spark plugs were black as coal (dry black powder). The car had Bosch Platinum 2' plugs and a Bosch O2 sensor.

I installed new Bosch Platinum 2 plugs and a Bosch (universal) O2 sensor, not knowing they would be a problem.

Could the combination Bosch O2 sensor and Bosch Platinum plugs be a nightmare combination for this car ?

I took OldNuc's advice and got some NGK plugs (not installed yet). I think getting an NGK O2 sensor may also necessary. OldNuc suggested sparkplugs.com.

Is there any other places to order one (NGK O2) at a better price ?

OldNuc
11-15-2008, 01:39 AM
If you find one, let me know.:D The NGK-NTK O2 sensor is a new item in the line for USA so many of the normal outlets have not picked it up yet.

cboss
11-15-2008, 01:58 PM
The NGK O2 sensor goes for about $51 to $54 at the places OldNuc mentioned (ie. sparkplugs.com, etc.).

RockAuto.com has the Denso O2 sensor for about $35, but shipping is about $12 (their shipping is a bit high).

AutoZone has the denso O2 for $68.

Is Denso as good as NGK's ?

OldNuc
11-15-2008, 02:26 PM
I had a Denso in the motor and swapped it out for the NGK-NTK while easter egging a problem and the Denso did seem to function properly. The NGK does come with a longer lead and the clip to mount the [lug on the fan shroud in the hole provided in the shroud. Other than that there did not seem to be any noticeable difference. However, I did not run any of the performance tests. The Denso is the lowest cost overall as summit has a flat shipping of 9.99 and Sparkplugs.com is about as bad as any of the others. Probably the way to do it is to get an other set of plugs, wires and anything else you will need in the reasonable future to even out some of that shipping cost.

I replace the wires when I replace the plugs. I figure that a set of wires is good for 3 to 4 years at the most.

cboss
11-19-2008, 05:50 PM
The P0172 code came back on today and won't go away. I can clear it but it comes back immediately.

I found a salvage yard that had a 1995 SL1 or SC1 (not sure which). I checked out rockauto.com to find out what parts will fit my 1996 SL2 and the O2 sensor and fuel pressure regulator will fit (plus a number of other parts).

I got the O2 sensor and the fuel pressure regulator.

The FPR looks like OEM.

The O2 sensor is a direct fit (not universal) and it looks like the this:

http://info.rockauto.com/SMP/SMPDetail2.html?SG418.jpg

The sensor has the following numbers on it:

987 ND 065500-6410

I can't find this anywhere online to figure out what brand it is or whether it is the OEM version. The closest I can find as far as how it looks (photos on rockauto.com) is either a Standard or an Airtex

Does anyone know what brand this is ?

I installed the NGK plugs yesterday. It ran a little better, but after about 120 miles the SES light came on again (P0172). I'll run it a little longer and then pull the plugs to see their condition. Since the P0172 code is still here, I expect they will all be black again.

I have already spent too much money on the car so far and I can't keep buying new parts. If I can swap out used parts until I find out what makes a difference, then I can then buy a new part for what solves the problem.

cboss
11-19-2008, 10:58 PM
Tomorrow I plan to install the other O2 sensor I got from the salvage yard. I'll post my results when I do it.

OldNuc
11-19-2008, 11:45 PM
Using J/Y parts for trouble shooting is a very good idea. This code is very difficult to get rid of for many reasons. Going after fuel pressure is also good as a high fuel pressure will overwhelm the injectors and make the engine run rich.

the P0171/P0172 are the hardest codes to clear up.

I have no idea who made the O2 sensor but it does look stock OEM.

cboss
11-20-2008, 12:12 AM
I have a fuel pressure gauge and have tested it and the fuel pressure (at idle) is 36 psi which is the maximum of the allowable range. The J/Y FPR I got may produce a lower pressure so I'll see if that makes a difference.

The J/Y O2 sensor does look better quality and is definitely not a bosch. I'll be interested in seeing if it makes a difference.

cboss
11-20-2008, 12:28 AM
One other possibility to consider:

Is it possible, that the timing chain is off by just one tooth and could that cause the car to run rich (P0172) ?

If so, how does one test to see if the timing chain is "on the money" (meaning perfectly aligned) ?

OldNuc
11-20-2008, 10:01 AM
You take the cam cover off and crank the engine over until the top mark lines up with the chain plate and then look at the crank marks, they should also be in alignment. This can take a deal of turning to get the top lined up. Usually if you are a tooth off the timing is so far out that it is immediately noticeable.

silverwing314
11-22-2008, 10:12 AM
Hey I might not be a diagnostic expert, but I'll share some info I learned from my recently acquired 1996 Saturn SL1. It had the P0172 code with 164,000 mi and so definitely needed a little tuneup. I did the ngk spark plugs, spark plug wires, ects, thermostat, egr cleaning, and pcv valve. All that learned about from this excellent source. Afterwards, the code cleared.

Unfortunately, it has returned intermittently after around 200-300 miles depending on driving style. What I've been done to get rid of the code when it pops up is clean the pcv valve with throttle body cleaner. My plans for a permanent solution would be to buy an OEM pcv valve and troubleshoot my oil consumption problem/potentially restrictive exhaust which may be gumming up the pcv valve.

I don't recall reading the condition of your pcv valve, but you should hear a rattle when you shake it if it is still good. But try cleaning it or replacing as well since they are really cheap. By the way, make sure you push the valve all the way into the rubber grommet so that it clicks as it seats. Good luck

OldNuc
11-22-2008, 10:22 AM
I think this was mentioned earlier but it is good to bring it up again. With an older motor that is suffering from degraded compression and sucking oil it will tend to run rich as the O2 sensor does not play well in an oil bath. And, the oil entering the exhaust will restrict the VAT and muffler to some extent. All of this contributes to the P0172 situation.

This is not a fun code to get rid of either.:(

cboss
11-26-2008, 09:31 AM
I have done a lot of research on the internet about the P0172 code, particularly with Saturns.

Often those with this problem do common tuneup work (new plugs, wires, etc.) and the code disappears for awhile and then comes back.

I don't think they are getting at the cause though, since the tuneup work likely only helps marginally just enough to get the car slightly away from the limits which would generate the code. After driving for awhile, the engine still running rich, the tuneup benefits fade (as the plugs get fouled) and the code returns.

The key appears to be to find a way to diagnose specifically what is pushing the engine to go rich.

Now with my Saturn (96 SL2) there are a few clues I am working with which are key to making sure I actually fix the real problem.

(1) All four splugs foul quickly and are loaded with powdery black deposits (running rich). When I finally solve the problem, the plugs should remain clean.

This condition tells me the richness is not cylinder specific. If one or two injectors were bad, it would show up on those plugs only.

Since the condition is not cylinder specific, I have ruled out the coils, since the likelyhood of both coils being that bad is low. Probability would indicate that the poor condition of the plugs would be in pairs (two say real bad, the other two maybe not as bad).

Now the ignition module (underneath the coils) is not cylinder specific, so that may suspect at this time.

So how does one test to make sure the ignition module is not the cause of any problems ?

(2) The fuel trim values on the OBD II scanner I think are critical. Just because the P0172 code is not generated for awhile does not mean the problem is solved. If the fuel trim is indicating the computer is trying to lean out the mixture significantly (large negative percent values), then the engine is still running very rich. Before one thinks the code problem is solved permanently, the fuel trim values need to come back more inline with normal running values.

I consider the scanners fuel trim values critical to testing to see if the problem is truly fixed after a repair is made.

OldNuc
11-26-2008, 10:51 AM
Well, thats about the size of it. The engine is running rich and will eventually reward you with a P0172.

The ICM has no control on mixture but it could possibly (due to a failure) impact actual ignition timing. A incorrect actual timing/coil pulsing would result in incorrect timing and that would result in a richened mixture. The O2 sensor is supposed to keep the PCM informed of the A/F ratio and correct the fuel flow dynamically. If you have a JY ICM changing it might prove interesting but I would not go buy a new one just yet.

A thought. If the timing goes haywire randomly the PCM will never correct fast enough to prevent the O2 sensor from going rich sufficiently to set the P0172.

What does your scanner show for spark advance at idle? And, what is it showing for the max/min values? does it show the knock retard by any chance?

Its nice to see someone dig into a problem and not just tape over the SES light.:D

You have run through all of the possible causes except either replacing or sending out for a thorough cleaning and flow balancing of the injectors. The things can stick periodically. I have seen diesel injectors do this and its a real pain to identify.

The car I am presently driving arrived on my doorstep in a basket with a note that said "feed me". It had a mismatched set of injectors so I just replaced all of them.

cboss
11-26-2008, 04:54 PM
I installed the junk yard O2 sensor (upstream) and the fuel pressure regulator.

I tested the fuel pressure and it was about 36 psi at idle (possibly 37).

The car appeared to drive reasonably well for awhile, but after about 20 minutes and then parking the car and letting it idle, the idle was a bit rough like it has been and the P0172 code came back. Continuing to let it idle, I could clear the code and it would come back immediately.

I think by diagnosing the idle problem I may find the cause of the problem.

First, the idle is hard to describe, but I'll do my best.

It sort of chugs. Kind of like a person when they hyperventilate. The idle speed is good, about 850 rpm, but the idle is kind of like the engine is gasping for air at times. The idle is actually good, except for the surging like effect (gasp for air). Its not like it wants to die or anything and the idle speed is good. Its hard to describe.

Now when I read live data on the scanner the following is curious:

The ignition advance stays consistant when you give the engine a little gas (higher than idle RPM). At 1500 to 2000 RPM the advance is about 25 to 30 degrees.

When you let off on the gas and the engine idles, thats when the ignition advance is crazy, IMO. The values range from -1 to 30 and flucuates rapidly.

The fuel trim is strange.

If I rev the engine and let off on the gas, the ST fuel trim will stablize for a few seconds at about 0 and then it will then begin to creep down (negative direction) and then in a few more seconds the ST fuel trim will get as low as -30.

If I give the engine some gas and get the RPM's up, the ST fuel trim will stabilize around 0 for a second or so and then if I keep the RPM at say 1500 or 2000, the ST fuel trim will do the same as noted above, creeping down again back to as low as -30.

The LT fuel trim is at -21 right now.

Another I noticed while idling for awhile (say 15 minutes), the temperature gets high enough to kick on the radiator fan and its about 35 degrees out. When driving the car, the temperature stays at just below the half way mark.

I haven't installed a new thermostat yet, but may I should. I don't think it causes the richness problem though.

I have a new ECTS, but it was purchased from AutoZone (or Advance, can't remember which).

I need some more diagnositic steps to follow at this point.

What to check next and how to test.

I did test the OHM values for all the fuel injectors and they are in the right range for a 96 SL2.

I have the NGK plugs in too.

cboss
11-26-2008, 05:03 PM
I think the ST fuel trim creep may be a clue.

If I rev the engine and no matter whether I let it drop to idle or keep the RPM at say about 1500 rpm (or 2000), for a few seconds the ST fuel trim stabilizes at about 0 and then it begins what I would call a creep. Its almost as if it was counting down, 0, -3, -5, -6, -9, -10, -12, etc. (not specific sequence of course, but continually down).

This ST fuel trim creep, will bring the fuel trim down quite a bit within say about 5 or 10 seconds.

What is there about giving the engine a quick rev, that would stabilize the ST fuel trim (meaing neither lean nor rich) ?

I think testing the fuel trim values under different circumstances (different RPM, rev the engine, slowing drop rpm) may provide some valuable info.

OldNuc
11-26-2008, 05:42 PM
I installed the junk yard O2 sensor (upstream) and the fuel pressure regulator.

I tested the fuel pressure and it was about 36 psi at idle (possibly 37).

The car appeared to drive reasonably well for awhile, but after about 20 minutes and then parking the car and letting it idle, the idle was a bit rough like it has been and the P0172 code came back. Continuing to let it idle, I could clear the code and it would come back immediately.

I think by diagnosing the idle problem I may find the cause of the problem.

I agree with this avenue of investigation.


First, the idle is hard to describe, but I'll do my best.

It sort of chugs. Kind of like a person when they hyperventilate. The idle speed is good, about 850 rpm, but the idle is kind of like the engine is gasping for air at times. The idle is actually good, except for the surging like effect (gasp for air). Its not like it wants to die or anything and the idle speed is good. Its hard to describe.

The PCM will do an outstanding job of controlling the RPM, the RPM is not a real good predictive tool for what is happening at idle.


Now when I read live data on the scanner the following is curious:

The ignition advance stays consistant when you give the engine a little gas (higher than idle RPM). At 1500 to 2000 RPM the advance is about 25 to 30 degrees.

When you let off on the gas and the engine idles, thats when the ignition advance is crazy, IMO. The values range from -1 to 30 and flucuates rapidly.

I have seen this exact same thing before. Its not a real good sign.


The fuel trim is strange.

If I rev the engine and let off on the gas, the ST fuel trim will stablize for a few seconds at about 0 and then it will then begin to creep down (negative direction) and then in a few more seconds the ST fuel trim will get as low as -30.

If I give the engine some gas and get the RPM's up, the ST fuel trim will stabilize around 0 for a second or so and then if I keep the RPM at say 1500 or 2000, the ST fuel trim will do the same as noted above, creeping down again back to as low as -30.

The LT fuel trim is at -21 right now.

That is not far from normal behavior. You have to get to -100+ on LFT or SFT to set the P0172 and there is a lean drop when you open the throttle.


Another I noticed while idling for awhile (say 15 minutes), the temperature gets high enough to kick on the radiator fan and its about 35 degrees out. When driving the car, the temperature stays at just below the half way mark.

This is totally normal behavior. The car will do this even if the thermostat is stuck open. The T-stat controls the speed of heat up and where the normal driving low temperature is at.


I haven't installed a new thermostat yet, but may I should. I don't think it causes the richness problem though.

No, it does not.


I have a new ECTS, but it was purchased from AutoZone (or Advance, can't remember which).

Go ahead and change the ECTS unless you can look at the start ECTS and IAT readings and see that they match withing 3 or 4 degrees.


I need some more diagnositic steps to follow at this point.

What to check next and how to test.

The first thing to do is find out why the timing is not steady at idle. this is most likely the prime symptom of the problem. So, the best place to start is with a dry and wet compression test. Do this from a warm engine. be sure to block the throttle as far open as possible. Also pull the PCM-B fuse so as not wash down the cylinders with gas. Crank the engine until the gauge quits increasing. For the wet test squirt in a couple squirts of engine oil and crank the engine over 2 times to clear the excess and then test the cylinder. This is the wet test for that cylinder. Repeat for the other 3 cylinders and post the numbers back here so we can all stare at them. Keep track of how many compression strokes for the gauge to quit increasing for each test this is also significant data. with all but one cylinder open you will find they are easy to count.


I did test the OHM values for all the fuel injectors and they are in the right range for a 96 SL2.

I have the NGK plugs in too.

Injector coils are good and the plugs are the correct ones. The black plugs confirms the rich condition.

Not to jump to any conclusions lets see what the compression test shows. Does your scanner show knock retard?

nivlem7
11-26-2008, 05:58 PM
Your fuel pressure readings look O.K.

I fought with this code for over a year - turned out to be the aftermarket fuel filter. Replaced with OEM, and it went away.

Just a thought.

OldNuc
11-26-2008, 06:26 PM
I thinkin one of the branches of this thread, or Gordian knot, we covered the fuel pressure issue. I was also burnt by an aftermarket filter/regulator.:hmpf:

cboss
11-26-2008, 07:45 PM
I had a mechanic install a new fuel filter. It was an inexpensive one ($20) from Advance Auto.

The 96 SL2 though, does not have the fuel pressure regulator built into the fuel filter, so I doubt this would be a problem.

From what I have read here, the fuel filters with the FPR built in, are most likely to be a problem with aftermarket products.

Is it possible for an aftermarket (non-FPR builtin) fuel filter for a 96 SL2 to cause a fuel pressure problem ?
Wouldn't that show up on the fuel pressure gauge when tested ?

I agree the ignition advance at idle is an important clue and it needs to be tracked down.

What effects timing advance (or retard) in the engine ?

I have a compression gauge I recently bought from Advance.
I'll try to do a compression test when possible.

cboss
11-26-2008, 07:48 PM
Originally Posted by cboss:

Now when I read live data on the scanner the following is curious:

The ignition advance stays consistant when you give the engine a little gas (higher than idle RPM). At 1500 to 2000 RPM the advance is about 25 to 30 degrees.

When you let off on the gas and the engine idles, thats when the ignition advance is crazy, IMO. The values range from -1 to 30 and flucuates rapidly.




Originally Posted by OldNuc:

I have seen this exact same thing before. Its not a real good sign.

What caused this problem on the car you saw this on ?

OldNuc
11-26-2008, 08:32 PM
Part of the exhaust valve lodged in the CAT. From about 1500 on up it ran fine. But it would miss at idle. You do not have a piece of valve gone but you might have one that is a bit sticky and closing late. Hence the dry/wet compression test and count the cycles to no increase.

Another poster here had an issue with a 96/97 with one of those filters. Produced low fuel pressure because it disassembled inside the housing and blocked flow. You should not have any problem.

There is a timing table in the PCM and it is designed to get max advance until the knock sensor retards it. There are different values for various RPM and engine load conditions same as an old style distributor curve. so the PCM wants 20 and tt-he knock sensor says you can have 15. That is why I asked if you had an output for knock sensor retard. I think the idle value from the PCM table is 13 or 15 degrees.

cboss
11-26-2008, 10:34 PM
I decided to invest in the alldata.com subscription.

Very interesting stuff, particularly the TSB's.

I found one TSB, which I don't think I have seen mentioned on the forums, about the P0172 code (for 96 SL2).

In essence, the wire harness (wire) going to the O2 sensor (upstream) has microscopic perferations (holes) in it. Not visible to the eye.

When the wire is exposed to salt (on roads) and water, it shorts out and the O2 sensor produces intermitant errors in its readings. This causes the engine to surge at idle (which is what mine does) and will generate the P0172 code.

Repair requires replacing the entire length of wire (about 5 feet).

Has anyone who had the P0172 code ever had the O2 sensor wire (not the wire on the sensor itself, but the wire harness leading to the sensor) replaced ?

It must be a serious problem for Saturn to write up a TSB about it.

OldNuc
11-26-2008, 10:48 PM
I doubt if it is a real common problem. You may have stumbled onto the solution though. The O2 sensors have air as a reference for the base O2 value. Does that give a part number? If it does punch it into Get Saturn Parts and see what it costs. The car would have to be in one of the horrid high salt areas of the country. That wire is well protected in the loom and very little of it is exposed. If you could record real time O2 sensor data you would see it fail and it would be real obvious.

If that wire can be had I would buy one and try it. You could pull one from a junker but they probably changed the design in later years. What is the TSB number?

cboss
11-27-2008, 07:09 AM
The TSB mentioned is:

96-T-32A

Dated April 1997

OldNuc
11-27-2008, 09:38 AM
Does not pertain to the 98 and later years so I do not have it in ALLData. Did it give a part number for the wire?

OldNuc
11-27-2008, 09:55 AM
The TSB mentioned is:

96-T-32A

Dated April 1997

That TSB is only applicable to the 95 and 96 model year.

cboss
11-27-2008, 10:02 AM
My car is a 96 SL2, so the TSB would apply!

OldNuc
11-27-2008, 10:17 AM
i know. but ther is a change in the design of the later cars so as to remove the problem. I suspect that it is a different connector and plain old wire back to the PCM but can't prove it yet. that is why I was looking for a copy of the TSB or the repair part numbers at the bottom.

This could actually be your problem if your car has been in a lot of salt.

cboss
11-27-2008, 12:45 PM
Ok, I reread the TSB and it was for a specific group of VIN numbers and mine does not fall into that group.

I also did the diagnostic procedure in the TSB and mine did not indicate a problem.

Maybe tomorrow I can do the compression test.

The P0172 code is coming on quicker now, after I changed the FPR and O2 sensor with the JY ones.

I have an idea I want to follow through on.

I want to go through some of the PCM grounds and check the wires to see if the ground is poor or the wire is worn.

One PCM ground, just above the coils, the wire appears overly flexible (meaning it could be breaking inside). I need to check this out.

I noticed two different PCM grounds on the diagrams found on alldata.com

Are there any wiring problems (ie. bad grounds) which would have a direct impact on timing , particularly at idle ?

cboss
11-27-2008, 02:56 PM
My goal right now is to define all the probable causes of the erratic timing at idle. The engine stabilizes timing fine at RPM's above idle (ie. 1000 RPM and above).

The timing at other RPM's appears to be within proper range.

Why only at idle ?

What electrical causes could there be ?

What mechanical causes could there be ?

It has to be something that would only effect timming at idle.

To find this, likely should lead to me the cause of the richness problem.

Has anyone ever had a problem with timing at idle only which was traced back to the:

- Crank Position Sensor ?
- Ignition Module ?
- One coil or both at same time ?
- a bad ground ?
- a bad PCM ?

OldNuc
11-27-2008, 05:06 PM
OK, The base spark timing is generated in the PCM based on RPM and the CPS modified by the knock sensor input. Nothing more. There is an RPM vs advance table in the PCM and a closed throttle setting. If there is any flakiness with the TPS signal it will change the timing or if the engine has a weak cylinder it will also change timing.

