EGR Simulator

[alpine67]

I'm done trying to deal with the electronic EGR valves on the 97 up S Series cars. Unless you have a factory valve, good luck keeping these things working correctly. Or maybe its something else that causes them to malfunction. Regardless, I'm looking for a work around that can be constructed to disable the EGR valve but still send a correct signal to the PCM so that the service engine light isn't triggered.

To make this work, I need some info that maybe someone might have on this forum. I need to know the voltages the PCM is looking for when it runs the test at startup and after running conditions are set. And if there is a time element involved, i.e. frequency of checking voltages, how many seconds, minutes into a run cycle this happens, and the like. Are there any former Saturn techs here that might have references to this? Would it be in the factory shop manuals (I know, I should check first...).

Any help appreciated. If I can construct something that would work, I'll gladly make the info available to anyone who wants to try it.

[Sid]

I was thinking basically the same thing. My thoughts are to leave the EGR connected except for the pintle position and create a device to simulate the pintle position using as an input the pulse width modulated signal from the PCM (pin A on the EGR). This would allow the EGR to function but tell the PCM that it's working OK even though the pintle may be sticking intermittently.

I don't have much time to work on this unfortunately.

(my 96 has the analog EGR which I believe is the same as the later Saturns)

[Waiex191]

Wouldn't you need to simulate the O2 sensor also?

[alpine67]

Sid, that's exactly what I had in mind. I want to leave it in place, just trick the PCM. EGR's changed in 97 (I believe, can't confirm w/o looking at the manuals) from analog to digital.

Waiex191, that's not difficult and I'm doing that too. a resistor and capacitor wave form filter wired into the leads from the downstream O2 sensor will do it. There are how-to videos on YouTube.

[02 LW300]

Don’t you need to simulate the drop in map when the valve opens also? That may be an OBD1 thing though.

[alpine67]

Originally Posted by 02 LW300

Don’t you need to simulate the drop in map when the valve opens also? That may be an OBD1 thing though.

I'm not sure. I would think that might be compensated for by the ECM. I still haven't dug through the factory shop manual...

[ultravioletnk]

Originally Posted by alpine67

I'm done trying to deal with the electronic EGR valves on the 97 up S Series cars. Unless you have a factory valve, good luck keeping these things working correctly. Or maybe its something else that causes them to malfunction. Regardless, I'm looking for a work around that can be constructed to disable the EGR valve but still send a correct signal to the PCM so that the service engine light isn't triggered.

To make this work, I need some info that maybe someone might have on this forum. I need to know the voltages the PCM is looking for when it runs the test at startup and after running conditions are set. And if there is a time element involved, i.e. frequency of checking voltages, how many seconds, minutes into a run cycle this happens, and the like. Are there any former Saturn techs here that might have references to this? Would it be in the factory shop manuals (I know, I should check first...).

Any help appreciated. If I can construct something that would work, I'll gladly make the info available to anyone who wants to try it.

a number of people have just installed a solid gasket between the egr and its flange, i've also heard of blocking the ports with coins! either way the egr is still plugged in so the pcm doesn't have a complaint. if your issue is that the pcm is complaining no matter what you do, you might have a bad connection (frayed wire or loose pin in the connector). if you suspect the egr itself, personally, i'd rather snag a spare egr or two from the jy than buy the crappy aftermarket replacements.

[Sid]

My service manual (1996) has an "insufficient flow detected" DTC (diagnostic test code) P0401. I've never had it show up. I don't know how people can block off the EGR and not get an SES light with this code.

[ultravioletnk]

Originally Posted by Sid

My service manual (1996) has an "insufficient flow detected" DTC (diagnostic test code) P0401. I've never had it show up. I don't know how people can block off the EGR and not get an SES light with this code.

that particular code seems to be remarkably fickle, but you're right that the code in question would be expected. here is the fsm page on that code.

[ultravioletnk]

Originally Posted by alpine67

I'm looking for a work around that can be constructed to disable the EGR valve but still send a correct signal to the PCM so that the service engine light isn't triggered.

i feel like the best question to ask first would have been, what code are you getting? if it's p0401, see the link in my previous post. it contains the diagnostic information needed to determine the cause of the problem.