The idea is to run down the cause of the timing variation at idle.

Did you try another ICM module yet?

Look at the TPS voltage on the scanner and se if it is steady. Graph it if you can. It should not change.

The grounds on the back of the block are for the O2 sensor and they will cause grief. The front one on the transmission is the PCM ground and it is a very floppy wire but the wire is good because the car runs. Cleaning them will not hurt a thing. Use one of those kitchen yellow sponges with the green Scotch Brite on the back. A well used one works best. Don't overclean all of the lead alloy coating off the rings. Just shine it up a bit.

Only at idle points to a slow moving valve, intermittent 0 throttle position contact in the TPS, low compression in one cylinder, or a bad ICM or CPS. And I probably missed a few.

If there is something like a misfire the timing will change. Your variation is much worse than a misfire. I was only seeing about 6 degrees change when the cylinder was about dead.

cboss
11-27-2008, 09:57 PM
intermittent 0 throttle position contact in the TPS

I installed a new TPS. At idle the TPS reads 0 on scanner.

Is this not correct ?

Did I installed it incorrectly ?


As far as the PCM ground on the front, the wire appears to be overly flexible close to where it is connected. This is usually an indication of the wire breaking in the strands, so it is weak. It can still pass electricity, but it will do it poorly. I need to get a new metal connector, cut the wire and install a new connector on it and reconnect it (clean connection too).

The thing about grounds is that a computer is totally dependent upon very exact voltages and a bad ground can really mess things up.

I work with computers and an old trick I learned years ago for a computer which is on the fritz, was to take every chip which is a plugin type (not soldered) and pull them, clean the contacts and reseat them. You would be amazed at how often this solved, what appeared to be terrible problem.

Auto computers are no different. Just a lot of long wires, instead of one big circuit board. Sensors all over the place, connected by wires to a central computer (or two) and bad contacts can wreak havoc.

Its worth a little time to check some of the grounds and connectors to make sure they have clean connections.

The junkyard O2 sensor appears to be weak. I am now getting the P0133 code (too slow), so I need to swap back the previous one (new universal Bosch). I'll see if I can get a new quality 02 (Denso or NGK).

OldNuc
11-27-2008, 10:45 PM
I installed a new TPS. At idle the TPS reads 0 on scanner.

Is this not correct ?

Did I installed it incorrectly ?

That sounds right. You are getting a reading in % throttle position.


As far as the PCM ground on the front, the wire appears to be overly flexible close to where it is connected. This is usually an indication of the wire breaking in the strands, so it is weak. It can still pass electricity, but it will do it poorly. I need to get a new metal connector, cut the wire and install a new connector on it and reconnect it (clean connection too).

Be sure to solder the terminal on after crimping. The hand crimper does not make a decent long term weather proof crimp.


The thing about grounds is that a computer is totally dependent upon very exact voltages and a bad ground can really mess things up.

Oh, how well I know that. There is a coper compound the name of which I can not remember at the moment that you can buy from Eastwood that does wonders for automotive ground and power connections. I have used it for years. It is a bit hateful to work with as you can build a nice short with it also.:( I will get the name and part No in the AM.


I work with computers and an old trick I learned years ago for a computer which is on the fritz, was to take every chip which is a plugin type (not soldered) and pull them, clean the contacts and reseat them. You would be amazed at how often this solved, what appeared to be terrible problem.

Not amazed at all, it works. The problem is most klutzes break at least one pin on every chip or are human Van DeGraf generators and fry them all.


Auto computers are no different. Just a lot of long wires, instead of one big circuit board. Sensors all over the place, connected by wires to a central computer (or two) and bad contacts can wreak havoc.

Its worth a little time to check some of the grounds and connectors to make sure they have clean connections.

Yes, it is a very good idea. There is a splice pack right in front of the windshield washer fluid bottle and the spillage goes right into it. It is bolted to the frame rail and after unbolting the black plug should pull off of the multi prong contact strip. Then the plug will come apart. Keep the wires in order and clean that one up as it causes front end lighting problems.


The junkyard O2 sensor appears to be weak. I am now getting the P0133 code (too slow), so I need to swap back the previous one (new universal Bosch). I'll see if I can get a new quality 02 (Denso or NGK).

That would be a good Idea. Order the NGK-NTK from Summit Racing its the cheapest and see if there is any other item you want as their shipping is a flat 10.00. The dogbone inserts and poly grease boots for the tie rod ends and ball joints are a good thing to have. I have part numbers if you want them.

OldNuc
11-28-2008, 11:05 AM
This is the link to the contact coating material. Don't use the brush in the bottle, you want the thinest coat of this stuff you can get on the connecting bolt and rings. It does help quite a bit. Not for small pin connections though.

http://www.eastwoodco.com/jump.jsp?itemID=1369&itemType=PRODUCT

cboss
12-01-2008, 11:45 AM
I just purchased a use coil pack (both coils and the ignition control module) on EBay, which was from a 2001 Saturn SL. I checked the online parts stores and I think the 2001 coils and ignition module are compatble with my 96 SL2.

I got the entire coil pack for $22.39 (shipping included).

The ignition module is kind of pricy new and its from a newer model (2001) so it should have some live left in it.

This will give me a good chance to test the ignition side to see if it is causing any of the problems, particularly the ignition module.

Weather hasn't been very good lately so I haven't had a chance to do a compression test yet.

I swapped out the O2 sensor (the JY one) and put the Bosch Universal back in for now. No SES yet.

OldNuc
12-01-2008, 02:02 PM
That will be a good test. There are documented mixture problems that have been traced to non OEM specification O2 sensors. You removed a generic sensor and replaced it with what has proved to be a sick one from the J/Y.

cboss
12-01-2008, 02:31 PM
I ordered a new O2 sensor.

Found a brand new NTK (21501) O2 sensor for Saturn SL on Ebay. It was offered for $38 (BuyNow) and you could use the "Make an Offer", so I offered $30 and got it. They'll ship it priority mail for just $6.75.

I'll get a brand new NTK sensor for a total of $36.75.

I double checked the part number with the one at summit racing and its the same.

Just curious. You refered to NGK or NTK in one of your posts. I assume its the same company, just a different branding ? (NGK or NTK)

When I get the parts, I'll install the new O2 sensor and just the ignition module for a start. I'll clean up some of the PCM grounds too.

OldNuc
12-01-2008, 02:37 PM
its actually NGK-NTK, the NTK is a ceramics division of the parent company. I think NGK is another division of the same parent. Japanese companies suffer from inbreeding.:D They are related and NGK website lists the O2 sensors. They also say NTK on the box.

Did you catch the part number for that copper compound?

cboss
12-01-2008, 04:50 PM
OldNuc,

You wouldn't happen to know the actual math equation used by Saturn engines for the fuel injector pulse width ?

As a computer programmer I found this very interesting.

ie.

Chrysler fuel injector pulse width equation:

Pulse Width = (RPM X MAP/BARO) X TPS X ECT X IAT X Battery Voltage X (STFT X LTFT)

What intrigued me was how each sensor value is integrated into the equation.

This would mean, that rather than one single sensor (item) being at fault, multiple sensors (or problems) could be slightly off and the cumultive effect would be the fuel injector pulse width being off.

If for example, the STFT and LTFT are slightly off because of the O2 sensor, the battery voltage being off (too much juice from alternator), the MAP being off, the total effect would be a problem.

In essence, rather than look for a single culprit (ie. bad O2), multiple problems could combine together to push the mixture over the limit.

This would explain how often with the P0172 code, people fix one thing and all appears better for awhile, but as time goes on and the fix has less benefit (ie. plugs starts to foul again) the code may return.

While one particular part may be the "biggest" culprit, they can all combine to create a problem.

I found a very interesting article about Saturn fuel injection and how it works on the different engines:

http://www.teamscr.com/fuel.html

OldNuc
12-01-2008, 06:31 PM
OldNuc,

You wouldn't happen to know the actual math equation used by Saturn engines for the fuel injector pulse width ?

As a computer programmer I found this very interesting.

ie.

Chrysler fuel injector pulse width equation:

Pulse Width = (RPM X MAP/BARO) X TPS X ECT X IAT X Battery Voltage X (STFT X LTFT)

I have never spent the time to dig it out of the Powertrain Control Manual but that looks real close. Its not directly spelled out in the manual so you have to put the pieces together yourself. Keep in mind that the STFT and LTFT are continuously monitored and modified based on feedback from the O2 sensor.


What intrigued me was how each sensor value is integrated into the equation.

This would mean, that rather than one single sensor (item) being at fault, multiple sensors (or problems) could be slightly off and the cumultive effect would be the fuel injector pulse width being off.

If for example, the STFT and LTFT are slightly off because of the O2 sensor, the battery voltage being off (too much juice from alternator), the MAP being off, the total effect would be a problem.

In essence, rather than look for a single culprit (ie. bad O2), multiple problems could combine together to push the mixture over the limit.

This is true in both theory and practice. The secret here is that all of those other sensors will set a code if they are out of limits. A couple will set a code if they are just a bit out and others will only set a code if they fail totally. And one, the ECTS is renowned for never letting on that it has failed.


This would explain how often with the P0172 code, people fix one thing and all appears better for awhile, but as time goes on and the fix has less benefit (ie. plugs starts to foul again) the code may return.

The secret to the P0171/172 codes is that they only set if all the other devices you listed above are not setting a code and they only set when the fuel trim hits its limit rich or lean.


While one particular part may be the "biggest" culprit, they can all combine to create a problem.

This is very true and it makes fixing this problem "difficult" at best. A good scanner is a very big help in tracking all of these inputs dynamically.

It is also important to remember that the fuel trim is derived from O2 sensor input under specific driving conditions and base data from the PCM. The result is the operating fuel trim. The O2 sensor design is critical to the proper control of the STFT and LTFT and resultant A/F ratio. The manufacturer has specified a set of output characteristics for the sensor used in their cars. If you start reading up on O2 sensors you will see this has been a problem for many owners/racers/tuners over the years.


I found a very interesting article about Saturn fuel injection and how it works on the different engines:

http://www.teamscr.com/fuel.html

I saw your other post and read the article. Quite a bit of good information in that article. for those that are asking "what happens when I ad this hop up goodie" it provided some numbers to go with the theory.

cboss
12-03-2008, 10:21 PM
I won't get the O2 sensor and coil pack until probably Friday.

I worked on the car today doing some seemingly small stuff, but it has made a difference in normal driving (not the idle yet though).

I put a new connector on the PCM ground wire and reconnected it. I cleaned it up good.

I took off the coil pack and cleaned all the contacts and connections and reinstalled it.

Lastly, something I should have thought about earlier, I removed the MAP sensor (which was a new one) and I examined the hole it connects to. The main hole is large (about 3/8 to 1/2 inch diam.), but it becomes quite small at the bottom and it was all gunked up with crud. This would likely effect the MAP sensors ability to work. I cleaned it out real good with carburator cleaner.

I also cleaned out the throttle barrel with carb cleaner as best I could.

I noticed a significant improvement in RPM's about 1000 and driving.

The idle still has a sort of surge to it (more like grasping for air) which I need to fix.

The rear O2 sensor values on the scanner seemed better afterward and the front O2 values seemed better too.

OldNuc
12-03-2008, 11:19 PM
Take the throttle body off and clean it up real good. The IAC seat and pintle will be caked with hard gunk. I thought you had cleaned out the MAP port. That will cause all kinds of dynamic mixture problems....

Good grounds and clean helps...

cboss
12-04-2008, 09:17 AM
Does the throttle body come off very easily on the DOHC (96 SL2) ?

Is the intake manifold part of it (one piece) or separate ?

What causes the throttle body to get dirty in the first place ?

I understand how a bad PCV valve could cause it to get dirty.

The other rubber hose which connects to the Valve cover and the air intake section (the big plastic intake section from air filter to throttle body), I disconnected it from air intake section and it blows out air (oily) and dirties up the intake. Is there any way to decrease this ?

OldNuc
12-04-2008, 10:08 AM
Does the throttle body come off very easily on the DOHC (96 SL2) ?

Is the intake manifold part of it (one piece) or separate ?

It is separate. There are 2 nuts and the throttle cable holding it on. There is also a brace over to the EGR. In the How - To Library first post is a good richpin video of how to get it off and how to clean it.



What causes the throttle body to get dirty in the first place ?

The blow by combustion gasses and oil vapors are sucked into the intake and condense back to goo. The air intake, that other hose will eventually blow out little puffs of this gook after the blow by exceeds the capacity of the PCV side to maintain the engine at a very slight negative pressure. This gooks up the intake system before the throttle plate.


I understand how a bad PCV valve could cause it to get dirty.

It will get dirty with a good PCV also. The PCV limits the volume of air pulled into the intake. You always get the oil vapor and blow by gasses pulled into the intake.


The other rubber hose which connects to the Valve cover and the air intake section (the big plastic intake section from air filter to throttle body), I disconnected it from air intake section and it blows out air (oily) and dirties up the intake. Is there any way to decrease this ?

All of this blow by usually is indicative of a need to rebuild the engine. Valve seals and rings are the primary causes of this. A catch can can be rigged on the air inlet to catch any oil that blows out and still allow air for the PCV system.

cboss
12-06-2008, 03:35 PM
I watched richpins videos on cleaning the throttle body.

I was thinking it may be possible the hole for the IAC may be clogged which could affect the idle (and fuel trim at idle).

One question before I attempt to clean it:

When I reinstall the throttle body after cleaning, do I have to replace the old gasket ?

If not, does it need to be coated with anything, before reinstalled ?

cboss
12-06-2008, 03:56 PM
After cleaning up the PCM ground (new connector), cleaning the coil pack, cleaning the MAP sensor hole, the cra has run better and no code for a couple of days. The idle is still a little off though.

Today I drove the car about 60-70 miles, around town driving. No code until my drive was over. The P0172 code has come back on again and when I cleared it, it came back on immediately. I can clear it a couple of times and it comes back on immediately.

My wife noted that in the past, the weather changed when the code cam on and a snow storm was moving towards out area today and just about when the code came back on, the sky started to get dark, the air wet and very cold. Later snow showers came.

I don't know if the damp air makes a difference (effect something electrical) or the change on air pressure (would effect MAP sensor).

I got the NTK O2 sensor in the mail today (brand new).

I'll see of it makes a difference.

OldNuc
12-06-2008, 04:16 PM
The lower air temperature richens the mixture by computing to a higher air mass entering the engine. So, the PCM adds gas to keep the A/F ratio at 14.7:1.

Lets see if the O2 sensor will straighten it out. Thee are a bunch of variables here as you know and it can be difficult to find the contributor.

cboss
12-06-2008, 05:10 PM
I think I am getting somewhere.

OK, I installed the NTK O2 sensor.

It was the perfect time to do it, since the weather change tipped the scale and the P0172 code is coming on again.

I ran the car for a few days after the cleaning I mentioned abobe and no code. Today for some strange reason the code comes on and I can't clear it for long. The weather is definitely a factor.

The weather is one key to the puzzle. Not that the problem was actually fixed, but the problem was close to the acceptable margins, but the weather tipped the scale. This may explain why some people think they fixed the problem, only to have it appear days or weeks later. They don't take into consideration slight changes by things like the weather , bad gas, etc.

I was hoping the new O2 sensor would solve the problem, but I think you will find this interesting.

I installed the new NTK O2 sensor (can't get better) and the fuel trim appears to be improved. I can tell it makes a difference.

Now I changed the O2 sensor today with the bad weather here and the code coming on immediately.

Guess what ?

The code still remains!

I decided to get out my scanner and to try some tests.

This is interesting:

I let the car idle until it was up to temperature (it is 32 degrees out now and snowing). I waited until it started to pant a little, which is what id oes when the idle is poor. The code comes on.

I then cleared the code and it comes back on immediately.
I then cleared the code and it comes back on immediately.
I then cleared the code and it comes back on immediately.
I then cleared the code and it comes back on immediately.

over and over again I did this.

Now I gave the car just a very little bit of gas and tried to keep the RPM's low, but just above idle speed.

I cleared the code and waited.

It doesn't come back on!

This means the code is only being generated during idle!
(or very low RPMs)

Now I assume the code could come on during driving, if one is braking (no gas and technically at idle) or when one is shifting on the highway and you let up on the gas.

I think the idle problem could be related to a dirty throttle body and in particular the air hole for the IAC. If that gets clogged, the engine would starve for air at idle which could explain the tendency to pant (best word I can find to describe it). This would create a very rich environment, since no air or little air, means very rich.

Next step is to remove and clean the throttle body.
(I had already installed a new IAC)

Is there anything else unique to idle that I should look at ?

OldNuc
12-06-2008, 07:05 PM
If you have one of those rough idles it may just do what you are seeing. It would be rich. Remove the throttle body and take a couple of can s of carb cleaner to it. Do not get carb cleaner into the TPS. The TPS and IAC come of before the cleaning starts. If its an old TPS while its off check the black and blue wires for a no skip smooth sweep from closed to full open. Remember , there is a slight windup on it when the throttle is closed. How - To Library and to the videos. All you ever wanted to know including what sized wrench is needed to get it off. And how to get the throttle cable off and back on after cleaning.

Once its all back together, let it warm up and check the minimum idle air setting. The value should be 500 RPM or a bit less. To check this you stick your finger slowly over the IAC air inlet port till its totally blocked off. If the engine does not stall out then the scanner will read the RPM for you. If its not right then the fix is to adjust the stop screw setting. When you cleaned the throttle body you should have found it. Supposed to have "keep your sticky fingers off of me cap on it", it is not supposed to be tweaked.

I think that the lowering temperature is being detected by the IAT sensor and that is not helping. You can pull it out of the air box and artificially heat it up to 85 degrees by holding onto it and see what that does to the generation of the code. Interesting experiment. Check the scanner and see if it displays intake air temperature.

cboss
12-06-2008, 08:38 PM
I noticed something strange when using the scanner.

When I examine the "live data", with the engine off (key on), the RPM (which should be zero) is set at 61.

Is this normal ?

If not, what would cause it ?

The IAT appears fine, since with engine off and cold and temperature is close to the air temperature (comparing it to a thermometer).

The ECTS is a new one, bought at AutoZone (or Advance, can't remember which).

Has anyone ever had a problem with aftermarket ECTS's ?

OldNuc
12-06-2008, 09:01 PM
ECTS is usually not problem. The idea of artificially warming up the IAT is to artificially lean out the mixture a bit and see if the code goes away. The RPM usually hangs onto the last real data and does not go to zero. If the idle is within the proper range that should not be indicative of a problem

cboss
12-07-2008, 02:24 PM
I cleared the code yesterday.

I drove the car about 60 miles today and no code.

The temperature out is very cold (in the 20's).

It appears the cold does not kick the code, but maybe the humidity. The times in the past when the code came back and would not easily be cleared, the air was very wet (about to rain or snow).

Could there be something effected by humidity ?

Something electrical ?

OldNuc
12-07-2008, 02:55 PM
Its a thought but there is no humidity input to the pulse calculation. The electrics are reasonably waterproof. Carburetor equipped cars run better in a high humidity atmosphere. Had to do with the moisture acting as an antiknock agent. Did the timing change? Did you ever locate a readout of knock retard?

Its not cold, so it must be the amount of moisture in the air. Fuel trim is miles away from the set point. What happens if you induce a fine mist into the intake?

cboss
12-09-2008, 10:24 AM
My scanner can display the item:

Ignition Advance

But does not display knock retard.

I did install a new knock sensor (from autozone) with little change.

Could the knock sensor wire be a problem ?

Another question:

When I remove the throttle body to clean it, do I need to replace the gasket or coat the old one with something ?

OnDaGround
12-09-2008, 10:29 AM
replace it.

cboss
12-09-2008, 12:10 PM
The gasket is cheap enough so no problem in replacing it.

Should the new one be coated with anything before installing it ?


On another subject:

What do you think of adding a Oil Catch Can to the PCV line ?

OldNuc
12-09-2008, 12:26 PM
Actually if the thing does not delaminate when you take off the throttle body you can reuse it. It will stick to either the manifold or the throttle body. It is a high temperature fiber and graphite bonded to a piece of perforated metal. Just like an exhaust manifold gasket. As the throttle body does not get hot you can keep using the same one till it starts to shed the fiber/graphite. They are cheap and you can get one at any parts store. They are designed to be installed without any coatings added.


Adding an oil separator is not a simple design if you want it to work right. You need to locate several very coarse commercial stainless steel pot scrubbers, SAMs club has them. Then a piece of 2 1/2 to 2 3/4 exhaust tubing about 10 inches long. And a welder. If you are interested I have a drawing somewhere. The thing has to condense and collect oil vapor and keep it in the bottom of the tube so that it can be periodically drained.

cboss
12-09-2008, 01:04 PM
How about this simple design:

http://www.mirage-performance.com/sonata/CatchCan/index.html

OldNuc
12-09-2008, 01:13 PM
That will work. Replace the filter with one of the pot scrubbers. Wire it into a cylindrical shape. Harbor Freight sells the separators cheap.

cboss
12-09-2008, 04:48 PM
I removed the throttle body today.

The IAC hole wasn't dirty as all, but the back side of the throttle plate was very gummed up. I made sure the vacuum line holes were cleaned well too.