[Sid]

Originally Posted by Sid

I was thinking basically the same thing. My thoughts are to leave the EGR connected except for the pintle position and create a device to simulate the pintle position using as an input the pulse width modulated signal from the PCM (pin A on the EGR). This would allow the EGR to function but tell the PCM that it's working OK even though the pintle may be sticking intermittently.

I don't have much time to work on this unfortunately.

(my 96 has the analog EGR which I believe is the same as the later Saturns)

After some more thought on this, I doubt the pintle position can be left disconnected. It's almost certainly used in the EGR solenoid control function as a feedback signal. But, could a device use the input pulse width modulated signal that drives the solenoid, to determine the acceptable range for the pintle position and then clamp it to within that range? Another indicator or indicators could show that there is an EGR failure. If this worked, it would keep the PIA SES light from coming on but there would still be an indication of the failure.

[Sid]

Originally Posted by ultravioletnk

personally, i'd rather snag a spare egr or two from the jy than buy the crappy aftermarket replacements.

Just to confirm your statement, my original EGR was on the car for 22 years. Since that time, I've purchased 3 after market EGR's. Two of them are Delco but I don't know if they're any better quality then other after market EGR's or just rebranded. I recently cleaned one of the after market EGR's with a foam carbon cleaner. It worked for about 45 minutes. Then I got my original down from the attic, cleaned it and it's on the car now. It's worked for a little more than an hour so far...

I can't see any difference between the after market EGR's and the original other than the original is stamped "made in USA".

[alpine67]

Sid, that is my assessment as well. I think it has to stay in the loop. But just like the filters that can be applied to the downstream O2 sensor to make the PCM think its functioning correctly, there is probably a way to build something that would input the correct signals to the PCM. The problem I keep running into is how to measure the voltage at the right time, since the PCM does multiple checks at various times of engine operation. Its far above my pay grade and electronics experience. BTW, the code is indeed P0401.

I've been through the aftermarket replacements, jy replacements, cleaning the original valve and it always returns after some time. I think the valve design is the problem, not the actual operation of the engine - probably too sensitive to changes. Or is it the programming of the PCM?

Lots of questions!!! But thanks for all the input - hopefully we'll figure something out.

[Sid]

alpine67, thanks for the info. My car was producing a p1406 DTC. I got a 1997 PCM and put that in to eliminate the PCM as being the problem. The 97 PCM produces a p1404 DTC. Now I have the original EGR cleaned up and installed. It's worked for about 2 miles so far. It only had about 80k miles on it when I took it out of the car. I'm thinking that the originals are going to be wearing out and it appears the after market replacements are a bust.

For my birthday, I got myself a low end PicoScope and finally got that connected up in my car yesterday. I wanted to watch the PWM signal that the PCM produces to drive the solenoid in the EGR and the pintle feed back signal. Unfortunately I had the pintle position connected to the wrong place. The PWM signal is ok though and it provides some useful information. I'll attach the scope picture.
* the PWM signal frequency is about 130hz. (has a period of about 7ms). So there is about 7ms every cycle of that signal to perform some computation. I think something like a low end Arduino could execute a lot of instructions in this time. It also has adequate I/O. It could probably work.
* The PWM signal never drives the EGR solenoid the full range of 0 to 14 volts. I had a DC voltmeter on this signal a while back and it never went much below 6 volts. (at 14 volts, the EGR is closed. The PWM signal goes lower to open the EGR). So, it might be possible to amplify this signal somewhat and drive the EGR solenoid harder. This might make it able to handle more carbon buildup and work longer.

[alpine67]

Originally Posted by Sid

alpine67, thanks for the info. My car was producing a p1406 DTC. I got a 1997 PCM and put that in to eliminate the PCM as being the problem. The 97 PCM produces a p1404 DTC. Now I have the original EGR cleaned up and installed. It's worked for about 2 miles so far. It only had about 80k miles on it when I took it out of the car. I'm thinking that the originals are going to be wearing out and it appears the after market replacements are a bust.