The IAC (according to directions) can have the pintle turned to move it in or out some, so I turned it back a little (originally I set it as 1 1/8 inch the max).

I cleaned the gasket surfaces well.

I use a new gasket and made sure it was tight.

I ran the car and tested with scanner. There is some improvement. The fuel trim seemed better (I cleared computer, by disconnecting battery for awhile) and the wierd timing at idle has improved some, but stll vacilates at idle.

It is a very wet day (dizzled all day and will rain tonight) so this is the best time to test it. I'll take it for a drive tonight since I have to go out. I'll drive some 40 miles, some on highway.

I also disconnected the PCV hose and sprayed carb cleaner through the hose to clean it. The hose was pretty dirty and initially the cleaner wouldn't go through easily.

I need to work on the timing at idle a bit.

Oh, the MAP sensor seemed to have better readings.

I wonder if there was a slight vacuum leak with the old throttle body gasket ?

OldNuc
12-09-2008, 05:04 PM
It would not be the first time that one of those high temperature gaskets leaked. If you go to Harbor Freight and buy a tube of Hylomar Racing Formula you can spread a thin ring of that on both sides and it will not leak. You can get the Hylomar back apart.

Info:
http://www.hylomar.us/index.shtml

cboss
12-09-2008, 09:42 PM
I took the car this evening. A 20 mile drive , some at 35 mph, most at 55 and 65 mph. After an hour came home (same trip opposite direction). A total of about 40 miles.

The car seems to run better on the highway and the idle appears improved. The weather was still drizzly out.

No P0172 code yet.

The computer has to relearn stuff, since I disconnected the battery when working on the car (just didn't want to accidently spark the battery, since I was working in that area).

I think the cleaning helped some, the new gasket may have helped some too.
I also think cleaning the PCV hose made a difference.

Some observations:

The vacuum line hole for the fuel pressure regulator is an awfully small hole (the hole in the throttle body). Does the size of the hole effect the amount of vacuum ? Has any done a MOD to increase the hole size to see what effect it would have on the fuel pressure ? Just curious.

The hose for the PCV is not very large internally. I assume the hose is the OEM. It is not simply a hose. It appears to be a hose with a small internal diameter, with a swivel rubber connector on each end which has a larger opening, to connect to intake manifold and PCV valve (in valve cover). The middle section (between the two swivel ends) has a secondary foam like rubber cover over it (to protect it I assume).

Does anyone have a photo of an OEM PCV hose they can post so I can compare it ?

When I cleaned the PCV hose out (sprayed carb cleaner through it and it was dirty), it appears the other hose which connects from the valve cover to the air intake (plastic section) which sucks air in, it has less blow back now. It is not very strong blow back now.

I am curious about the PCV valves, rubber gasket (seat) which is in the valve cover. How tight should the PCV fit in it ? Do they leak when older (even if the appear in good condition) ? What effect does such an air leak have on the PCV system ?

OldNuc
12-09-2008, 11:18 PM
I took the car this evening. A 20 mile drive , some at 35 mph, most at 55 and 65 mph. After an hour came home (same trip opposite direction). A total of about 40 miles.

The car seems to run better on the highway and the idle appears improved. The weather was still drizzly out.

No P0172 code yet.

This is a good sign.


The computer has to relearn stuff, since I disconnected the battery when working on the car (just didn't want to accidently spark the battery, since I was working in that area).

I think the cleaning helped some, the new gasket may have helped some too.
I also think cleaning the PCV hose made a difference.

Actually cleaning is quite important. the PCV system is supposed to maintain the engine internal area at a very slight negative pressure and the volume of air sucked through that hose is calculated into the fuel pulse width as a constant depending on MAP pressure. If the PCV hose was plugged that may have been a major contributing factor.


Some observations:

The vacuum line hole for the fuel pressure regulator is an awfully small hole (the hole in the throttle body). Does the size of the hole effect the amount of vacuum ? Has any done a MOD to increase the hole size to see what effect it would have on the fuel pressure ? Just curious.

The regulator is a sealed chamber so there is only a pressure and not a flow except when the vacuum changes and there is no real reason to have rapid changes. This volume of the connecting hose and chamber is small and reacts rapidly to changes in manifold pressure. Works fine, not broke.


The hose for the PCV is not very large internally. I assume the hose is the OEM. It is not simply a hose. It appears to be a hose with a small internal diameter, with a swivel rubber connector on each end which has a larger opening, to connect to intake manifold and PCV valve (in valve cover). The middle section (between the two swivel ends) has a secondary foam like rubber cover over it (to protect it I assume).

Does anyone have a photo of an OEM PCV hose they can post so I can compare it ?

What you have is the original undamaged OEM part. That is how its made.


When I cleaned the PCV hose out (sprayed carb cleaner through it and it was dirty), it appears the other hose which connects from the valve cover to the air intake (plastic section) which sucks air in, it has less blow back now. It is not very strong blow back now.

That is a sure sign that it was severely restricted and you have fixed something that was broke. The air inlet hose should have a net flow into the engine when at idle.


I am curious about the PCV valves, rubber gasket (seat) which is in the valve cover. How tight should the PCV fit in it ? Do they leak when older (even if the appear in good condition) ? What effect does such an air leak have on the PCV system ?

The grommet has a grove in it about 1/4 inch down from the top and the ridge of the PCV valve should snap into that grove. That will make an air tight seal you be somewhat loose. If it leaks you will see an oil stain spreading out from the PCV valve and grommet. If that happens to be the case then you get a new one. They are almost a generic item if you take the old one to a NAPA store they usually have them. An air leak at the PCV valve grommet reduces the vacuum in the crankcase and that will increase oil leakage and oil dilution by combustion products blow by. It is supposed to be loose as it seals on that ring and grove. You have to look for the oil smudge around it. Clean it all of and watch it for a week.

cboss
12-09-2008, 11:46 PM
For those of you struggling with the P0172 code!

It appears so far, that some of the less expensive fixes, do help a lot.

So far the following has helped: (these I have mentioned earlier)

(1) Check all vacuum lines carefully!
The one to the Evap Solenoid was oily and had slipped off. I cleaned it good and made sure it stuck.
The vacuum line to the FPR was about to break and finally broke and I fixed it. I cleaned the vacuum line to the FPR using carb cleaner.

(2) Clean MAP sensor hole. This was very dirty and clogged. A MAP sensor may be fine, but the hole is dirty.

(3) Install new Air Filter. (Mine was beat)

(4) Install new PCV valve. Make sure you clean the inside of the PCV valve hose well.

(5) Remove Throttle Body (not difficult and takes about 30 minutes) and clean it very well. Definitely use a new gasket. Clean surfaces for gasket well. If not installing a new IAC, clean old one's pintle well.

(6) Install new NGK plugs.

(7) Clean Coil Pack contacts (remove coil pack and clean and reinstall).

(8) Install new ECTS.


For the expensive stuff, I would start with:

(1) Upstream O2 sensor. Buy OEM or NTK or Denso.


While I am not finished yet (still tracking down a slight problem at idle), the above suggections may solve a number of P0172 code problems for others and is worth trying. Other than the O2 sensor, they are all cheap and only require a little time and effort.

One definite test in diagnosing the problem, pull the spark plugs (even if you are not changing them) and examine their condition.

If only one or two are black, then you have a cylinder (or two cylinder) specific problem.

If they are all black, the the problem is not likely cylinder specific (ie. a bad coil), but is more likely universal (effecting all cylinders).

If you still have problems, invest a few dollars a buy a fuel pressure guage. Harbor Freight has them cheap (mine was from Advance for $40). Make sure the fuel pressure is right. If it is too high, the FPR could be bad or you missed a vacuum leak to the FPR.

Fuel injectors, IMO are not likely the problem if all four plugs are black (not cylinder specific and what are the odds all four injectors are bad). Either way, do a leak down test with the fuel pressure gauge any way to see if any of the injectors are leaking.

If any of my suggestions are in error, I am sure some of the more experienced Saturn guys/gals will correct the error.

OldNuc
12-10-2008, 12:01 AM
Good list, nothing wrong with it at all.

cboss
12-14-2008, 05:03 PM
Ok, no code for a few days.

Today I drove about 20 miles, highway speeds, then a little later around town. Then coming home on the Interstate (65 mph), I commented to my wife that it looked like a storm coming in (dark clouds). Just a few moments after I made that comment, the SES came one (P0172 again).

When I got home (just idling), every time I cleared the code, it immediately came back on.

The change in weather obviously has some effect.

I checked the scanner and it had a freeze frame (I assume the computer captures it the moment the code is set).

Here is the freeze frame data:

Load - 50.1%
Coolant - 201 degrees
STFT - 0%
LTFT - -19.5%
MAP - 18.6
RPM - 2780
Speed - 65 mph
TPS - 18%

I cleared the code a couple times and the SES came on immediately and I got another freeze frame at idle. The data is:

Load - 29.4%
Coolant - 194 degress
STFT - -7.8%
LTFT - -21.8%
MAP - 11.5
RPM - 885
Speed - 0
TPS - .3%

I checked a thermometer on the house and it was about 50 degrees out.

The MAP read 27.7 with engine off, which would be the barometric pressure.

What to check now ?

I knew I hadn't solved the problem yet, but I was hopeful for the moment. Now back to the drawing board.

OldNuc
12-14-2008, 05:36 PM
There are some bits of information regarding disagreement between the MAP and the TPS. I will look. There is nothing wrong with the fuel trim.

OldNuc
12-14-2008, 08:59 PM
I will run a scan tomorrow, its 50 degrees colder here.:( The engine load looks high, but maybe not so I want to compare. Nothing on the MAP/TPS.

cboss
12-14-2008, 10:45 PM
Is the dealer the only place one can buy a replacement rubber grommet (mine is orange colored) for the MAP sensor ?

The closest dealer is about 80 miles away.

Does the MAP sensor grommet ever pose a problem ?


Could the change in air pressure (when it is about to snow or rain) be just enough to push the sensors over the edge to make it too rich ?

The air temperature does not appear to have any effect.

It is definitely the air pressure (when the sky has that "its going to rain or snow" look).

OldNuc
12-14-2008, 10:57 PM
The MAP grommet could be a vacuum leak. Take it up and coat the grommet with silicone grease. Put it back an see if it changes anything. The item that is interesting is the engine load at idle. Seems high but I could be wrong. And it does seem that changes in barometric pressure are related.

The grommet just might be a generic part that can be had at a place like NAPA.

You could call the local weather office and get a set of uncorrected barometric pressures for your local area at the time the code came back. Sometimes the local TV has a weather station and will have the data. You want uncorrected data though for the local area.

cboss
12-15-2008, 04:47 PM
I can't get rid of the P0172 code now.

I decided to change out the ignition module and coils I bought (used from a 2001 car). They improved the fuel trim and the car ran better for a little bit (still a code though).

I did a test run and then when home, I cleared the code and it came back immediately.

I figured since the code won't go, it would be a good time to test the fuel pressure again. At first it was at 37 psi (just above the acceptable range). I tested it with engine off, key on and the normal pressure is 42 which is OK (fuel pump fine).

I started the car again and checked the fuel pressure (at idle) and the pressure was worse. I think maybe the used FPR I installed awhile back may be going too.

The strange thing is the fuel pressure is not consistant.

The needle fluctuates rapidly, between 33 and 44 psi.

What causes the fluctuation ?

Is it simply a bad FPR or could something else cause it to fluctuate like that ?

OldNuc
12-15-2008, 04:57 PM
The regulator is controlled by vacuum from the intake. Pull the vacuum hose off of the regulator and see if there is gas in it. That would be a leaking regulator diaphragm. With the vacuum line disconnected you should have a steady fuel pressure. On the high side but steady. If its still jumping around either the regulator is shot or the fuel pump is not doing well. The regulator can fail so as to not smoothly control pressure. Usually caused by sticking internally. Experience has shown that only the dealer sells a regulator that is worth carrying home. Several aftermarket units have exhibited the symptoms you describe.

cboss
12-15-2008, 07:18 PM
Is this an OEM fuel pressure regulator:

http://cgi.ebay.com/ebaymotors/95-97-Saturn-SC-SL-SW-Fuel-Pressure-Regulator_W0QQcmdZViewItemQQ_trkparmsZ66Q3a2Q7c65Q 3a1Q7c39Q3a1Q7c240Q3a1318QQ_trksidZp3286Q2ec0Q2em1 4QQhashZitem250327062082QQitemZ250327062082QQptZMo torsQ5fCarQ5fTruckQ5fPartsQ5fAccessories

OldNuc
12-15-2008, 07:49 PM
Its an OEM manufacturer but no idea if it meets the OEM specs.

GetSaturnParts lists the real one for $58.48 + the 7.95 to ship it. They are cheaper and its a real Saturn part. Part No. 21007222

cboss
12-15-2008, 09:24 PM
I just ordered a new FPR and PVC valve from getSaturnParts.com.

No matter what, I need a new FPR (not a JunkYard one) and PVC valve.

GetSaturnParts was a lot cheaper than a dealer price ($73.50 at dealer).

cboss
12-19-2008, 11:43 AM
I tested the fuel pressure again yesterday and no more fluctuation. The pressure is 37 psi, just slightly above the acceptable range.

I think I may have messed up the fuel pressure test last time, because I pressed the relief button on the gauge and let out some gas (into a gas gan). Maybe doing this while the engine is running may mess the diaphram in the fuel pressure regulator so it fluctuated.

Its back to normal now.

I should be getting the new FPR monday (and new PCV valve) to install it.

I checked the car with the scanner yesterday when I did the fuel pressure text again and I noticed something. I watched the O2 sensor values while the engine was heating up and they appeared quite normal during while the engine was heating up. Even after it went into closed loop it was fine. But after awhile as it got closer to running temperature, it started to run rich. The O2 after the CAT was quite lean for awhile, but then when the engine got up to temp the O2 (after CAT) stayed around .760 to .800 volts (I think I have the decimal in the right place).

OldNuc
12-19-2008, 01:09 PM
As the O2 goes rich the fuel trim goes in the leaner direction. It is not a perfect control loop so there is some offset. What is it telling you for the A/F ratio? That should be 14.7.

How does the IAT and ECTS compare before starting the engine after it has set for 5 to 8 hours?

This very well could be a consequence of all of the oil you are burning.

metz281
12-19-2008, 06:19 PM
Okay, so in summation you bought it recently (so you don't know much of its history) and it runs rich and fouls plugs, also has erratic idle and makes chugging noise while accelerating. Good find on the Evap purge vacuum leak, not an easy one to uncover. :) Definitely a contributing factor. You've covered just about everything electrical one could possibly think of (and then some), so I'm thinking it's mechanical. Being as you've covered the exhaust (I think) and replaced the FPR, isn't valve timing the only thing left? My guess is, and THIS IS ONLY A GUESS, valve timing may have jumped, or some Yo-Yo (prior to you owning) replaced the chain and didn't do so good. Twin cam motors are sensitive to the int-exh being in sync or the combustion gets all fouled up. OldNuc suggested this in post #35, I just second it.
2 things to check: Is the motor noisy in the chain area? Are there signs of being opened recently. The factory silicone is gray, most shops use black, orange, whater is lying around.:ugh:

Only takes a few minutes to check (by removing V/C) alot long to fix. 96 should have the plastic one with the reusable rubber gasket.

OldNuc
12-19-2008, 07:00 PM
Cranking it around and checking the timing is not a bad idea. I think the compression was even but a bit weak. It eats oil and that will cause problems.

cboss
12-19-2008, 09:05 PM
OK, some details which may help others see what is going so far.

The car is 1996 SL2.

The engine runs very good, IMO and on the highway it runs sweet (nice and quite and smooth). The gas mileage appears to be very good. I estimate some where from 35 to 40 mpg.

The CAT is newly installed, but the resonator or muffler could be clogged, so I can't rule out the need to replace them if the exhaust is restricted. The car didn't have a CAT on it when I bought it, but is what installed immediately. If the car had been driven without a CAT and running rich, I would not rule out the resonator and muffler being fouled up.

The car has 175 K miles on it, so burning oil would be expected and runnng rich there is likely a lot of carbon buildup (I have plans to progressively clean the engine, do MM piston soak, possibly Auto-RX).

Each part that I have replaced has made a difference and when running above idle it runs very good.

The parts installed:

- NGK plugs
- new spark plug wires
- new MAP sensor
- new IAC
- new ECTS
- new TPS
- new NTK O2 sensor
- used ignition module and coils from a 2001 SL2 (compatible)
- new air filter
- new fuel filter (no builtin FPR so not a problem if not OEM)
- new evap solenoid
- new PCV (not-OEM)

I installed a new non-OEM FPR from Advance and it ran at 41 psi so I brought it back to the store and asked for my money back. I told them, if it runs at 5 psi higher than the max allowed, the part is not acceptable IMO. I got my money back. You can't advertise something as OEM replacement, when it truly is not.

I installed a used FPR from a 95 SC so it isn't considered replaced yet. The old FPR was running at 37 psi (slightly high). This FPR is also running at 37 psi. The new one (OEM from getstaturnparts) will come monday. I expect a significant improvement when it gets installed (nothing other than OEM is acceptable for the FPR). So until that gets installed, the fuel pressure is not yet under control.

Every thing I have done has improved how the engine runs.

Not done yet.

The removed the throttle body and cleaned it good and installed a new gasket. While the IAC air path was not clogged, cleaning helped some and I think the gasket helped (could have had a slight vaccum leak).

What really helped, was cleaning the MAP sensor hole in the TB. It was badly gunked up. Also I sprayed cleaner through tghe vacuum line fro the FPR and fixed the line when it broke off at the elbow connector. I sprayed cleaner through the PCV valve line as well. Cleaing the vacuum line for FPR, PCV and MAP sensor are critical and made a big difference. I have a new PCV valve coming with the new FPR which is OEM and I believe that will help too.

I have a feeling when the FPR's get old, they can fluctate at times. They work, just barely above acceptable limits and then possible because of air pressure changes, it has intermitant problems. I don't think they have to completely fail to be a problem. The new OEM FPR should make a big difference and I will report back when I get it installed.

When I got the car, it has platinum 2 plugs installed (and they were black). I installed platinum 2's are first, before I came across the Saturn forums. When I learned about the problems with these plugs, I changed them for NGK's.

I took out the plugs yesterday and cleaned them, checked the fuel pressure (37 psi and constant) and so far no SES light. I think my mistake of bleeding the fuel pressure gauge while the engine was running, may have caused the FPR to fluctuate. When I tested a few days ago, is was first at 37 psi and after bleeding the line while running it started to fluctuate. This may indicate the old FPR's may weaken and have a tendency to fluctuate intermitantly. If the internal diaphram is weak with age, but not broke, it is possible that slight changes (ie. air pressure changes or anything else that could effect it), may push it over the edge and it startes to fluctuate.

Not the ECTS I installed was bought from AutoZone (or Advance) and I am not sure whether there is much difference between them and OEM. That is a consideration.

OldNuc
12-19-2008, 09:14 PM
The variation in any 2 ECTS sensors from any source will be noticeable when at room temperature but becomes insignificant when at normal operating temperature.

The oil burning does cause problems though. You can try all of the possible fixes as they do not cost much and may do some good. Auto-Rx does work but does cost more. the resonator is a straight through pipe so its not a restriction but the muffler could be.

cboss
12-19-2008, 09:32 PM
Continued ...

The problem with comparing the temperature of the ECTS compared to the AIR temp sensor is that it has to be done when the air temp is consitantly the same for some time.

In the winter, the nights are very cold and as the day warms up the air temp sensor will read a number of degrees different than the coolant sensor. The reason is that the engine being metal and water (coolant) take longer to warm up than does the air (if the car is outside). The engine holds the cold.

The air temp needs to be consistant fro a number of hours before comparing them. When I do that compare in late afternoon), the temps are the same.

Since the PCM is basically a computer (electronic) it is very sensitive to voltage drops because of bad grounds.

I cleaned and replaced the connector for the front PCM ground and that helped. I haven't done this with the PCM ground on the back side yet, but plan to and that may help.

As far as I am concerned, the engines actually runs very well. It is quite and smooth, especially cruising on the highway. It has great pick and acceleration and power. I can go up hill (winding rural hilly roads) at 55 in 5th gear (manual). Thats good power! I can't do that with my 1990 Geo Prizm (5 speed manual).

This is a well running engine.

Now there is a problem at idle. It does idle good (RPM at about 850). It doesn't stall or die. It doesn't rev. It is reasonably quiet.

Now I am a bit of a perfectionist and so I sense what I call a slight panting or chugging (most people would not even care). It is not loud or overly distracting. I just know that it means the idle is not quite perfect and there is some problem being overlooked, which is better to track down rather than ignore. Also the OBD-II scanner, shows when at diel the timing advance fluctuates, rather than stay at a fixed (or close to fixed) level. This obviously indicates that there is something amiss, which should not be ignored but tracked down.

The engine is in basically good condition, IMO. It is as smooth and powerful as my Geo Prizms engine was when I bought it ten years (it had 149 K miles on it). My Geo now has over 330 K miles on it and only recently have I has it started to get a tick (noise). It still runs great on the highway with plenty of power. It burns oil (the valve seals most likely and possible gunked oil rings), but it still has lots of power. My Saturn should be able to give me another 100 miles before a rebuild is necessary.

I can handle the oil usage and have plans to work on that to improve it.

I am willing to do some engine work on it, if necessary, but I will exhaust all other efforts first before I go that route. If the engine is basically sound, there is a lot that can be done without a major rebuild.

The OBD-II scanner is a critical tool in taking care of the engine. Some people may be happy just getting rid of the SES light. I am using the scanner to see how the engine is really running (live data) to fine tune it, not just get rid if the SES light). Since codes like P0172 can occur when the engine runs outside of its margins (limits), fixing one thing or another may push it back within margins for awhile, but often the problem returns. I am trying to go through the system one key are at a time to find what most effects the running rich and one thing at time cleanup.

I am patient and will continue to diagnose the engine, one thing at a time. As a computer programmer (my trade), I am very adept at diagnosing problems, one step at a time. The key is patience, test, test, test and get valuable data (need input). The scanner has been a definite worthwhile purchase. This is my first Saturn, and I have already figured out more about the car and whats wrong with it, than my nearby mechanic (and he's good) was able to. My impression is that many mechanics today are not that adept in diagnosic work.

OldNuc
12-19-2008, 10:15 PM
Before starting the car in the morning just turn it on and read the IAT and ECTS. They should be withing 5C. The PCM preforms this check every startup from cold and uses these values for several checks and calculations. If they are not within 5C it can/will cause problems. And I would have to go dig out just exactly what goes wrong but its worth the effort to verify that it is within those limits.

Most mechanics do not even try to understand the information you can get from a generic scanner. This is why engine construction is fast moving to the throw away designs. Its assembled once and if it needs to come apart its junked and replaced.

cboss
12-20-2008, 01:19 AM
If you have (or currently are) experienced the P0172 code (bank 1 rich), please feel free to get into this discussion.

This discussion is not just aboiut my current problem with P0172.

Its for the benefit of all who are dealing with this code.

I have researched the internet a good deal about this code and often I find most discussions about end abruptly with no solution posted.

For once it would be nice to have a definite set of steps for diagnosing this problem. My experience currently offers a good sounding board for discussing this problem. What ever I learn I will post here and the comments of others offer valuable insights. Sometimes it is the little things we overlook.

One good example, was the idea of cleaning out the vacuum line to the fuel pressure regulator. Or the need to clean the hole where the MAP sensor connects too.

Join the discussion and let's build an extensive database about P0172 diagnosis.

I would like to thank especially OldNuc.

Your insights and experience have been invaluable!

metz281
12-20-2008, 08:31 AM
Okay forget the timing chain for now, I'm thinking injectors. You live in a cold climate, cold fuel does not atomize as well. So if those injectors aren't spraying perfect, properly, fuel will kinda drip/ rather than spray and that burns best. If you want an example think of the feedback carbs, with the preheater plates at the base and tinfoil tube going to the air cleaner to help warm the air and help with atomization. I wouldnt worry too much about a few psi on fuel pressure, it takes close to 60 psi to make it force more fuel in there. You could try 4 injectors, but I would suggest 4 from a low mileage donor at a pick and pull (if there are any in your area). I've never had the problem with 172, but I'm farther south, makes me wonder is this a northern issue? With the mpg and performance you describe, its odd that the pcm sees it as a problem. Good luck, and I'm no expert, I'm just learning like the rest of us.........

Another thought, can you get to the pcm and see the service # make sure its right, like maybe someone put sohc (pcm) in dohc car?

OldNuc
12-20-2008, 09:33 AM
With regard to injectors. They are an electromechanical device and as such will wear out. the nozzle will erode. Data on actual service life of the Saturn injector is non existent except anecdotal information bases on limited experience.

When looking for a solution to poor performance in a well used oil burning engine I decided as the injectors had many hours of operation and being aware that the industrial service life for these types of devices would be less than the hours I had on the installed set (140,000 miles) I would replace them. In the interest in collecting data I found that the newer designs have multiple nozzle holes as compared to the original injectors which have 1 larger nozzle orifice. As you point out above, atomization is important. The smaller multiorifice injector was developed to provide improved atomization of fuel.

Armed with this information I went searching for a replacement. Found that BOSCH made an injector that would fit and function that also had multiorifice nozzle configuration. Bought them here: http://www.fiveomotorsport.com/Injector_SetsSATURN.asp Bottom of page, right hand side. The result of installing these was and is improved fuel economy and smoother running. These were not installed to chase a P0172 though.

When I got into the P0172 search I found the problem was excessive fuel pressure and the cause was using a lower cost filter/regulator unit. Restoring the OEM unit fixed it. The important point here was that the fuel trim was driven to the negative limits for both short term and long term at almost all operating regimes and became excessively rich during deceleration.

As to cboss problem. The engine is burning and pumping oil. Oil in the exhaust interferes with O2 sensor performance. The oil contaminates the cell and interferes with the response and accuracy. This is just the design of the cell and shield. During deceleration the injectors shut off completely and combustion temperature drops. The oil burns as a lower octane fuel and when mixed with the fuel also serves to lower the octane of the resulting fuel. this will impart timing and O2 sensor performance. As you have read above the P0172 code comes in with a vengeance following a high speed run and return to idle and seems to have a relationship to atmospheric conditions. The O2 sensor relies on the atmosphere as a reference to provide the O2 signal to the PCM.

The cause of this P0172 is most likely not a sensor failure or degradation but primarily the result of excessive oil use. Its an obvious problem and there is the possibility that solvent cleaning of the rings will reduce it somewhat and that may solve the problem.

On an other note. The program in the PCM firmware and associated tables is both encrypted and not published except in the most general terms. Access to that firmware routine and the data tables would make solving this problem quite easy.

metz281
12-20-2008, 12:38 PM
Ive had 6-8 saturns and only one did not use oil others were qt/ 300/500/800. Never had the 172. Heck the one I just built has hollow converter and the 420 hasn't come on yet. Thats why the question. He's already beat the electrical side to death, so its almost got to be mechanical, unless theres a TSB for pcm REFLASH or something. Seems OBD monitors were a little more laxed in the early days unless something was seriously wrong. But hows he getting the claimed mpg if its fouling plugs and running rich?

cboss
12-20-2008, 05:51 PM
Ok,

I have been plugging away at this for possibly a month now, so I can say I have some experience with the P0172 code.

At first I just started replacing stuff that "may" be associated with the problem. Before long the money adds up and you get to the point that you have to step back and stop "assuming" things and start analysing things.

The OBD-II scanner, IMO was critical to this change in method.

Now everything I do I am examining the effects on the timing, O2 sensors, fuel trim, MAP, etc.

I have significant improvement so far and much of what I have done has made a big difference.

Well a few days ago, the P0172 code came on and stayed on. Couldn't get rid of it.

I waited a day and then cleaned the plugs, added some oil and then did a fuel pressure test again. This time the fuel pressure was close to normal (37 psi, slightly high but no fluctation).

Gut feeling:

The fuel pressure regulator is suspect and rather than it just fail, I think it may be simply acting poorly at times. I haven't cut one open (both I have appear to be OEM, but are simply very old, the original on the car and used one I picked up at a JY) yet to see how it works internally, but there must be some kind of diaphram since it works with vacuum.

Rather than the FPR simply fail, old age may simply cause it to not flex properly, so external factors could intermitantly cause it to act crazy. The air pressure (weather), my bleeding the fuel line while engine running and pressure guage connected, etc.

Remember, I did installed a new FPR awhile back, but it was a non-OEM from Advance and when tested, it ran at abbout 41 to 42 psi, which is about 5 to 5 psi above the maximum allowed. Aftermarket FPR's are just plain trouble. Also those who have cars with the embeded FPR in the fuel filter have the same trouble because the FPR is in the fuel filter. I don't think the fuel filter itself is a problem, but the FPR is.

Lets' see what different an OEM FPR makes on Monday, when I get it.

Next, I installed an aftermarket PCV valve. This contributes to problems with the oil burning. I am getting an OEM PCV Monday too, so lets see what difference that makes.

Now as far as the excessive oil usage, the exhaust is quite clean, so I doubt I have significant leakage at the valve seals. My gut feeling is that the oil rings (and PCV) are the primary contributors.

Now remember, this car had been running very rich when I got it (plugs black as coal) so it had a history of running rich. It also has Platinum 2 plugs in it which contributed to the problems.

The NGK's I installed are staying a bit cleaner (still some black from running rich, but better).

I did a engine flush when I first got the car (Gunk) and then new oil and new filter. I added Rislone the last round when I added oil, because it can help clean rings, whereas most oil additives don't say anything about effects on the rings. I have gone through a few quarts of oil so far (the car was using a quart for less than 200 miles - I Know scary!). I felt though the engine was running very well and sounded good and appears to be in good overall condition ( should know what a good engine sounds like, since I bought my Geo Prizm with 149 K miles on it and I now have over 330 K miles on it).

Rather than look for a quick fix for the oil (haven't done a MM soak yet), I am going the route of slow and consistant improvement.

I have started adding oil (supertech from walmart, so nothing fancy with it) with a small amount of Marvel Mystery oil in (simply add 4 ounzes to the quart bottle of oil and shake good). no matter what critics may say, I have a lot of confidence in MM. The idea is clean slowly.

If my estimate is correct, the last round of adding oil I am now getting about 300 to 400 miles per quart of oil.

The running rich likely has an effect on oil usage and the condition of the oil rings (carbon build up). As the richness progressively decreases and the engine gets cleaned slowly (MM, Rislone), the oil usage should drop.

This may be way out, but common sense (and a lot of what I have read) would indicate that cleaning the engine out should take place slowly and this requires something simple (which I plan to do).

I am going to start changing the oil and oil filter at very short intervals (rather than 3000 miles). The idea is to keep new oil at all times so the gunk cleaned out gets totally removed from the system. For now a better oil filter (cleans more out) (ie. 99%) would be better, but change the oil and filter and run until the oil gets pretty dark (gunk is coming out into the oil) and then change it again and a new filter. Even if I have to change it every 500 miles or so, I will.

This will let the marvel mystery oil do its stuff in clean oil, clean out the gunk and completely remove it and do it again.

True this will cost a few extra dollars, but I think it will make an improvement. The MM works wonders, but you need the engine to be clean so the gunk can go somewhere (rather than be pushed around the engine).

By using simple oil (walmart), rather than the fancy stuff, then you don't have a lot of oil chemicals competeing with the MM and the oil is cheaper so I can change it more often.

I already have an significant improvement in oil usage and I haven't even done a piston soak yet.

I would like to get the oil usage to a more acceptable level using this way (lets' say at least a quart per 1000 miles) and then try something like Auto-RX.

There is no bad noises from the engine to suspect the timing chain or other mechanical stuff.

Now as far as the P0172 code, I believe this code can often be the combination of multiple factors, rather than one big problem. Its kind of like the domino effect. One problem pushs another which pushs another.

The "easy to do stuff" first list which I posted, is a good start in developing a game plan for dealing with P0172.

You would be amazed how much the simple things have made a difference so far.

cboss
12-20-2008, 06:06 PM
One reason I encouraged others to get into this discussion, especially those who have dealt with the P0172 code or are dealing with it now, is that the more data brought to the table on the subject the more likely a detailed plan can be devised fro dealing with the code.

I have scoured the internet (non-Saturn sites as well) about this code and it is amazing how little info on solutions there are. Yes, there is the standard ' this code may be caused by the following stuff, but rarely have any online discussions followed through from beginning to end and offered details on how the problem was solved. Saturn users are not alone in this. All makes of cars deal with this.

The way the computers work is basically similiar on all cars. The basics to engine design are not all the different on different models. There are common factors.

I have learned so much about how my engine works and how computers work from this adventure, that it was actually worth the experience. I have a greater appreciation for how the computer works and how so many things are interrelated.

Rather than looking for one broke part, one must examine how everything is related to each other.

A car which is over 10 years old (my Saturn is 12 years old) has to have a number of systems which have age related deteriation. Cars built in the last 20 years are far better built than one may think. Its amazing they last so long.

Older cars need to be nursed back to health slowly and methodically.

Back to proper testing.

Fuel Pressure gauge invaluable
OBD-II scanner (with live data) invaluable
Volt Meter (for checking wiring or sensor/coils resistance) invaluable

Maybe its the computer programmer in me, but I need data and lots of it.

No more assumptions.

I need real data.

If something improves because of some fix, then I need real data to prove it.
If something helps oil usage, then I need data.

Note: Every time I add oil I top it off to full mark and then reset the trip mileage meter on the car back to zero. I track the exact miles between adding oil to see if any changes in how far I got before I needed to add more.

No assumptions!

Like I said, I have a significant oil usage improvement and no piston soak yet.

OldNuc
12-20-2008, 06:48 PM
Ive had 6-8 saturns and only one did not use oil others were qt/ 300/500/800. Never had the 172. Heck the one I just built has hollow converter and the 420 hasn't come on yet. Thats why the question. He's already beat the electrical side to death, so its almost got to be mechanical, unless theres a TSB for pcm REFLASH or something. Seems OBD monitors were a little more laxed in the early days unless something was seriously wrong. But hows he getting the claimed mpg if its fouling plugs and running rich?

The fuel trim numbers look fine. It comes up with the P0172 after a decent run and return to idle. the only thing that sets the code is the short and long term fuel trim at the low stop for 3 seconds I believe. And about the only thing that can dynamically change fuel trim is the O2 sensor. The O2 sensor is sensitive to oil contamination.

Its interesting.....

OldNuc
12-20-2008, 07:00 PM
cboss

We will see what happens when you change the regulator, and PCV. The closing of the throttle shuts down the fuel totally and that raises manifold vacuum until the EGR opens which lowers it and this may cause the regulator to fail to control. Oscillating pressure will drive the engine rich. An engine can consume a prodigious quantity of oil and not smoke noticeably. The oil does significantly reduce the octane of the fuel and that results in a retarded spark timing. And retarded timing will make an engine run rich. High manifold vacuum sucks more oil past the rings and down the intake guides also. Might want to consider sending the injectors out for a good cleaning and flow balance.

cboss
12-21-2008, 07:25 PM
Drove the car yesterday. No code.

Drove the car 180 miles today on long trip. No code and I estimate I got 40 mpg (only a rough estimate, could be a little less).

The engine runs sweet on the highway, nice and quiet and smooth.

Look forward to getting the FPR and PCV valve tomorrow.

OldNuc
12-21-2008, 07:57 PM
These are good signs. What is the oil consumption? Each diamond on the stick is 3.2 oz of oil

cboss
12-22-2008, 10:58 AM
The oil usage is improving (meaning use less oil).

I need a little while to document it.

I set the trip counter each time I top off the oil and then drive a certain amount until the level drops to certain level on the dip stick and then I calculate the oil usage. I am in the middle right now of such a cycle. When I get to a certain point, I'll post the latest oil usage.

I just got the FPR and PCV today.

The weather turned real cold (10 degrees), so I may have to wait for a warmer day before working on the car.

The FPR looks like the one that was in my car originally and the one I got from the JY (both of them ran at about 37 psi). Definitely OEM (I got it from getsaturnparts).

Now the old FPR's ran at about 37 psi (the Haynes book says range should be from 31 psi to 36 psi). They obviously were running high and possibly intermitantly fluctuated (ie. when air pressure changed).

Just as a comparison, then one I got from advance (non-OEM), ran at 41 to 42 psi right out of the box. I brought that one back and got my money back.

The new OEM FPR should make a difference. I'll post the fuel pressure gauge results, when I install it.

Now the new PCV valve I got from getstaturnparts is interesting.

I can tell there is a difference, just by looking at it.

It weighs about twice as much as the ones you get from autozone or advance. It appears to be heavily built (also says made in USA). I have no doubt it will make a difference.

I would have to say this right now, even before I get to install them, I think both the fuel pressure regulator and PCV valve on old Saturns should be changed out and have new OEM ones installed. No AutoZone or Advance Auto parts for these two parts.

OldNuc
12-22-2008, 11:44 AM
The flow through the PCV is a function of the weight and the spring that holds it open. The original idea was that they were calibrated to maintain a given pressure in the crankcase, usually a negative pressure.

It is well documented that the Saturn is a very picky engine and most "close but not right" aftermarket parts fail to preform.

cboss
12-23-2008, 04:30 PM
First I checked the oil usage.

When I got the car, it used a quart in 200 miles or less.

Oil usage is nearly 400 miles for a quart.

I haven't even done a piston soak yet.

The last thing I used was rislone in the oil and I now add 20% MM to the oil when I add it. As the engine runs better (not running as rich) it may help as well.

I installed the new Saturn OEM fuel pressure regulator today and tested the pressure. It is now 35 psi. Thats two psi less than the old one. It is well within the acceptable range of 31 to 36 psi.

I do think the old FPR's , while possibly not fully failing, may develop an intermitent problem, possibly encouraged by air pressure changes, where they fluctuate. When the pressure was fluctuating, it was bouncing back and forth between 33 psi and 42 psi very quickly. This could explain why the P0172 code may come and go.

The new FPR should rule out fuel pressure in the future if the code comes back. So far no code yet (took it for a quick run for 15 minutes at highway speeds).

I installed the new PCV valve as well.

I topped off the oil again (and cleared the trip mileage meter) so I can track the oil usage with these new fixes.

I will post my results after a few days of driving.

OldNuc
12-23-2008, 04:50 PM
This actually progress.

cboss
12-23-2008, 10:06 PM
I drove the car another 40 miles this evening (highway speeds) and no SES light yet.

The weather is also changing (rain coming).

The idle is much improved.

The ignition advance though (on scanner) still shows some fluctation of the advance during idle. Just give it a little gas and the advance is consistant (no fluctuation). I don't know if it is related to tge P0172 code problem or is a separate problem.

Driving on the highway, the engine is sweet! Very quiet and smooth, like a new engine.

I had replaced the knock sensor (one of many sensors I replaced), but I was wondering today if anyone has ever had problems with the aftermarket version ?

The MAP sensor, IAC and TPS are also aftermarket, but I think they are working fine so far.

It would be good to discuss, which sensors are more prone to problems when aftermarket than others.

I am definitely convinced the fuel pressure regulator and PCV valve should always be OEM Saturn parts. No doubts there.

OldNuc
12-24-2008, 09:59 AM
The ignition advance will shift back and forth a bit at idle. Its normal. You can play with the wires and plugs and it will tend to stabilize. The pair for #1 and #4 should be clipped so as to run parallel to each other all the way up on the head.

Knock sensor either works or fails. The aftermarket TPS does not have a good reputation for being within specs right out of the box, or a long service life. The OEM TPS is only about $10.00 more. The others are probably OK. In most cases the OEM sensor will be a better bet. The ECTS and IAT are exceptions as just about any one you find will be OK. They should be matched as to the measured resistance at the same temperature. A pot of boiling water makes a good test bath. And, that is the range you are interested in, along with room temperature. The pair(ECTS-IAT) have a large impact on mixture strength.

The fuel trim will shift from negative to positive as the fuel pressure moves from the high pressure limit to the low pressure limit. Rarely is actually at zero trim.

nivlem7
12-25-2008, 06:36 PM
I didn't slog through all 6 or so pages of this thread, but just wanted to throw out the possibility of fuel filter, if it hasn't been mentioned already.

I believe OldNuc alerted me originally to the fact that aftermarket filters may give you too high a pressure at the fuel rail - once I replaced mine with OEM, the P0172 code I'd been fighting for a while went away and hasn't returned.

cboss
12-25-2008, 10:57 PM
As far as aftermarket fuel filters, they only pose a problem for the later models where the fuel pressure regulator is built into the fuel filter.

The models 1998 and before, all have a separate fuel pressure regulator which is located in the fuel injector rail.

Thats what my last few posts were about, installing a new OEM fuel pressure regulator. With the models which have a separate fuel pressure regulator, it is not the fuel filter which has the problem, but the fuel pressure regulator, if it is an aftermarket part.

I had previously installed a fuel pressure regulator (herein using the abbreviation FPR) which I bought from Advance Auto. Fortunately I tested the fuel pressure after I installed it. The pressure was about 41 to 42 psi, where as the max for the car is 31 to 36 psi (at idle with vacuum line attached). The aftermarket FPR was 5 to 6 psi higher than the maximum allowed.

I took the aftermarket FPR back to advance and got my money back.

The current FPR I just installed, I got from getsaturnparts.com and its a real Saturn part and it makes a big difference.

For those following this thread:

If your car is a later model with the fuel pressure regulator built into the fuel filter, then do not get an aftermarket fuel filter. Its not the filter part which is a problem, but the built in fuel pressure regulator which causes the problem. It will push the fuel pressure too high. A Saturn OEM part is required here.

If your car has a separate fuel pressure regulator, which is located in the fuel injector rail, then the fuel filter itself can be an aftermarket one (mine is), but the fuel pressure regulator must be replaced with an actual Saturn OEM part. Nothing else is acceptable, unless you try it and test the fuel pressure with a gauge and it is within range.

cboss
12-25-2008, 11:19 PM
I would like to point out an observation about fuel pressure regulators on older cars (ie. 1991 to 1997).

I can't prove this definitively yet, but a theory I have about the FPR's is this:

I think when they get very old, while they may not actually fail outright, the internal diaphram may weaken and intermitantly fluctuate. This may explain why the P0172 code may come and go for no reason.