For my birthday, I got myself a low end PicoScope and finally got that connected up in my car yesterday. I wanted to watch the PWM signal that the PCM produces to drive the solenoid in the EGR and the pintle feed back signal. Unfortunately I had the pintle position connected to the wrong place. The PWM signal is ok though and it provides some useful information. I'll attach the scope picture.
* the PWM signal frequency is about 130hz. (has a period of about 7ms). So there is about 7ms every cycle of that signal to perform some computation. I think something like a low end Arduino could execute a lot of instructions in this time. It also has adequate I/O. It could probably work.
* The PWM signal never drives the EGR solenoid the full range of 0 to 14 volts. I had a DC voltmeter on this signal a while back and it never went much below 6 volts. (at 14 volts, the EGR is closed. The PWM signal goes lower to open the EGR). So, it might be possible to amplify this signal somewhat and drive the EGR solenoid harder. This might make it able to handle more carbon buildup and work longer.

Just had some time to jump back to this. I removed the EGR valve, cleaned it again, and tried to run a small piece of braided cable powered by my cordless drill through the passages in the head and intake manifold without any luck. Put the EGR valve back on and cleared the code. No code for 3 days and around 150 miles of mostly highway driving, but P0401 came back again. This is infuriating!

I don't know enough about circuit design to build anything, but it sounds like what you've discovered could very well be a start!

[fetchitfido]

The AlldataDIY FSM data on the related codes goes a bit more in depth on how the EGR works, I've attached them. I also have the GM ESI FSM for '98-'02 S-Series but those files are larger.

They read like Sid is onto something with the PWM signal not being right. If commanded position over 80%, the voltage must be within 10% (0.5v). Under 80% it must be within 24%(1.1v).

Also of note: "An open in the PCM diode feedback circuit will result in unstable valve operation resulting in vehicle surge"

[Sid]

Thanks for the info fetchitfido. I don't quite understand the P0404. My service manual (1996) also shows a P1406 pintle position error which is what I'm getting now. It only shows the tolerance of 10%. Does your 99 also have a P1406 DTC?

[fetchitfido]

My GM Techline TSI/ESI says P1406 isn't valid for the S-Series for any year and P14xx codes aren't valid until around '98. Forum search's show otherwise, but I don't have the printed FSM's to argue with the electronic versions.

I found a link to a P1406 info page, had to use Wayback to retrieve it: http://web.archive.org/web/200607131...StickyEGR.html

I was able to grab the full P1406 FSM via a '97 K1500 4.3L V6 4wd. Looks clearer on the GM doc but looks better formatted on the AllData. Both are about 30kb over the forums file size limit so I'll try my google drive again.
GM: https://drive.google.com/file/d/17f5...ew?usp=sharing
AllData: https://drive.google.com/file/d/1r1w...ew?usp=sharing

[Sid]

Thanks again fetchitfido. That's interesting information. I was wondering whether other vehicles had problems with the analog EGR's. So there are.

The case study on the P1406 "Sticky EGR Valve" pattern failure is informative too. I've been trying to identify some of the problems we have with the EGR valves.

I've cobbled up a test bench (actually a test card table) to take some measurements from some of the EGR valves in my collection. Here's some data on my original EGR that has been working in my car. I cleaned it up with some spray foam carbon cleaner and a hair dryer to warm it up. The spray foam has a tube attached that gets down into the VGR. It's well used but was working.

I'll include the results of static DC measurements that might be useful to anyone trying to make a simulator. At the moment, I'm leaning more toward an EGR controller who's purpose would be to get the EGR solenoid/pintle positioned to where the PCM says it should be. Eventually I'll start another thread on it if it looks like it would work.

[alpine67]

This is good info!

I'm starting to wonder if there isn't something else going on with my SC2 that is masked by the P0401 code since it goes away on its own periodically but always comes back. I've noticed that on warmer days it will reappear more often. I do have the brass water temp sensor, not the resin one, but wonder if it is borderline failing and not setting a code. May test it with an ohmmeter and boiling water or maybe just replace it to see what happens since they are fairly cheap.