I had an old one in my car and periodically I would get the P0172 code and I couldn't get rid of it. I would then let the car sit a day, clear the code and then go days with no code, until it returned.

I installed a used FPR (OEM one) from a 95 SL (I got from a JY) and it did the same thing.

I tested both FPR's and then pressure was 37 psi, just slightly above the maximum allowed (which is 36 psi). Both once in awhile would fluctuate pressure when tested, but not consistantly. When fluctuating, it would jump quickly between 33 psi and 42 psi.

If my theory is correct, the FPR has not failed, but is simply weak (thats why the pressure is slightly high) and can intermittantly fluctuate.

The real test now is to see if the code comes back after I installed the new OEM FPR. I will update this thread regularly so you will know how it goes.

Its only a theory, but it makes sense.

I was thinking of getting my dremel tool out and cutting the old FPR in half (carefully) to inspect the insides. It would be interesting to see the actual condition of the internals of a fuel pressure regulator. Maybe it could be dismantled, without cutting it. The key is to see inside.

So far, the critical parts which require an OEM part are:

- Fuel Pressure Regulator (or fuel filter if FPR is inside)
- PCV valve

As far as quality parts (or OEM):

- Oxygen Sensor must be high quality, either OEM or possibly an NTK or Denso

OldNuc
12-25-2008, 11:28 PM
The 98 and newer have the pressure regulator integral with the filter. Its the 97 and older that have the regulator on the fuel rail. As the fuel pump ages the discharge pressure decays to the point that the regulator is regulating with its valve almost closed. This will cause regulator chatter and sticking. This is what you saw. A new regulator will have a new diaphragm, spring and valve which will tend to stabilize the performance. The 98 and newer cars have a higher pressure fuel pump also. You probably have identified the actual root cause of your P0172.

cboss
12-25-2008, 11:33 PM
Thanks OldNuc,

I fixed my post.

I thought the 96 to 98 were pretty consistant in parts. I didn't realize they changed the FPR to a FPR/Fuel Filter combo in 98.

OldNuc
12-26-2008, 09:37 AM
What is significant with the pre 98 cars is that the actual fuel pressure to the injector is a function of the manifold vacuum. This changes the amount of fuel delivered to the engine as the manifold absolute pressure changes as long as the injector pulse width is constant. This would imply that tight control of fuel pressure vs. manifold absolute pressure is critical to proper performance. Much more so that the 98 and newer cars.

This also would lead to the conclusion that the fuel pressure regulator and PCV valve are critical components for the 97 and older cars.

Fuel pressure control is also important for the 98 and newer cars but the PCM may be more adaptable to variations in the PCV performance.

Conclusion would be that an OEM regulator or filter/regulator is recommended. And an OEM PCV is recommended for the 97 and older cars and a good idea for the 98 and newer. There is an aftermarket fuel pressure regulator available for the 97 and earlier cars that may be an acceptable replacement.

cboss
12-29-2008, 09:13 AM
The OEM Saturn Fuel Pressure Regulator was installed on December 22, 2008.

It is now December 29, 2008 and no SES light (P0172) yet!

I'll post every so often so you can know if this has permanently solved the P0172 code problem or not.

So far, so good.

OnDaGround
12-29-2008, 09:18 AM
This thread is still alive? Replace the whole car.

cboss
12-29-2008, 06:19 PM
The car is a nice car worth the effort.
(see photo)

Gas mileage is great (at least 36 mpg, possibly more on long highway trips).
Engine runs "sweet" on highway (smooth and plenty of power).

My last car (which I still have) I bought when it was 7 years old, with 149,000 miles on it (1990 Geo Prizm). I still have it and it has over 330,000 miles on it and still running.

My goal is to get 300 K miles out my Saturn.

OldNuc
12-29-2008, 07:23 PM
Based on the continued positive results (no P0172), I would suggest icing the champagne and scheduling the dancing girls. It is acting fixed.

cboss
01-02-2009, 05:52 PM
While I am a little disappointed, I am not totally surprised.

The P0172 code is back again today.

I do have a feeling the problem with the idle is a clue and I was expecting the problem may return and it has.

The air temperature is cold today (in the mid 30's) and the air is wet.

I drove about 20 miles today around town.

Late in the day the SES light came on.

When I got home, I immediately tested the fuel pressure and it is 36 psi, well within range and no fluctuation.

The fuel pressure test did show up something.

I turned the key on so the fuel pump would pump a few seconds and then watched the pressure to see if it leaked down. It wasn't leaking down before, but now that the SES light is on, the pressure leaked down at least 10 psi in 5 minutes or so. It drops slowly, not quickly, but definitely too fast.

Another test:

Idle the car and clear the code and it immediately comes back on.
Do it again and the code comes back immediately.

Now give it just a little gas to be above idle and hold it.
Now clear the code and it does not come back immediately.
It does not come back, unless I let it go back to idle.

My gut feeling is that maybe the fuel injectors (possibly effected by the cold and dampness) can intermittantly leak.

A leaking fuel injector shows up more at idle (when the fuel should be little) and it may be associated with the idle problem.

Now, the question, is it worth installing new injectors ?

OldNuc
01-02-2009, 06:19 PM
I would send them out to have them cleaned and flow tested. That is a $82.00 + your shipping to Witchhunter cost item from Witchhunter. A total replacement with new BOSCH injectors is $160.00. Refurbed GM is $160.00 and up, and new GM is beyond consideration. There is nothing in a can or bottle that will fix them permanently an as you now have this problem chased down to a very intermittent occurrence the temporary help from a can of Techron may not prove conclusive.

Witchhunter--
http://www.witchhunter.com/index.php4

cboss
01-03-2009, 12:40 PM
Ok,

I cleared the code yesterday so today I would start clean.

I drove the car over 40 miles today (most on highway, some on back roads) and no SES light yet. Idle is a little better than yesterday.

The weather is still damp, but slightly warmer today (but still cold).

I'll see how long before the code comes back.

I am getting the impression there is something intermitantly wrong. I don't know if the weather plays a part or not.

Some questions:

(1) Can aged fuel injectors be affected by the weather to produce an intermittant problem ? (ie. cold or dampness affect how they work)

(1b) (also can injectors exhibit problems intermittantly ?)

(2) If the exhaust (ie. muffler since I had a new CAT installed) is restricted, will it produce worse symptoms when the weather changes ?

(3) When I did the leak down test (fuel pressure drop) how can you tell if it is the injectors leaking or simply the fuel pump check valve causing the slow pressure drop ?


(4) When a set of fuel injectors get old do they tend to all exhibit similiar problems (ie. leak) or do you usually only have one or two have problems at first ?

Lastly:

(5) Can old fuel injectors (with problems) cause the idle to be poor and the timing to fluctuate at idle ?

I do think tracking down the problem with the idle will lead me to the true cause of the P0172 problem or at least a contribitor to it.

OldNuc
01-03-2009, 02:30 PM
Ok,

I cleared the code yesterday so today I would start clean.

I drove the car over 40 miles today (most on highway, some on back roads) and no SES light yet. Idle is a little better than yesterday.

The weather is still damp, but slightly warmer today (but still cold).

I'll see how long before the code comes back.

I am getting the impression there is something intermitantly wrong. I don't know if the weather plays a part or not.

A bad IAT sensor could be causing the problem.


Some questions:

(1) Can aged fuel injectors be affected by the weather to produce an intermittant problem ? (ie. cold or dampness affect how they work)

Yes, they are both electrical and mechanical. See Witchhunter site for a good description of how they work.


(1b) (also can injectors exhibit problems intermittantly ?)[/qote]

Yes, not often but its possible.


(2) If the exhaust (ie. muffler since I had a new CAT installed) is restricted, will it produce worse symptoms when the weather changes ?

Low probability. The muffler if its restricting usually does this all the time.


(3) When I did the leak down test (fuel pressure drop) how can you tell if it is the injectors leaking or simply the fuel pump check valve causing the slow pressure drop ?[/quote]

Pull the fuel rail and look for a leaking injector. Set them on paper towels and look for a wet spot.


(4) When a set of fuel injectors get old do they tend to all exhibit similiar problems (ie. leak) or do you usually only have one or two have problems at first ?

They age individually and exibit different symptoms.


Lastly:

(5) Can old fuel injectors (with problems) cause the idle to be poor and the timing to fluctuate at idle ?

Yes


I do think tracking down the problem with the idle will lead me to the true cause of the P0172 problem or at least a contribitor to it.

Yes, and the SOHC motor is known for sticky intake valves at low RPM in engines that are consuming oil. The 20,000 mile fix is to slowly suck a can of Chevron Techron injector cleaner in to the running engine through the PCV hose. This is done while maintaining a 1500 RPM idle. Once you know the valve stems are clean you can look further. A valve that is closing slowly at idle will cause problems with smooth idle performance.

cboss
01-04-2009, 04:24 PM
I drove the car for about 35 miles today and no code.

All of sudden the code returned (stop and go driving around town).

I took some freeze frame data and cleared the code.

I drove then about 25 more miles, most on the highway and the code appeared again on the highway.

I did notice a significant decrease in power (especially on hills) and the car hesitates even on the highway. This is more prominant than any time before.

When I got home, I left the car running and tested the fuel pressure. It was 35 psi (and no fluctuation).

Turned the car off, cleared the code (got a freeze frame first) and started it up. At idle the SES light came back immediately (doesn't want to go) and I got another freeze frame and then cleared the code.

All the Scanner Monitors read OK!
(Misfire, Fuel System, Comp.,Catalyst, Oxygen Sensor, Oxygen Sensor heater, EGR all OK)

Here are the freeze frames:

Freeze Frames ____ #1 _____ #2 _____ at idle
-------------------------------------------
Calc Load ________ 49% ___ 50.5% ___ 27.8%
Coolant (degrees) _ 201 ____ 201 _____ 212
ST Fuel Trim _____ -1.5% __ -2.3% ___ -7.8%
LT Fuel Trim _____ -19.5% _ -21.8% __ -21.8%
MAP ____________ 17.4 ____ 20.6 ____ 11.8
RPM ____________ 2868 ____ 2231 ____ 842
Speed __________ 24 mph __ 52 mph __ 0 mph
TPS ____________ 15.6% __ 16.8% ___ .3%

I strongly suspect the fuel injectors!

Likely buying those new ones (Bosch) for $159 is the best bet, rather than simply clean them.

One other thing:

I did a leak down test with the fuel pressure gauge.
Turned key on.
Got about 40 psi.

Waited to see if the pressure dropped. It dropped slowly.

It dropped 10 psi in 3 1/4 minutes.

Unless the check valve was slowly leaking, it would be the fuel injectors leaking. In the past, I have tested it and the pressure did not drop at all even after 10 to 20 minutes (when the car was running better) so at least then the check valve was fine.

Considering the drivability problems today, along with the pressure leakdown being too quick, it would appear to point to the fuel injectors.

cboss
01-04-2009, 04:47 PM
Oh one other thing.

I think I may be able to rule out the weather having an effect.

Where as before, I got the code when the weather was very cold and wet, today:

The weather was actually 60 degrees (quite hot compared to what it has been recently) and very dry and the P0172 code came on and won't leave.

OldNuc
01-04-2009, 05:47 PM
Ruling out atmospherics clears a bunch of sensors. It would be nice to see the powertrain control manual for your car as the later ones only set the P0172 code based on fuel trim. Your short term trim is a bit rich and the long term is at max correction for a rich condition. If the short term goes to about the same percentage as the Long Term that would do it right there. It very well could be the injectors hanging up. I would almost pull the rail and look.

cboss
01-04-2009, 08:59 PM
One question about fuel injection on the 1996 SL2 :

The Haynes book calls the engine a MEFI (multiport fuel injection).

How does the PCM fire the injectors on a Saturn Multi-port engine:

(1) Each injector syncronized to each cylinder ?

(2) A batch of injectors fired together (ie. a pair) ?

(3) All the injectors fired simulataneously ?

I need to know this to better understand how the injectors effect the overall richness.

If each injector is syncronized to each cylinder, then to have too much fuel would require an injector leak.

If all the injectors fire simulatenously, if one or two injectors are clogged (cause a lean condition), the computer will enrich the entire system to compensate.

I am wondering what the original condition of the spark plugs may tell me (when i first got the car).

I pulled all the plugs and from left to right this was there condition:

- Very black soot
- Very black soot
- black soot not as bad as first two, but still bad
- slightly sooty (cleanest one of the bunch)

What does the plug condition tell me about the fuel injectors (assuming they are the problem) ?

OldNuc
01-04-2009, 09:42 PM
The injectors are synced to the cylinder. One of the functions of the CPS and the phantom Cam Position Sensor.

To gain any meaningful information from spark plugs you install a brand new set and then make a performance run. Come back in and pull the plugs and read them while still too hot to touch. The pictures in the book are gross fail pictures. It is possible that you are having injector problems though as that does cause a black fouling. The oil consumption does not help as it also shows up on the plug. Depending what is in the oil it can leave white deposits on the center electrode. I would pull the fuel rail and see if there is a leaker in the bunch. I suspect that a new set would help but I would want to be sure myself before spending the $$$. If they are no leaking then you have to ask, are these worn out from all of the mechanic in a can that has been dumped through them? They are old enough that on age alone you could justify replacing them.

cboss
01-05-2009, 08:54 AM
OldNuc,

The new Bosch injectors found at:

http://www.fiveomotorsport.com/Injector_SetsSATURN.asp

Are they good ? (BOSCH - 19# fit 96 SL2)

The price is right ($159).

OldNuc
01-05-2009, 09:42 AM
I have 4 of them installed and they are working just fine. They are of an improved design and the nozzle has 4 holes instead of 1 larger one. This provides improved atomization of the fuel.

cboss
01-06-2009, 08:58 AM
Question about fuel trim:

I understand how the computer uses the O2 sensor to check the fuel mixture (rich or lean), but before the computer calculates the fuel trim to compensate, what factors determine how much fuel is being injected ?

- Fuel Pressure (fuel pump and pressure regulator)
- MAP sensor (depends upon calculated load using air pressure testing)
- TPS (how much gas requested by the accelerator pedal)
- ECTS (engine coolant temperature)

What else ? .......

Going by the list above, since I have already installed a new Pressure Regulator (OEM) and done a pressure test (35-36 psi) at idle, installed a new MAP sensor, new TPS and a new ECTS would that not only leave the Fuel Injectors left to cause a rich fuel condition ?

Now assuming the fuel injectors are at fault (for discussion sake) , how can injectors cause a rich condition ?

I understand how they can be clogged and restrict fuel (lean), but how can bad injectors cause a rich condition ?

OldNuc
01-06-2009, 10:04 AM
There is a default fuel map in the PCM. This is where it starts. The trim is the deviation from the base table. This why its important to verify that the PCM is the correct one for the car and any later firmware updates are applied. The 96 year is the PCM with the tightest limits.

There is 1 short term fuel trim cell which is used to update 1 of the 4 long term fuel trim cells. The long term cells are Idle, Deceleration, Cruise, and Acceleration. The fuel trim window is 101 to 147 (-21% to 15%) and 128 = 0%. The P172 will set anytime the following condition exists: the average of the idle and cruise long term fuel trim is 101 (-21%) and the short term fuel trim is below 101 (any negative value%) for 3 seconds.

cboss
01-06-2009, 10:25 AM
Would it be advisable at this point to bring the car to a Saturn dealer to put it in their computer scanner and have them update the PCM ?

I don't want them doing any work on it (I'll fix any problems).

OldNuc
01-06-2009, 11:30 AM
Before spending $$$ check this site. Start with the VIN of your car on the left side box. This will work down to a list of the various calibration updates. See how many there are. More than one would indicate that an update is probably available. It usually takes a couple of tries to get through this.

http://tis2web.service.gm.com/tis2web

cboss
01-06-2009, 05:27 PM
Only one item showed up when I went to that web site:

Misfire Diagnostic enhancement (Bulletin # 97-T-30/20A)

I read the 97-T-30B (overrides A) and it does not deal with the P0172 code.

OldNuc
01-06-2009, 06:38 PM
How many CAL IDs came up in the final summary?

cboss
01-06-2009, 09:05 PM
There were two CAL IDs:

21008856
21008831

They appeared to be treated though as one item, since all the other columns had one item listed and it was all blocked together.

OldNuc
01-06-2009, 09:14 PM
I suspect that one is for the auto and one is for the manual. More than 2 would indicate a revision. So, a reflash would just make the dealer happy. Have you ever seen that long term idle fuel trim come off of -21%?

cboss
01-06-2009, 10:08 PM
The LTFT has never gotten higher than 21.8 to the best of my knowledge.

There is something you commented on awhile back, that I would like to bring up.

You once mentioned that the calculated load appeared higher than normal.

If you look at an earlier post where I posted three freeze frames (for the P0172 code), do any of the calculated loads appear too high ?

If the calculated load is too high, and I installed a new MAP sensor and cleaned the hole it connects to, what could cause the higher than normal calculated load ? (I assume it is calculated based on the MAP sensor)

OldNuc
01-06-2009, 10:31 PM
I finally decided that the cause of the increased load indications were the data was being taken while accelerating. I was looking at a possible problem with the IAT sensor. I think you told me that cold it was close to the ECTS. That's how you check it. The limit is 5C.

cboss
01-07-2009, 06:13 PM
I order a set of Bosch Injectors today.
$159 for a set of four is really fair price.

The evidence tends to lean towards problem injectors.
Besides, a car with 175 K miles on it, with what appears to be the original injectors, its about time for new injectors.

I'll keep the old ones and when I get a few extra dollars, I'll consider having them cleaned (for backups in the future).

OldNuc
01-07-2009, 06:29 PM
After removing the old ones squirt some very light oil or LPS-1 into them and put them in a plastic bag. That will keep them from sticking. Don't use WD-40.

At that mileage you are right, they are ready to either replace or clean.

laser3kw
01-07-2009, 07:09 PM
Just an obtuse thought, Does the pcm look at speed sensor as part of a load calculation? would the speed sensor contribute to the problem?

Cheyne
01-07-2009, 07:18 PM
I order a set of Bosch Injectors today.
$159 for a set of four is really fair price.

The evidence tends to lean towards problem injectors.
Besides, a car with 175 K miles on it, with what appears to be the original injectors, its about time for new injectors.

I'll keep the old ones and when I get a few extra dollars, I'll consider having them cleaned (for backups in the future).

I installed these same injectors just two weeks ago. Very nice. :yes:

cboss
01-07-2009, 10:10 PM
Cheyne, did you have any problems (ie. poor idle, hesitation) with the old injectors ?

Did you install the new Bosch injectors to fix a problem or simply to improve the engines running ?



As far as calculated load, it appears the MAP sensor is key to calculating load.

Some interestign info on this web page:

http://www.aa1car.com/library/map_sensors.htm

OldNuc
01-07-2009, 10:24 PM
The engine RPM, temperature, the intake temperature, and the MAP signal tell you how much air is being pumped into the engine. The TPS is used to confirm the MAP reading. This is where the load comes from. The only sensor you have not changed is the IAT and if the connector is good, which it should be and it is within 5C of the ECTS reading before you start the car from cold then its probably good. Hook up the scanner and play with the wiring and connectors on the ECTS and IAT. You just might find an intermittent connection.

Cheyne
01-08-2009, 07:13 AM
Cheyne, did you have any problems (ie. poor idle, hesitation) with the old injectors ?

Did you install the new Bosch injectors to fix a problem or simply to improve the engines running ?

I had a very minor misfire that I was trying to diagnose and wanted to rule it out. I knew that the injectors certainly needed cleaning or replacing at 170K. Since I'm far away from witchhunter, I decided to replace so that the car was not down for a week or two (daily driver). Now I have the spare injectors to send off for cleaning if needed later.

Ended up being the wire dress and some wire clips over the #3 plug solved the problem.

Cheyne
01-08-2009, 08:45 AM
It would be nice to see the powertrain control manual for your car as the later ones only set the P0172 code based on fuel trim.

Don't have the 96 FSM, but the 97 is also only set based on fuel trim, too rich.

Possible causes listed:
Fuel pressure too high - fixed
Leaking fuel injectors - will be addressed
Vacuum leak around the MAP sensor grommet - did you fix?
O2 sensor contamination - replaced
Restricted exhaust or air intake

OnDaGround
01-08-2009, 09:16 AM
The car is a nice car worth the effort.
(see photo)

Gas mileage is great (at least 36 mpg, possibly more on long highway trips).
Engine runs "sweet" on highway (smooth and plenty of power).

My last car (which I still have) I bought when it was 7 years old, with 149,000 miles on it (1990 Geo Prizm). I still have it and it has over 330,000 miles on it and still running.

My goal is to get 300 K miles out my Saturn.

Well indeed it is... good luck. Sorry I cannot bring myself to read the whole thread to figure out where you are or are not to help you.

OldNuc
01-08-2009, 10:05 AM
The 96 is just a bit different in some respects and that is why I asked. I would be surprised if it were but it is always worth checking.

cboss
01-08-2009, 10:36 PM
Vacuum leak around the MAP sensor grommet - did you fix?


The grommet looks in good condition, but I would like to replace it if I can.

The nearest Saturn dealer is about 80 miles away so not easy to drop by there.

I emailed GetSaturnParts.com and they said it was not in their parts lists, so I can't get it there. You would think it would come with the MAP sensor, but the guy at GetSaturnParts said maybe it comes with intake manifold. He couldn't find it in any parts lists.

Anybody know where I can pick one up (will a dealer actually carry them) ?

OldNuc
01-08-2009, 11:58 PM
Do you have any GM dealer close? That is a generic part. This does not come up on Get Saturn Parts but this is the part number from the parts book, 01635948. Its a standard GM part.

cboss
01-09-2009, 04:08 PM
Thanks OldNuc!

Having the part numbers a big difference.
I ordered the MAP sensor grommet from a local GM dealer. They don't have it in stock, but should have it by monday. It cost $6.91 (w/tax).

Its a small part, but definitely worth getting.

I should get the injectors monday too.

Another question:

Although I installed new spark plug wires already, they were an inexpensive brand from AutoZone (about $17).

I found a good deal on some Holley High Performance Ignition wires for 1991 - 1996 Saturns.

Do the high performance wires make any difference ?

OldNuc
01-09-2009, 06:44 PM
The high performance wire question is a good one. If they are carbon impregnated fiberglass or Kevlar and have a higher quality construction then they very well could be worth the cost. But, carbon wires have a very definite life and with the best of them 50,000 miles or so is about the max. The more times they get plugged and unplugged and manipulated the shorter the life.

The wire wrapped wires under almost any name are smoke and mirrors. The wire is wound on the same fiber-carbon core and then the wire is wound on and to get it all to work it is then coated with a conductive latex mix. There are exceptions to this though. The spark is a high frequency highly damped wave and travels only on the surface of the conductor. Keeping this in mind you can see that the wire wound wires that are then over coated are relying on the overcoat for conduction.

Adequate ignition performance is obtained with the basic fiber/carbon wire. Higher quality construction, stainless steel clips, silicone boots, silicone insulation, and sealed connections between the clip and the conductor are worth the expense over the cheap ones. Generally speaking the acceptable wires are in the 25 to 35 price range. The wire wrapped wires are prone to producing excessive RFI which is very bad in an electronic car. The idea with the wire wrapped wires is the act like an inductor. If they are over coated that does not happen. The only ignition wires that function with electronic car are either the wire wrapped or the fiber/carbon. Like spark plugs this also generates many claims supported by nothing but warm air and bold claims.

Some of the people running modified cars will add their preference/experience here I suspect and you will have material for further research.

cboss
01-12-2009, 09:40 PM
I haven't gotten the P0172 code for a few days.

Got the new injectors today.

I installed the new injectors.

Started car and it seemed to run OK.

I let it warm up first and then took it for a quick drive. I didn't get more than a few hundred feet down the road and the car choked and the SES light came on. The car ran very poorly, so I came back to the house quickly and tested for codes.

I got a cylinder #4 misfire.

The fuel trim seemed high too.

I checked the connectors to the injectorsd and also took a OHM reading with a volt meter. The Bosch injectors read higher than the original OEM injectors.
The OEM injectors should be 11.5 to 12.5 OHMs.
The Bosch Injectors read 15.8 ohms. I checked the #1 and #4 cylinder injectors and both read 15.8 ohms, so the #4 injectors is probably OK.

I cleared the code and thought about it for minute.

I decided to try something. If the PCM was used to the old injectors and long term fuel trims, maybe that would cause some running problems with the new injectors, so maybe it would be wise to clear all the PCM's trim memory. I disconnected the battery and let the car sit for awhile.

I came back, connected the battery and started it up.

The car ran fine. I took it for a test run and this time it ran quite well and no SES light. I drove it about 20 miles at speeds from 35 (back road) to 65 (highway). Acceration appears to be good. The engine sounds quieter too.

When I got back I put the scanner on it.

The idle sounds somewhat better (say 20% better) and smoother.
The idle is about 850 (when at running temp).
The fuel trims seem a little high at idle, but may be better at higher RPM.

The fluctuating ignition advance at idle is improved. It still fluctuates, but not as much (not as low and not as high).

The idle still seems to pant (Best description I can think of) and the advance is still fluctuating.

There is an improvement with the new injectors, but i don't think it has solved all my problems yet, particularly the idle.

While the CAT is new, I am considering that maybe there is exhaust backpressure from a restirction in either the resonator or muffler. Thats my next area to test out.

I need to jack the car and visually inpsect them (see if they rattle at all when banged lightly with a rubber mallet) and too look for any dents in the pipe.

I read somewhere than some pipes are double layered and the inside layer can some times peel off internally and restrict the exhaust.

I didn't get the new grommet for the MAP sensor yet.

OldNuc
01-12-2009, 09:52 PM
Yes. You have to dump the trim tables or it is a bit entertaining when you change major components. The resonator is a glorified cherry bomb without the fiberglass. If there are no holes in it its fine. The muffler can be the problem. If yours is held on with a clamp then its been replaced once at least already.

A you have seen an improvement with the injectors that should move you in the right direction. if it ever warms up enough that I think its fun to be in the garage i will run a scan and remember to look at idle advance and fuel trim numbers. Its too cold to run the car with the door closed and I burn about $20.00 in propane and diesel fuel getting it warm enough to be habitable. I do have work to do so it will happen.

cboss
01-13-2009, 03:57 PM
I installed the new MAP sensor grommet today. The old one comes out with the MAP sensor when you remove it ( meaing it is loose). The new grommet is a lot tighter.

The P0172 code came back today and keeps coming back even after I clear it (when engine is warm). The code won't be set though until the engine gets warm (is that a clue?). I drove the car for a little while this morning and no code and then midday it showed up. The weather is slowing changing (a cold front is moving in). Don't know it it affects it or not.

I swapped out the air temperature sensor with the one I originally pull from the engine (coolant) (it appeared in good condition). The air temp sensor was one of the original types (ceramic, rather than metal).

Would pulling the O2 sensor out and running the car tell me whether there is an exhaust restriction ? (see if the scanner shows any difference and if the idle gets better)

I still think solving the idle problem (kind of pants and fluctuating ignition advance) may be the way to go. What would cause this condition ?

OldNuc
01-13-2009, 05:16 PM
If you remove the O2 sensor it will all promptly shift to open loop as the O2 signal will drop to the bias level. If the exhaust is restricted it usualy shows up as a drop in high speed performance. Try it as if it changes the idle then you can suspect the exhaust.

The idle could be nothing more than a slow closing intake valve. It is a characteristic of that engine to do that in some versions. The fix is to feed it a can of techron into a warm engine through the PCV hose while holding the idle at 1500 or so. A slow closing valve will really foul up the idle. The timing shifting back and forth is what you would expect to see with a slow valve. If there is a sticky valve it is being caused by the buildup of crud on the intake manifold side of the valve.

cboss
01-13-2009, 10:23 PM
The P0172 code came back today as I mentioned above.

Tonight, I cleared the code, drove the car about 20 miles at highway speeds and the SES came back on half way on the trip.

2 hours later, I cleared the code again and drove home , 20 miles at highway speeds and no code yet.

The idle does sound a bit better.

I am runnng the car through cleaning cycles using SeaFoam now. I know there are different views about such things, but I have seen some improvement.

I had added 5 oz of seafoam to the oil and have been driving it for a few weeks now.

I changed the oil/filter yesterday and added 5 oz of seafoam again and plan to change the oil/filter on shorter cycles than normal (say 1500 miles, rather than 3000).

Rather than do a piston soak and expect miracles over night, I and trying the slow approach.

I am tracking the mileage (using the trip counter) when I add oil to see how far I get before I lose a certain amount. My oil usage is improving slowly.

I am a little confused about the intermitant nature of the P0172 code. Why I can go a few days with not code and then it comes back and can't be cleared for awhile (comes back immediately).

My analysis continues.

OldNuc
01-13-2009, 11:17 PM
The seeming randomness of the code is what makes it interesting. I would try the techron into the intake to clean up the back of the valves. Somehow I feel its related to the idle variations.

cboss
01-14-2009, 02:20 PM
I cleared the codes yesterday.

I drove the car this morning 20 miles (highway speeds) and no code.
The temperature out is 12 degrees (warmed up later).

Later, I drive the car again and while warming the car up, the SES light comes back on. I drove 20 miles home (highway speed). I also noticed a slight decrease in power coming home (ie. on hills).

The code is not set when the engine is cold or warming up (even when closed loop). The idle also is better when cold or warm.

When the car gets to about 180 degrees or above near running temperature, then the SES light will come on. Once the engine is warm, every time I clear the code (while idling) it comes right back on. If I give it a little gas, while idling, it won't come back on (it can though come on while cruising on the highway).

The fact that sometimes the code won't show for a few days and then it comes back and can't be easily cleared (comes right back) is baffling.

When the SES won't clear easily, I have checked the fuel pressure with a gauge and it is perfect (about 35 psi at idle).

I have new injectors now, a new OEM fuel pressure regulator, a new fuel filter and the pressure from the pump is right on the mark.

This would rule out a fuel problem.

I have a new MAP sensor, new MAP sensor grommet, new air filter, new Air Idle control and I cleaned the throttle body.

This would rule out the air supply.

I have new spark plugs, new wires, and a newer used (from a 2001) ignition module and coils. The engine is getting proper spark.

As far as computer control, I have a new O2 sensor (NTK), a new ECTS, swapped in a newer air temp sensor, new MAP, new TPS so I should be getting good data for the PCM.

I have a new CAT installed.

I have a new Evap Solenoid and I fixed any vacuum leaks.
I have a new PCV valve.

Whats left ?

- restriction in exhaust pipe, resonator or muffler ?
- sticking EGR valve ?
- gas filled vapor canister ?

The intermittant nature of the problem is baffling.

Would that not indicate something mechanical or electronic which has an intermittant problem ?

I need a plan of what to test next!

This is happening for a reason and I just haven't hit on the true cause yet.

When I read the freeze frame when the P0172 code is set, the fuel trim is high (rich) but I am not sure whether it is beyond range. The LTFT is sually -21.8 and the short term is -17 to -21.

cboss
01-14-2009, 02:24 PM
I haven't tried using techron, but I have used SeaFoam sucked up the PCV hose into the intake, to do a cleaning.

OldNuc
01-14-2009, 03:11 PM
Sea Foam is a fuel stabilizer. Techron is an actual cleaner. The techron will do a better job.

cboss
01-14-2009, 11:03 PM
Maybe if I break down the symptoms into possibly separate problems.
While one thing could cause all the symptoms, it is possible there are multiple problems that combine together.

Let's look at the current symptoms separately:

(1) The slightly rough idle (worse at temperature of engine rises). If it is a sticking intake valve, like you suggest, it may explain the idle and the fluctuating ignition advance at idle.

(2) P0172 code, with accompaning decrease in power and some hesitation.
What is intriguing is that this is intermitant.

What would come and go randomly ?

- EGR valve sticking ?
- bad wire connections ?

I would think a restrictive exhaust would not be intermitant, unless a big chunk of say the old CAT is bouncing around the resonator or muffler, and sometimes the restriction is worse than others.

(3) Rich fuel trim.
Even when the code stays away for a few days and the car runs good, the fuel trim is kind of rich IMO.

The pressure is right, new FPR and new injectors. This would seem to only leave something with the PCM making it too rich (ie. bad ECTS)

I was wondering about whether the ECTS can be a problem when it is not an OEM Saturn part ? It has been said it is not, but would it be worth testing the sensor (put it in ice water and then boiling water and testing the OHM's) ?

Another question:

If the EGR valve sticks at times would not this show up on the scanner for the EGR monitor ?

Could an EGR valve stick, yet the monitor still pass ?

cboss
01-15-2009, 12:07 AM
This is a "shot in the dark" so to speak.

Sometimes problems can be the result of the unexpected. The key is examining what factors are related.

Coming back the the engine Load question again.

My engine has an engine load of 29% at idle.
OldNuc, you thought that seemed high.

I took a look again at RichPins video of using a OBD-II scanner and on his SL, at idle his Load was only between 22% and 23%.

For discussions sake, lets consider that his 23% is the proper baseline.
That would mean my engine is 6% different than normal (29%-23%=6%).

Now if the load should be 23% and mine is 6 points different, then:

6/23 = about 25%

My Load is about 25% higher than normal and I would assume that would mean a significant increase in fuel being requested by the PCM.

That definitely would create a rich condition.

Now why ?

Lets' consider than the normal problems have been fixed already (vacuum leaks, temp sensor, etc.).

Is there anything else that could effect the engines "calculation" of Load, that is easily over looked ?

While researching how Load is calculated, I found two factors I haven' considered yet:

(1) air conditioning clutch
(2) power steering pressure sensor

The PCM takes both into consideration when calculating Load, since they add to the load of the engine (when AC is ON and when turning steering wheel at slow speeds).

There may be other "hidden" factors.

The point is:

What is some part of the car, which effects Load calculation but is not obviously considered involved, is defective and it increases the Load calculation so it is false (too high) ?

The PCM would pump in more fuel than is necessary and a rich condition would result. This would also explain why it is so hard to figure out, since it simply is just never considered.

Some Load baselines from other SL2 (96 to 99) would be helpful here.

OldNuc
01-15-2009, 10:50 AM
Check the MAP. At idle it should be 20 to 21 inches of vacuum. The absolute pressure would be the vacuum subtracted from the local uncorrected barometric pressure. That would be 9 to 10 inches Hg. What are you seeing?

cboss
01-15-2009, 03:30 PM
I checked the vacuum.

The air temperature right now is about 18 degrees.
I live about 2500 ft above sea level.

The barometric pressure (MAP w/key on) is 27.7 in.hg

I checked the MAP value when the engine first started (cold) and when it warmed up a bit to about 167 degrees.

Engine Cold:

MAP fluctuated between 8.8 and 9.7

which is a vacuum of 18.0 to 18.9 in/hg


Engine Warm:

MAP fluctuated between 10.6 and 11.9

which is a vacuum of 15.8 to 17.1


Note: The MAP appears to fluctuate at idle, much like the ignition advance does.

OldNuc
01-15-2009, 07:01 PM
Bottom line you lost 2 inches of vacuum. Its -20 here this AM and never touched zero all day. It is supposed to be warmer and I will go see what mine reads. The computer is in the house so it will at least work in the morning. The vacuum seems low, not enough. I still like a sticky valve. Those are classic valve sticking symptoms, not enough vacuum and twitching.

cboss
01-23-2009, 09:55 AM
OldNuc,

While surfing the web doing research I came across an article on testing vacuum and it stated that the specs for engine vacuum are based on tests at "sea level".

The article also said that for every 1000 feet above sea level, you decrease the vacuum at idle by 1 in/hg. I would assume this is because the air gets thinner the higher above sea level you go.

I live at above 2500 feet above sea level (in the mountains. actually the valleys are 2500 ft and the mountains higher) which would mean my vacuum at idle should be 2.5 in/hg less than specs.

Lately the SES light has been coming on every day (and I then clear it for the next day). Yesterday, I drove about 50 miles (most highway, but also around town) and no SES light.

When the SES light comes on and stays on (can't easily clear it for long) the engine runs poorly. When I get a day or so with no SES light, the engine runs quite good.

I do think there is an intermitant nature to the problem.

If I get a chance, I may install a new CPS today and see what difference that makes. I read somewhere the CPS can intermitantly act up when it gets heated up.

I am also shopping for a used EGR valve (on Ebay) and if I can get one cheap enough, I'll swap it out to see if it makes a difference.

OldNuc
01-23-2009, 10:44 AM
If you are measuring vacuum then that is referenced to barometric pressure and that is altitude dependent. Read the MAP with the engine off and compare that to the uncorrected barometric pressure. Should match.

It could be pressure dependent and that would be interesting. If the PCM captures the barometric pressure at start and then calculates the absolute pressure it will be independent of barometric pressure unless the barometric pressure changes, as when a storm front rolls through.

cboss
01-23-2009, 05:44 PM
I installed the new Crank Position Sensor today.

I took the car for a test drive (about 20 miles, most at highway speeds).

The power was much better. I didn't have to give it much gas to keep up to speed.

When I came to a stop sign, the idle was simply perfect (best I have seen yet).

When I got home, the temperature was a little high (slightly above half mark) and when I let it idle, I used my scanner on it and the advance was more stable. As the car got a little hotter, the "woof, woof" in the idle started to come back and the advance was fluctuating more like it was before.

I think I need a new thermostat.

From the scanners live data, I could tell the upstream O2 sensor was cycling better (from high to low and back again). This is good.

I have a question about the downstream O2 sensor.

When idling, should it also cycle (from low value to high and back again) ?

Mine stays about .700 volts. That is a bit rich.

If the upstream O2 sensor cycles properly and the downstream stays rich (above .500 volts) and does not cycle, would this indicate a restricted exhaust ?

The CAT is new, but the resonator and muffler may be stopped up. I am considering changing them out also.

OldNuc
01-23-2009, 06:44 PM
Any exhaust restriction tends to drive up the rear O2 reading. The rear O2 sensor has no control function other than to monitor the CAT. As long as it is not setting the dead CAT code it is fine.

Your temperature is going to cycle between 1/2 and 3/4 if the thermostat is functioning properly.

When the idle starts woofing in the exhaust that will change the timing. The question is why is it doing that.

You could open up the exhaust behind the cat and see if it makes a difference.

The CPS seems to have helped.

cboss
01-24-2009, 03:29 PM
The CPS has helped and improved the engines running.

Sadly, the SES light came back on again today.

I took a look at richpins video of the OBD scanner data for his car and I noticed the downstream O2 sensor stayed at about .680 volts.

Mine stays at about .780 volts and sometimes peaks to .800.

The upstream (front) O2 sensor appears to be cycling properly with high and low values, so this may indicate a restricted exhaust. Since the CAT is new, I would suspect the muffler is clogged. The muffler pipe is very black (carbon) inside and it sputters crabon flakes. If I put my hand near it when the engine is running I get lots of black flakes on my hand.

I got a small crack in the radiator today (and it leaks), so I need to clean it and put some epoxy on it to fix it.

I'll keep updating my progress.

OldNuc
01-24-2009, 03:39 PM
Unfortunately, epoxy will not properly or permanently bond to the radiator tank. Its new radiator time.

The muffler is probably restricted. As a low buck experiment I would cut the inlet pipe and slip on a piece of straight pipe and clamp it to the hanger and see how well it ran that way before spending on a new exhaust for no gain.

cboss
01-24-2009, 09:19 PM
Actually, I have found the putty type epoxy (you just knead it) works great to fix all sorts of things. It has great adhesion. I roughed up the surface with sandpaper first and I cover a much larger area than the hairline crack, so it has something extra to grab to.

I fixed my geo Prizm radiator with it.

If the epoxy doesn't hold, I'll go to step two.

In the past I have repaired a plastic radiator by using a soldering iron. You simply melt the plastic at the crack and let it flow back together. Then you cover the repair with some kind of epoxy to strengthen it.

I got a lot of mileage out of my Geo's radiator after fixing it.

Worse case scenario, if all else fails, I'll but another radiator.


Now back to the P0172 code.

I drove the car this evening to town and the SES light came on again. But as I drove it a little longer, the light went out on its own.

cboss
01-25-2009, 07:20 PM
It appears the new CPS may be making a bigger difference than first thought.

The P0172 code does come back, but when I clear it and drive some more, the idle seems to get better and better and it takes longer for the code to come back. I won't know if this will continue until after a few weeks.

If this continues, then my guess is the CPS may have been at the core of the problem and the car is progressively blowing out the extra carbon in the muffler (old) and the CAT (which was recently installed).

I'll track the downstream O2 sensor to see it shows a lower value. Currently it has been around .780 volts and should be lower than that. If the exhaust is restricted (plugged with carbon) it may push the value higher than it should be.

If I get a chance, I may install a new muffler this week.

OldNuc
01-25-2009, 08:23 PM
Plastic welding can work. I would be concerned about the catastrophic failure that can occur. That crack is originating at one of the cooler attachment points and is subjected to all of the vibration that the line causes. A small steady state leak can be monitored and made up to for a long time.

Blowing the carbon out could be the solution. You could do a steam clean of the system and that might fix it also.

cboss
01-26-2009, 09:06 AM
What do you mean by a "steam clean" of the system ?

As far as the radiator, the crack is currently a very tiny hairline crack, just enough for fluid to leak out slowly. I think the epoxy repair is enough for it right now. The epoxy putty stops the leak and it also strengthens the area (I covered a very large area with the putty). I also made sure the epoxy has some non-flat areas to grab onto, so it gives it more grabbing strength.

I find the putty type epoxies tend to have a better adhesion to surfaces than does two part liquid epoxies. Also there is a trick to getting it to grab better. Also it is important to rough up the surface with sand paper.

When you push the putty into place, it tends to stick to your fingers and wants to pull off. The trick is to take a piece of plastic bag and lay it over the putty when you are pushing on it. The bag will stick to the putty, but your hands won't stick to the bag. You can get better adhesion of the putty when you push down on it and when it hardens you can peel the bag right off. The epoxy putty will bond well to the surface, even plastic.

The epoxy puttys also can take a lot of heat.

OldNuc
01-26-2009, 01:11 PM
Search for steam clean engine in the How-To Library and the Tech forum. It does work but to clean out the exhaust you have to run the process for about 15 to 30 minutes. You suck water into the intake at a controlled rate while holding the RPM at 1500 or so.

The problem is the crack continues to spread under the patch. Sometimes if you get a real good bond it will hold it.

cboss
01-27-2009, 07:21 PM
I did some research on the web and it appears there is a lot of negatives to trying to repair a plastic radiator.

Mine has a 3 1/2 inch hairline crack, which has broken through three of the ribs (which strengthen the plastic), so it is a good possiblity it could burst.

I decided on getting a new one.

While I could get a good deal online, I am wary of the quality I may get, so I chose to buy local.

Autozone wanted about $170.
Advance wanted about $150.

Amazingly the best deal locally is NAPA,
They have one (ordered it so will get it tommorrow) for about $135.

They said it was made by Modine.

I already took the old radiator out (initially was going to patch it) and its a good time to get to the thermostat and replace it.

Once I get this all installed, it will be back to solving the P0172 code.

Oh, I also want to clean the EGR valve a little better. I only scraped the carbon off the pintle before (Haynes manual said don't use any cleaner on it).

I think I'll clean it better this time around and make sure the pintle moves smoothly.

The muffler needs to be swapped out too.

cboss
01-27-2009, 07:27 PM
Just a note:

I wasn't about to jack up the car and remove the plastic cover underneath, just to remove a radiator.

I tend to be good at getting my hands into tight spots, so I angled the radiator forward so I could get my hand behind it and with a box wrench I removed the bolts holding it to the AC's radiator.

I was able to do it, but I think I may look for some stubby (short length) box wrenches for the future for tight spots.

Seemed like a big job at first, but it really was quite simple. Just a few bolts and out she comes.

OldNuc
01-27-2009, 07:39 PM
Thats how I get them out also. Just take the fan off the back and then remove the clips and lean it back towards the engine and unbolt the condenser. Taking the bottom off is a real pain...

cboss
01-28-2009, 12:50 PM
I installed both the new radiator and thermostat today.
Everything went smoothly.

cboss
01-31-2009, 11:47 AM
The problem is starting to get worse. When I clear the P0172 codes, it comes back immediately. Before I could drive a few days before it came back, now it stays.

I read some stuff about EGR valves and some of the symptoms can be associated with a poorly running EGR valve, such as poor idle and over heating (the engine runs a little warmer than it should).

I broke down and bought a new EGR valve and installed it. It helped a little, but the problem is not fixed.

I am now leaning towards the exhaust system.

When I got the car three months ago, there was no CAT on it. The plugs were black (also they were platinum 2's). This means the car had been running rich for a long time and likely a previous owner (I got it from a dealer) removed a damaged CAT and put a straight pipe in.

I had a new CAT installed already and I have fixed a number of things that could have contributed to the engine running rich.

What I haven't changed yet is the exhaust pipe (from CAT going back), resonator and muffler.

If the previous CAT was severely damaged and the insides fell apart, they could have been blown into the rest of the exhaust pipe and muffler.

I took a quick look at the exhaust pipe and muffler and unless you can buy today a complete exhaust pipe, resonator and muffler as one unit, what is on my car may actually be the originals.

The exhaust pipe, resonator and muffler are one complete unit, rather than separate pieces. They are perfectly welded (or however the originals were put together) together as one unit and they are definitely not a muffler shop set. They look as if they were manufactured as one complete unit.

If they are the originals and the previous CAT (probably the original) was destroyed, they could be filled with junk.

I also read that some pipes with double layers of metal, can have the inside metal peel off and jam up the insides, so that would not be visible externally.

The pipe and muffler show age and are rusty, but otherwise look solid. If they are the originals, they must have made them very well.

The P0172 code has been intermittant through all of this, which is why I original suspected sensors or some other parts which could intermittantly have problems.

Now if the muffler is full of junk from the previous CAT, it could be bouncing around (acting like sand or small rocks) inside and some times it opens up and other times clogs up worse.

Autozone has the complete exhaust pipe (from CAT to muffler, including resonator) for about $83 by Maremont. Autozone has a number of Maremont mufflers from $50 to $70.

Advance has a muffler for $25 (universal fit).

Any suggestions on what exhaust pipes/muffler to get ?

Now about the previous discussion about my vacuum being a little low, even my mechanic pointed out the altitude will change the vacuum being we are in the mountains. I have a mechanic who has a shop right next to my home, which I use once in awhile. I normally do most of the work myself, but sometimes I get a mechanic to do some stuff.

The problem is that I don't find most mechanics are really good at diagnostics. They can fix stuff well, but when diagnostics get complex, they tend to be just as stumped as the rest of us.

I don't mind put some money into the car, since I plan on keeping it for the next 5 to 10 years (I keep cars until they die).

Any other suggestions beyond fixing the exhaust ?

Is there any way to tell if the exhaust pipe/muffler are the originals ?
(if they are, they definitely need to be changed)

OldNuc
01-31-2009, 02:45 PM
Vacuum is dependent on the local atmospheric pressure as you are measuring the depression from the barometric pressure. The absolute pressure is measured from zero and is not affected by the barometric pressure. As the MAP sensor is referenced to barometric pressure it is responding to manifold depression. That can be converted to absolute pressure by knowing the value of the barometric pressure and subtracting the depression from it, the result is the absolute pressure. If the PCM grabs the map value when the key is first turned on and uses that value to calculate the absolute pressure then it is no longer responding to local barometric pressure. You reported that when a low pressure front came through you ran excessively rich. The question is this. If you shut the car off and then go to key on, wait and then start it does that allow clearing the code.

Marimont or Walker exhaust is aluminumized steel and will last 5 to 10 years. the original is a low grade stainless steel and is single wall pipe. The muffler could be full of junk. You can buy a stainless steel replacement but its about 450.00.

cboss
02-06-2009, 09:04 AM
Question:

Obviously when an old Saturn burns a lot of oil it adds to the richness of the fuel (oil becomes fuel when burned). This can contribute to the P0172 problem.

When the oil progressively drops (ie. towards the bottom level of the dip stick) does this richness decrease or does it remain the same ?

Does adding oil and topping it off so it is full increase the amount of oil being burned, compared to say when it is lower at the level of the bottom of the dip stick ?

cboss
02-06-2009, 09:15 AM
I need to get better info when diagnosing the engine.

I purchased an ELM322 (for GM/Saturn) based scanner (RS232 based) for use with a computer for about $23 on Ebay.

I have an old laptop with Windows 95 on it which I plan to use.

I am downloading some free OBDII software from different places on the internet, but I am not sure how many of them will work on Windows 95.

I just need something to get started using a laptop/scanner combo.

I plan (hopefully) of developing some of my own software. Communicating with the ELM322 appears to be quite straight forward.

I write complex software which can work on legacy PC's such as Windows 95, as well as the latest Windows (XP, Vista).

I need better real time data.

What you get with a hand scanner is not enough. It only tells you the current value and you can graph the data and compare how different sensors (etc.) relate to each other. I would think obvious patterns would be more discernable when graphing real time live data.

OldNuc
02-06-2009, 09:22 AM
As to oil burning sometimes changing the brand of oil will have a significant impact on oil consumption. Its a shot in the dark. Additives don't cut it though. The free software for the ELM 322 or 327 is marginal at best but the plotting of the data stream is quite informative. And oil lowers the octane rating of the fuel and that starts the path to running rich.

cboss
02-06-2009, 10:56 AM
The free software is just a starting point, so I can experiment with the ELM322 scanner I bought.

I am a computer programmer, so I plan to write my own software. The ELM chips are easy to communicate with and it should not be difficult to write the communication code.

I want to write some software geared towards real diagnostics. To look for patterms and relationships between "live data" values for different sensors.

Already I have noticed one pattern to my current problem.

The P0172 code is never set when the engine is normal running temperature or below. I watch the temp gauge on the dash and normal running temp is slightly below 1/2 of the scale. For awhile the temp will be about 3/8's and then slowly move up to just below 1/2. At this point, the SES light does not come on, but at times I can sense when it will because I see a slight temperature creep. The temp gauge slowly creeps up to 1/2 and slightly above. When this temp creep occurs it is most likely the SES light will come on.

If I could keep the temp between 3/8 and 1/2 the SES light would never come on.

I need better real time data to see what is really happening and the relationship to this data to when the SES light comes on.

By writing my own software I could track the data exactly the way I want to, specific to my Saturn.

Just a little background about my programming skills:

I am a developer of GUI tools for programmers (meaning professional programmers use my tools to develop their software) working with the PowerBasic compiler (on Windows).

My GUI tool can be used to develop very fast and small (in size) applications which will run on Windows 95 to Vista. It can do advanced Graphics, even on Windows 95. The apps are small enough to fit on a floppy disk and they do not require complex install programs (no OCX's to install). You can copy the files and run.

It is perfect for developing advanced graphic orientated applications which will run on a minimal system.

For example the laptop I am using is a Pentium 150 mhz, 1 gig harddrive and 64 meg ram, running Windows 95 (original version or OS1). You don't much less than that.

If I create some OBDII software which will run on such a minimal system, then it would allow users to buy cheap used laptops (ie. on Ebay) to run the software.

OldNuc
02-06-2009, 11:14 AM
This will run on 3.1 so it might be what you are looking for. http://www.obd-2.com/

cboss
02-10-2009, 10:13 AM
Yesterday, I installed a new muffler (CAT had already been replaced) just in case there was some restriction in the exhaust.

The pipe from the CAT to muffler, appears sound (its the original). I banked the resonator and no noise (nothing inside bouncing around).

No improvement with the P0172 problem.

The ECTS wire had a new connector installed at some time by the previous owner, so I cut the wires where they were spliced (crimp connectors were used) and reconnected them by simply twisting and then soldering them and then coated them with plastic coating made for wires. I had already installed a new ECTS.

No improvement with the P0172 problem.

To recap, so far I have:

Installed new:
- ECTS
- spark plugs and wires
- MAP sensor and grommet
- O2 sensor (front)
- Air Intake Control
- TPS
- CPS
- evap solenoid
- air filter
- PCV valve
- CAT and mufller
- Fuel Pressure Regulator
- Fuel Injectors
- Fuel Filter (no embed FPR in it, since FPR is separate on 96 SL2)
- EGR valve

I fixed a vacuum leak (evap solenoid vacuum line and cracked line to Fuel Pressure Regulator).

I removed and cleaned the Throttle Body. Cleaned MAP sensor hole.

I am not sure what is left to fix ?

Any suggestions ?

OnDaGround
02-10-2009, 10:33 AM
I still stand by the motion to burn it.... I think its time you take it to a dealer. You have spent way too much time and money for something that could be so simple.

laser3kw
02-10-2009, 11:49 AM
Pull the valve/cam cover and check for broken springs and bad timing chain / gear teeth / chain slack/ guides. Also when you take it off, look to see if there is heave soot/ carbon on the cover and other parts. That would point to bad exhaust valve stem seals (replace both exhaust and intake if bad).

OldNuc
02-10-2009, 01:11 PM
I would agree that its time to look at the mechanicals. Take off the cam cover and see how it looks.

cboss
02-10-2009, 06:29 PM
I already purchased a new valve cover gasket, since I plan to install a new one to prevent any oil leaks from valve cover.

When I have the valve cover off, I will take some digital photos so I can post them.

What am I looking for when I have it off ?

OldNuc
02-10-2009, 07:28 PM
Something that does not look like the rest of them. Broken spring, scored cam lobe. Loose timing chain. Shine a light down into the timing cover area and see how far the tensioner is extended. Its on the lower end of the rear guide. Lots of gunk buildup.

laser3kw
02-11-2009, 06:47 AM
I got some GOOD pictures of the soot build up and absolutely trashed gears. I'll hunt them down when I get home to night.

cboss
02-11-2009, 03:52 PM
I took off the valve (cam) cover today and installed a new gasket. I then did a engine flush (instead of engine flush I used a can of Sea Foam and ran it a few minutes) and then drained the oil, changed the oil filter to a Purolator Pure One and added new oil (Mobile Clean).

The oil was very dirty (about the 3000 mile range so needed change bad). I have been using small amounts of Sea Foam in the oil (according to directions) to help clean out the engine.

I think the previous valve cover gaskey was not too old, but I am not sure if the bolts were torqued down properly (may have been a little loose). There was some oil leaking from the edge. The new gasket has been torqued down properly to at least 89 inch/pounds.

The CAM's look reasonably clean compared to pictures I have seen of other engines. The timing chain also looks pretty good (no broken teeth, chain looks tight).

Its hards to look down inside to see of all is well there though. I couldn't get a very good shot of it.

Heres the pictures I promised (attached to the post).

cboss
02-11-2009, 03:58 PM
I did notice one thing.

In the picture below, notice the red sealant where the timing chain cover mounts to engine. When the engine was new did the factory use red sealant like this ?

If not, would this imply that the previous owner removed the timing chanin cover and reinstalled it (ie. say to install a new timing chain) ?

cboss
02-11-2009, 04:10 PM
I tried to get a closeup image down inside the timing chain area. It was hard to see anything, but here are the shots. Maybe someone else can make sense of the images to see if there is anything of value in them.

OldNuc
02-11-2009, 04:30 PM
Theres more. Look at the cam journal caps, those are aluminum and collect carbonized oil at the same rate as the rest of the head. They are black yet the head is not. The sealant on the timing cover looks like RTV-106, the original was a dark color, black or grey.

Could you see how far the chain tensioner plunger is extended? The distance from the face of the tensioner body and the back of the shoe that bears against the guide.

What color is the RTV on the pan seal? If you can see it.

You will never be able to get a picture of the tensioner. Its on the lower rear guide all the way down at the end of the guide. with a bright light you can see it but a photo will be impossible. It also looks like the inside of the timing cover looks like the head should look. they should be the same color and have the same amount of crud. also the face of the block.

Wipe the oil off of a timing chain sprocket tooth and root and look at he 2 faces of the teeth on either side of a root and see if one is shiny compared to the other.

cboss
02-11-2009, 05:10 PM
I can't go back and look anymore since I installed the valve cover (and new gasket) already and I don't feel like taking it all off again.

There appears to be no serious damage that I can see or evidence of it (ie. black carbon from a blown exhaust valve).

I do believe the engine may have been worked on before. I don't know if the head has been worked on or not.

Here are some more closeup photos.

OldNuc
02-11-2009, 05:39 PM
I can't go back and look anymore since I installed the valve cover (and new gasket) already and I don't feel like taking it all off again.

There appears to be no serious damage that I can see or evidence of it (ie. black carbon from a blown exhaust valve).

I do believe the engine may have been worked on before. I don't know if the head has been worked on or not.

Here are some more closeup photos.

It looks like either a repaired head or replaced head. The timing chain was probably also replaced at the same time. Depending on the quality of the repair work on the head this could be your problem.

laser3kw
02-11-2009, 06:55 PM
I agree. The difference in the color is weird. Because the cams and the caps are about the same color, I would guess that the head is off another motor.I would guess if this is true, that who ever did it used the original cams and caps in a different head? Just guessing out load here. I would question whether any half way reputable head rebuilder would sell a "refurbished" head with out the original cam caps? I guess anything is possible. I wish you could have done some more investigating. Something is not right.

OldNuc
02-11-2009, 07:07 PM
As the motor is an oil burner and the timing cover area looks to match the cam journal caps and the head does not I suspect a quickie valve job on a replacement head. The problem with that is if the engine has weak rings the improvement in compression from the valve job will finish off the rings and they really eat oil. It is not as true today as it was 4 years ago but the conventional wisdom is to not do a head job on an engine with bad rings. You will have an oil burner. With the idle problems and the P172 issue I wonder about the head.

OldNuc
02-11-2009, 08:09 PM
Hate those typos... It supposed to be 40 years and not 4 years.

cboss
02-11-2009, 09:22 PM
Taking a closer look at the CAMs in some of my photos, it appears the black on the CAM shafts is a manmade coating of some kind done on purpose and not the product of discoloration.

Notice the area I circled in the photo below. The black color stops just before the cam as if it were painted by hand and the side of the cam not touched at all.

Some people may do their own thing when they rebuild a head and coat parts of the cam shaft.

The black color does not look like natural discoloration IMO.

cboss
02-11-2009, 10:05 PM
By comparing my photos with the ones in the following post:

http://www.saturnfans.com/forums/showthread.php?t=113726

IMO, the black on at least some of the parts appear to be painted over black, rather than original discoloration. The parts may have cleaned as best as possible and then certain parts painted black (or coated with something black).

OldNuc
02-12-2009, 12:05 AM
The raw forging of the cam shaft is a very dark brown/black when new. the lobes are ground on the forging and are bright polished. If the engine is fed a diet of cheap low quality oil every non wear part in it will be coated with a very hard black/dark brown varnish. It will not wipe off. That is what is on the shaft and the journal caps and missing on the head. the head is the hottest part of the engine and gets a coat of this varnish soonest.

I will hazard a guess that he timing chain broke in this engine and the timing set and head are replacements. I also would suspect that the parts came from eBay as a replacement timing set for $75.00 and not a bunch more for a "rebuilt head" without cams and the journal caps.

Last year I needed a core head and I bought one with the cams from a guy in Salina Kansas for 50.00 that he thought was a recent rebuild. It was clean except for the cams. I sent it to the rebuild shop for seals and a lookie-loo at its condition. The seals were good but it was in serious need of having the valve job done right. If it had been rebuilt it was done by a shade tree mechanic with no clue. These parts sell on eBay all the time.

cboss
02-12-2009, 09:21 AM
I have no doubt some rebuilding must have gone on here, since it is too clean for its age.

I do think though the engine is quite sound and runs quite well mechanically.

The P0172 problem has an intermittant nature, which tends to make me think that it is not mechanical, since a mechanical problem with the head would not be intermittant.

While the oil buring is likely due to stuck oil rings, one reason I installed a new valve cover gasket is because there was a good bit of oil around the edges of the valve cover, especially on the back side of the engine. I think the valve cover may not have been torqued down well enough or the previous gasket was simply poor. You can lose a good bit of oil from a leaking valve cover, so I need to see if oil usage improves now that I installed a new gasket.

Also, I had been using the cheap supertech oil from walmart up til now and a cheap supertech filter, since I was trying to clean up the engine with seafoam in the oil. The oil was very black and was getting a little thin (possibly the seafoam watering it down).

I installed a new Purolator Pure One (very good) oil filter and changed the oil, using Mobile Clean which is a better oil and has cleaning properties.

I want to see if these two things make a difference in oil usage.

The intermitant nature of the P0172 problem may be hard to solve, but I do have a few areas to yet inspect (I think I need to verify some of the data going to the computer, since an bad electrical connection somewhere could cause problems. Also the load the computer sees may be inflated for some unknown reason and I think that is worth investigating.

As far as the head rebuild, one can only guess about it. For all I know the rebuild is decent or at least acceptable. There are a lot of back yard mechanics who are actually quite good, so I shouldn't "assume" its a bad job, just because of a visual examination and just the colors on the parts. Who knows, some guys may like to paint some stuff. The photo oin my last post, if you look at it carefully, there is a line where the black color stops. The side of the cam lobes is not black and that part is not ground down. There is a lighter silvery color to the sides of the cam lobes.

It's all guessing anyway and not really facts. Unless one actually dismantled the head, I doubt one could know for sure.

I don't plan on rebuilding the engine or head at this point anyway. I just wanted to see if there were any obvious damage in the head or timing chain and neither appears to have any. No broken springs, no pieces of metal floating around, no chipped teeth on the timing chain, no obvious damage.

Since it is very clean inside (and likely some work was done and possible new seals), I doubt the valves are sticking because of old age and grunge (varnish build up).

I do think the leaking valve cover may have played a role, at least in oil usage. If you notice the spark plug wells, three are bright and clean, but one is dark and varnished. That well was leaking oil and it was burning. There was obvious oil leakage on the outside edges of the valve cover, so a good bit of oil may have simply leaked out.

cboss
02-12-2009, 09:27 AM
Another clue to the P0172 problem:

It appears to only kick in, when the engine is a certain temperature. As long as the temp gauge reads slightly below 1/2, it does not occur. When the engine idles, itgets a little warmer and when the temp hits 1/2 or slightly above, the SES light can come on.

Also the SES light does not come on when the engine is under any normal load above idle or cruise. It only comes on, when the engine is either idling or on the highway when no pressure is on the gas peddle (cruising with minimal load).

OldNuc
02-12-2009, 09:44 AM
Idle and closed throttle coasting are specific fuel trim tables. Idle is also a slightly richer fuel mixture.

Contemplate this. The IAT sensor provides and intake air temperature that modifies the mixture strength. It is important that it actually report changing temperature to the engine but it is possible to modify the output electronically to tell the PCM that the air temperature is higher than what it actually is. That will lean out the mixture. Read the dynamic analog signal. Digitize it. Send a manually adjustable precentage to the D to A converter and send that output to the PCM.

cboss
02-12-2009, 03:19 PM
Today is a warm (albeit windy) day.

I started the car up and I noticed something.

Initially while the engine is in open loop when first started, the idle is appears perfect.

When the engine warms just a little and it is going into closed loop, the idle starts to "woof" a little bit and tends to worsen as the car warms up. Now remember the car is drivable and runs great on the highway. Its just that the idle is not what I would like it to be, once it warms up.

I think the engine is mechanically sound, but something goes awry once the computer kicks in during closed loop, possibly getting something wrong (maybe a bad wire to a sensor).

OldNuc
02-12-2009, 03:42 PM
When you switch to closed loop the mixture leans out. Listen at the intake and exhaust and see if you can hear the change in idle tone more distinctly at the intake or the exhaust.

cboss
02-14-2009, 05:48 PM
Interesting day!

Now that I have a new valve cover gasket, I wanted to see if any oil was still leaking from the valve cover. There was a lot of build up around the valve cover, especially on the back of the engine and the right side (drivers side).

I sprayed the entire right side of the engine (in the area around the EGR valve and below and in the area of the coils and below the exhaust manifold) with engine degreaser and then used a small plastic sprayer to spray a minimal amount of water to clean it off.

I cleaned a number of the wires such as the fuel injector wires, the MAP wire and the ECTS wire. I sprayed electronic contact cleaner into the fuel injector connectors and the MAP connector.

I ran the car for about 15 to 20 miles, most at highway speeds from 55 to 65 and some at slower speeds about 35. No SES light!

Amazingly the car ran better than it has for some time. The pickup was great.

When I got home I let the car idle for about 5 minutes. No SES light. The temp gauge got hot enough to kick on the radiator fan (expected when idling and the air was very warm today). No SES light.

I then let it idle a little longer and finally the SES light came on. P0172 again and the idle degraded.

Once the car was off, I inspected the engine to see if there was any oil on the area I cleaned. Amazingly there was a good bit, but it was not coming from the valve cover. I think the engine leaks oil outside the engine and it heats up (air smells like oil) and the gets sucked into the engine through the normal air intake.

I took some photos and I am confused about where the oil is coming from.

Let me explain the photos:

(Photo 1) The area here below the EGR valve was previously very coated with oil and hardened. I cleaned this are very good and it was quite clean. Notice how much oil is in this area now (the wet oil, not the crud I missed when cleaning).

(Photo 2) The oil is either coming from this point or at least pooling there.

(Photo 3) The oil appears to be spraying in a fine mist onto the air intake (plastic intake tube) close to the area near the EGR valve.

(Photo 4) Notice there is no oil leaking from the valve cover gasket.

I should note that I disconnected the hose from the Valve cover to the air intake hose, so I could take a better picture.

OldNuc
02-14-2009, 08:04 PM
Well, thats progress. The oil droplets are because of the air flow. The area behind the EGR that is full of nice fresh oil is indicative of a leaking cam cover. The cam cover can leak profusely along that end and you will not see it at all until it drips off of the oil pan and lower support plate. the gunk that is built up demonstrates this.

If you want to check this theory go get a square of TP and while the engine is running and hot drag a torn edge of the TP along the joint. And then look at it with a bright light. You will see oil.

The cam cover floats between the gasket on the lower face and the elastomer washers on the upper face. The amount of squeeze is controlled by the physical length of the metal washer and spigot. the bolt torque does nothing after the washer hits the head. There are several ways to address this: the official approach is to take it back off wash the gasket in acetone and clean the cover and head mating surface with acetone. then reapply a dab of RTV on the timing cover/head joint and reattach. Be sure to place the corners of the gasket first and tuck the extra in along the length. Do not stretch it all into the front and back. Do not attempt to seal it with a liberal application of RTV, it will just make a mess and it will leak.

Now what you need here is just a very small amount of additional squish on the gasket. Shims under the washer on top run the risk of breaking the cam cover. By not tucking all of the extra gasket into the front and back you will gain thickness and the leak should stop.

Do you have a Harbor Freight close to you? There is another alternative.....

cboss
02-14-2009, 08:40 PM
The gasket is a new Felpro gasket from Advance Auto.

The cover was torqued a bit more than the recommended 89 inch/lbs (I read the torque wrench wrong).

No RTV was used at all.

I simply cleaned the edges with a paper towel, since it was just wet with oil, not stuck to either the head or the cover.

I didn't stretch the gasket at all. I pushed in the corners first and then easily pushed it in the grove.

What do I need to do now to stop the oil leakage ?

The previous gasket must have leaked a good bit since there was a lot of oil on the engine and it was baked on. This could be where a lot of the oil was going.

Can the burning oil which leaks out (or sprays out, like it appears to have done on the air intake hose) get sucked into the air intake and increase the richness of the fuel mixture ?


AutoZones web site says to add some RTV sealant to the grove in the CAM cover, before installing the gasket (no RTV on the head though). Then while the RTV is wet , install the CAM cover and torque down.

Is this the correct way ?

ssicarman
02-14-2009, 09:12 PM
AutoZones web site says to add some RTV sealant to the grove in the CAM cover, before installing the gasket (no RTV on the head though). Then while the RTV is wet , install the CAM cover and torque down.

Is this the correct way ?

Not with the composite valve covers. The gasket is installed into the groove ini the cover. RTV is only supposed to be used on the timing cover to head joints. It is the metal valve covers that get the RTV in the gasket groove and then only to the top of the groove.

OldNuc
02-14-2009, 09:33 PM
RTV can make a huge mess.

The RTV is supposed to be placed on the timing cover to head joint, this you need to do. The bead in the cover grove is better replaced with Hylomar as it is much more pleasant to work with. And as it will get a bit harder will give you more gasket squish.

The gasket will not seal unless it is clean and oil free. Hence the acetone.

The bolts will snap at a bit over 89in-lb, or pull the threads out of the head. I would not go past 85in-lb myself.

Did you use the torque pattern for the cover? http://www.differentracing.com/tech_articles/camcover.html

You can use RTV in the grove but it has to set for a period of time or it is a waste. The reason for asking about Harbor Freight is they sell Hylomar Universal Blue Racing Formula and it works MUCH better. If ther cover, gasket, and head are clean and free of oil the thing will seal. Be sure there is some extra gasket to tuck in on the ends.

And the oil fog is probably the cause of the problem. That and all of the oil soaked wiring connectors. The oil acts just like the dielectric grease, both are insulators. Fix the cam cover and your oil consumption may be cut by better than half.:D

What he says above is true. But you can get a bit more gasket pressure by half filling the grove with Hylomar.

cboss
02-16-2009, 09:30 AM
I have noticed a pattern.

I first assumed that the P0172 code was related directly to temperature. As the engine reached normal running temperature and then slightly above, the code would occur.

Now that I have installed the new valve cover gasket (decreased oil driping on engine and burning) and cleaned part of the engine, the P0172 code is not directly related to reaching running temperature.

The engine can now run a little bit while hot and no code. Time appears to be a factor. The longer the car is run at temperature, sooner or later the code occurs. It will also occur when the car is either idling or at cruise (highway speed, but no pressure on gas pedal).

Another interesting thing is the Freeze Frame values. I assumed the LTFT was always more than -20 % (ie. -21%) before.

The last few days I wrote down any freeze frames which differed from this pattern (more than -20%). Here are two interesting ones:

load ......... 67%
coolant .... 205 degrees
STFT ....... -0.7%
LTFT ....... -13.2%
MAP ........ 26.8 in/hg
RPM ........ 2154 rpm
MPH ........ 50 mph
TPS ........ 33.3%


load ......... 42.3%
coolant .... 201 degrees
STFT ....... + 5.4%
LTFT ....... - 19.5%
MAP ........ 15.9 in/hg
RPM ........ 2287 rpm
MPH ........ 37 mph
TPS ........ 10.5%

OldNuc
02-16-2009, 11:56 AM
Take a look at this and then look at the freeze frame data. It looks like there is critical info missing from the available data. Empirical evidence would point at oil leakage being a major contributor to this problem. I would check the cam cover gasket carefully and degrease the engine. Gunk makes a very good spray on degreaser and it hoses off with water. I also would be investing in a couple of cans of CRC electronic cleaner and clean out all of the plugs and sockets on the engine and ignition. Don't grease any contacts but a bit on the rubber boot will help. remember, ther problem is that they are probably full of oil now and oil is an insulator/dielectric.

Also, check the cam cover hose that connects to the intake tube for signs of oil. That can be a source of oil into the intake also.


The front oxygen sensor sends signals on exhaust gas oxygen content to the PCM to control a 14.7 to 1 air/fuel ratio under normal driving conditions. The PCM can make fuel corrections from a nominal 128 (0% correction) short/long term fuel trim value to a maximum window of 101 to 147 (-21% to 15%). If the PCM determines a rich condition exists (02 voltage above 450 mV), it will decrease injector pulse width (reducing fuel) to maintain a 14.7 to 1 air/fuel ratio. The short term fuel trim value will be between 101 and 128 indicating a rich condition. If the PCM determines a lean condition exists (02 voltage below 450 mV), it will increase injector pulse width (adding fuel) maintaining a correct air/fuel mixture. The short term fuel trim value will be between 128 and 147 indicating a lean condition. When certain conditions have been met to begin learning fuel control (EGT, closed loop, etc.) the PCM will use the short term fuel trim cell to update one of the four long term cells (idle, decel, cruise, accel) being used. During every drive cycle (trip) the PCM will use and update the idle and cruise long term cells and will keep track of how long the control system is operating in these cells. After approximately two minutes of being in each of these cells the fuel control system will enable EVAP purge (EVAP purge On) and the purge long term cells will be used. (Refer to EVAP Canister Purge section of this manual for information on purge long term fuel cells.) If the average of the idle and cruise fuel cells is 147 and the short term fuel trim cell is currently over 128, a lean fuel system DTC will set. This test is performed before the purge cells are used.

DTC PARAMETERS
DTC P0172 will set if the averaged long term idle and cruise fuel trim cells are 101 and the current short term fuel trim cell is below 128 for three seconds when:

* PCM is using the non-purge long term fuel trim cells.
* The system is in closed loop.
* Engine coolant is between 60 and 115C (140 and 239F).
* Air flow is between 2 and 80 grams/second.
* Manifold pressure is between 20-90 kPa (3-13 psi).
* RPM is between 500-4000 rpm.
* TP is less than 75% open.
* Barometric pressure is greater than 70 kPa (10 psi).
* IAT reading is between -30 and 80C (-22F and 176F).
* Vehicle speed is less than 121 km/h (75 mph).
* No ECT MAP, IAT TP Sensor, VSS, EVAP Cannister Purge, Front O2S or P0125 DTCs have been set.

cboss
02-16-2009, 12:52 PM
It looks like there is critical info missing from the available data

What specifically is missing ?

Also, the LTFT didn't appear that bad in the two freeze frames above, so why would the P0172 code occur then ?

OldNuc
02-16-2009, 01:15 PM
The freeze frame data for LTFT is not the average of the idle and cruise that is stored and used to reference the code. There are more points stored, your scanner is not pulling all of the data.

cboss
02-18-2009, 11:17 AM
I removed the valve cover and cleaned the groove, the gasket and the head surface (where cover contacts) thoroughly and reinstalled it.

Putting RTV in the groove doesn't work well. I tried it on part of it where it was most likely to leak and then had to clean most of it out. The gaskets tend to stretch larger, so you have to slowly push the gasket into the groove in a fashion which squishes it back to proper size ( a sort of push, squish or slide it back a little and the push some more gasket and the same).

All looks well with the valve cover now.

I'll track it to see if any oil is spraying out.

I think I should replace the PCV valve grommet (and the PCV hose) just to make sure it isn't a source of a leak.

I had the air intake hose off and I noticed the inside of the throttle plate is very black, but also oily (like wet oily carbon). I had cleaned the throttle body not long ago, but the carbon on it was dry.

Does the oily carbon on the back (not the front though) come from the PCV system or from the EGR system feeding back ?

I checked for vacuum leaks, especially around the intake manifold using propane. I used an unlighted propane torch and sprayed propane gas near the intake manifold and the engine did not speed up. It appears there are no leaks in the intake manifold.

OldNuc
02-18-2009, 02:48 PM
Raw oil inside the intake is from the PCV. Replacing would be a good idea.

Hmm, propane might be more fun than starter fluid.:yes:

If there is on oil around the PCV grommet its good but you can find a functional replacement on the HELP rack, usually. EGR is usually hot enough that the oil is all burned.

Hylomar Universal Blue Racing Formula is a bit harder and will fill the grove and not squish out and stays sticky for days. I am not a fan of RTV. Most of the things its now being used for is outside its original purpose.

laser3kw
02-18-2009, 06:44 PM
you may want to add a catch can to your PVC. Do a search to find the basic configuration.

OldNuc
02-18-2009, 07:17 PM
+1^^^If the proper PCV does not fix it then the blow by is so bad that a catch can is a very good idea, even necessary.

cboss
02-20-2009, 12:27 PM
The P0172 code is more consistant now and I can't clear for long.

First, I figured I do some common sense stuff.

Make sure I am getting a good spark.

I pulled the plugs and you can see their condition in the photos.

Some ash after a few months, but not as black as I would have expected, since it is running rich. These are NGK plugs.

The Plug from cylinder #3 was very bad though, particularly ash deposits.
Also fuel/oil additives can cause this problem as well.

I installed new plugs and switched to Champion 71RC12YC.

I have read about some debate of the effects of anti-seize compound on plugs, so I opted for these plugs. Champion describes them as having some special plating (tinning) on them so they don't require anti-seize compound. I installed them without anti-seize. I figure the better the contact of the plug to the engine (ground) the better the spark.

I was thinking of getting new wires, but opted to wait (more about this later).

Another thing I came across was info about whether to use dielectric grease under the ignition module. I found the truth about this finally.

Dielectric grease should never be used on the ignition module or any computer chip based component that requires a heat sink. The mounting area for the ICM is a specially designed area which acts as a heat sink. Have you ever looked at an electronic board of some kind (ie. motherboard) and seen certain chips/transistors connected to a block of aluminum ? The aluminum acts as a heat sink to decrease the heat in the chip (electronics worse enemy is over heating).

On a car, the ICM has a metal back and that touches the mounting area which is usually metal (aluminum), such as the transmission case on the Saturn. It is also not flat on the Saturn, but has air flow areas. This acts as a heat sink.

If you use dielectric grease under the ICM, it acts as a heat insulator and it is possible to burn out your ICM if it gets too hot.

The proper compund to use is called "heat sink compound" which has special minerals in it which quickly transfer heat (its prime purpose). It can be purchased at any Radio Shack (why some parts stores don't carry it is a wonder). IMO, do not listen to the parts store guy when he says use dielectric grease.

I removed my ICM, cleaned it up and used the proper heat sink compound on it.

The car runs still better and the idle is yet smoother, abeit I still have the P0172 code.

I have a gut feeling the problem lies somewhere with the electronics. When the weather warms up I plan on going through a thorough examination and cleaning of the wiring , contacts and grounds.

For example, I was having a problem for some time with the directional signals. After about 15 minutes the left direction signal on the dash would stop working (not light up). You can hear the flasher working and outside all the lights work correctly. Its just the dash light that stops working.

I was playing around one day under the dash, where the wiring has a big connector box (from the PCM). I unplugged one of the connectors and the reseat it again solidly. Guess what ? The direction light problem is now fixed.

I have also had problems with the tachometer. At first it would simply not work for about 10 or 15 minutes until the engine had run a bit and then it started working properly. I just thought maybe if it was cold out it was just sticking. Now it takes much longer to work and sometimes doesn't work after a 20 minute drive.

I went through most of this entire thread again today, reading back on the stuff I have dne and the results.

One thing I had forgot about was the AIT sensor (basically same thing as ECTS, but on air intake). Since I had installed a new ECTS, I had the old siting around. It was brass one and there was a good chance it wasn't bad. I swapped it out with the one in the air intake, since that one (AIT) was an original OEM, which was not brass. I drove the car this way for awhile.

What I had forgotten, was that at some point I swapped the original OEM AIT back in again (which is what is in the car now).

It may be best for me to simply buy a new one, exactly the same as the one I installed for the ECTS (so they are the brand). I'll likely do that soon.

I also read my posts looking for what it was I installed that for a short time produces an amazingly great idle (for the first test drive). It was the new CPS (crank position sensor) I installed.

In the back of my mind I an now always asking, "will I have problems with an aftermarket part for this item, compared to the OEM ?".

I figured I check some prices for CPS and I found something very strange.

I'll continue this discussion in another post.

cboss
02-20-2009, 01:16 PM
Now to touch on spark plugs wires.

OldNuc mentioned something about cheaper wires (wire wrapped) versus better ones in the $30 range. What interested me most was his comments about RFI interferance. Could a cheaper set of wires contribute to some electronic problems by producing interferance of some kind ?

I don't think my wires are the cause of the P0172 problem, but could they contribute to some of my problems (ie. fluctating advance at idle, possibly poorer spark) ?

I bought a set for about $17 (I think) at autozone, the cheapest they had.

Would getting a better set in the $30 range, with better insulation (from bleeding out interferance) make some difference ?
If so, what brand would be best (and where to buy it) ?

The next concern is the CPS (crankshaft position sensor).

I installed a new one, which was only $15 from AutoZone (Duralast brand).

Getting back to the quality of the parts installed (aftermarket compared to OEM or simply better brand), the question comes up about the CPS.

This is what intrigued me:

Click this link:

http://shop.advanceautoparts.com/ProductList.aspx?parttype=762&ptset=A&searchfor=Crank+Position+Sensor

Advance sells the BWD brand CPS's. Notice that BWD has two models:

CSS11
CSS11P

One sells for about $15, the other for about $33.

I went to BWD's web site they actually have three different sensors, CSS11, CSS11P and CSS11Z. They do not though explain the difference between them.

Why would they sell two sensors with almost the same model number (one letter different), all for the same car and charge $15 for one and $33 for the other ?

What difference is there between the CPS purchased in the $15 range and the CPS purchased on the $30 range ?

I believe GetSaturnParts sells their OEM version for about $28 (close to the $30 range).

Now it would be nice if the manufacturer explains the difference between them, but that doesn't seem available.

My question, is there something inherently bad about getting a cheaper CPS ?

When I installed the new CPS and took it for a test run, I had the best idle ever when I stopped for a stop sign. When I got home though, the idle went back to its roughness as before.

Also the P0172 problem appears to be directly related to the engines temperature. It will not occur unless the engine is up to full running temperature (or slightly above). CPS's are notorious for being heat sensitive (aka. the won't start when hot problem).

IMO, when two parts look identical and they are not overly complex (like a CPS), if one sells for twice the price of the other (by the same manufacturer) there has to be a reason (common sense).

Now true this may be simply a "shot in the dark", but the quality of parts we install in our Saturns (possibly any car for that matter) can become a real issue.

OldNuc
02-20-2009, 04:41 PM
Tune the radio to an in between station point and listen for RF noise. The pop of each pulse. Use both AM and FM. If you think you hear something turn off the car. If eveything is working right you should hear nothing.

Its not usually an insulation failure problem with plug wires but a conductor failure. If a plug fails or a connector comes loose then it will be an insulation failure.

BWD is a division of the same company as Standard Motor Products. SMP also has 3 versions of almost every part, the Blue Streak is the top of the line. You can cross those part numbers over to AirTex part numbers and their website is much more informative, sometimes.

Did you get the PCV changed and has it cut down on the oil in the intake?

cboss
02-21-2009, 12:21 AM
I already installed a new OEM PCV valve!

I wonder what the effect of using Sea Foam in the oil for awhile has on oil passing through the PCV valve, since it thins the oil ?

cboss
02-21-2009, 06:57 AM
Interesting!

I went out this morning to swap out the AIT sensor with the other one (the one I pulled from the engine originally) and to clear any SES codes.

I found one of the plug wires lose on the coil (it slipped off). That wasn't good. I cleaned it up and crimped the connector a little to make sure it stays put.

The real interesting thing though was the SES code was not P0172 again, but I got a P0341 (CAM sensor out of range). I had gotten this once before, but thought it was a fluke after I cleaned the engine a bit (cleaned the oil off).

Remember my comments above about the CPS.

This is a relatively new CPS, installed about a month or so ago.

Its a cheap one from autozone (duralast for about $15).

Why do some CPS sell for around $30 and others around $15 from the same manufacturer ?

Are the $30 ones better quality ?

The wiring may have a problem, but maybe the cheaper priced CPS's are poor quality right out of the box. It is possible the bolt loosened up and the CPS bounces around a bit too.

OldNuc
02-21-2009, 09:11 AM
Almost any oil additive thins the oil and lowers the viscosity and produces volatile vapors.

The low cost components are from a lower cost production source and probably a lower quality insulation.

cboss
02-24-2009, 03:54 PM
I installed a better quality CPS and better spark plug wires today.

The engine has a little more pep when driving.

When the engine idles for awhile and the engine is warm, the P0172 comes back immediately.

At first the engine was not bad at idle, but the loner it runs (and hotter) the idle starts to "woof, woof" (best way I can explain it) and I can almost tell when the SES light is going to come on.

While I am at a loss of where to go from here, my observation is that it is temperature related (partly).

I installed a new Stant thermostat (195 degrees) so it is possible the engine is a little warmer and so the SES light comes on more. The SES light won't come on unless the temperature is at max running temperature (about 1/2 way on gauge).

Coming back to the subject of the vacuum, OldNuc suggested the MAP was reading too high (vacuum low).

I installed a new MAP sensor, fixed any vacuum leaks and tested for leaks on the intake manifold gasket.

I guess two questions are:

How do I verify that the ECTS and AIT are reading the right temperatures ?

How do I very that the MAP reading is correct ?