2004 L300 Limp Mode with lost TCM connection

[squidflakes]

Hello everyone,

I had a Saturn S-series that made it to 291k miles so I was pretty excited when I found an L-series for my kid to take to college.

Its a 2004 L300 with 36,551 miles on it as of this morning, literally driven by a little old lady to church on Sundays. It had been sitting in her driveway for two-and-a-half years, undriven, and while the outside and interior were dirty, the engine looked almost new.

After a new battery it started fine and made the 60 mile trip back to my house on some extremely sketchy tires.

I got those replaced and picked up some new parts with no problem, but on the way back I stopped at gas station car wash to knock the first layer of grime off when I started having some significant issues.

The battery, oil, check engine, and limp mode lights all came on at the same time while I was in the wash. I feared that there had been some water infiltration somewhere and this was causing a short. After the wash cycle was complete I was able to get it started and moved to a parking spot where I disconnected the battery and gave the engine and wiring harness a cursory inspection. I didn't find anything, restarted the car, but was stuck in limp mode for the couple of mile drive home.

I immediately checked the trouble codes and got:
P0507 - Idle RPM high
U0101 - TCM network failure

I know both of these can be caused by insufficient voltage and thought maybe the alternator was bad. I put a voltage load tester on the battery and it came back just above 12.4V and the alternator test showing it was putting out 14.7V, so that seemed to clear the alternator and the voltage regulator.

I cleared the codes and checked again and I'm still getting the U0101 and the car is still in limp mode.

I pulled the TCM and checked the connectors for corrosion and bent pins but everything is clean. I checked the wiring harness that goes to the TCM and I couldn't find any obvious breakage. I don't really have a way of testing the connector itself without pulling the TCM open but I'd like to avoid that if possible.

One other thing I've noticed, and this may be unrelated, when I put the key in the on position there is a high pitched sound coming from the ECM and then there is a clicking like a relay or a solenoid is being cycled. This happens twice if I leave the key at ON, but otherwise doesn't seem to effect anything.

So, my questions are:

Does anyone have experience with the U0101 code and know what I can check next? I don't want to have to throw a new TCM at the car if it doesn't need it.

Is that noise from the ECM normal?


Thank you in advance for any help you can provide!

[02 LW300]

I purchased a non running LW300.
I had to replace the ecm on my low mile LW300. The car would crank but not fire and the dash was going crazy.
I had communication errors only on the bcm. There was no communication with the ecm on initial hookup.

Since your issues were caused by water, I would sweep all computer connections and see if they clear. There are ground studs under the hood on the drivers side. One is behind the shock tower and the other is below the fan relays in front of the batteries.

[fdryer]

Just to be sure, where is your TCM located?

[squidflakes]

Originally Posted by fdryer

Just to be sure, where is your TCM located?

Just behind the firewall on the passenger side, under a weather cover, that is under the large plastic faring that also holds the hatch for the cabin air filter.

[02 LW300]

No rodent evidence chewing on the harness? That happens here all the time.

[squidflakes]

Originally Posted by 02 LW300

No rodent evidence chewing on the harness? That happens here all the time.

None that I can see. I had to clean out a pretty large amount of leaf debris but that was mostly under that cowl behind the firewall air dam. I'll post some pictures if the engine and the harness. I very much could be missing something.

[02 LW300]

Was this a drive through car wash or hood up wand in hand type? If it was a drive through then the only water should have been in the cowl near the transmission harness. Does it have a sunroof? The front drains run down the A pillars and can wet the fuse panels since the drains behind the rockers are probably plugged with leaf debris also.

[fdryer]

One other thing I've noticed, and this may be unrelated, when I put the key in the on position there is a high pitched sound coming from the ECM and then there is a clicking like a relay or a solenoid is being cycled. This happens twice if I leave the key at ON, but otherwise doesn't seem to effect anything.

When leaving ignition ON without starting, longer than 30 seconds, the ecm enters throttle actuator calibration mode. The whining sound is the servo motor driven by the ecm digitally to calibrate throttle opening with programmed parameters for fully closed, initial opening and some other positions before finishing up with resetting throttle opening for cold engine startup, based on engine coolant temperature. The throttle actuator and pedal use dual position sensors, two in each assembly to ensure against a runaway engine driven solely by electronics; press the gas pedal to send signals to the ecm, the ecm sends commands to the throttle servo as the throttle sensors sends feedback signals to the ecm that should match pedal position. After calibration, whining stops in about a minute after calibration begins. The high pitched whine is high speed digital signals driving the throttle servo against a strong built-in throttle return spring if electronics decides to fail.

[squidflakes]

Originally Posted by fdryer

The high pitched whine is high speed digital signals driving the throttle servo against a strong built-in throttle return spring if electronics decides to fail.

I thought it sounded suspiciously like a PCM driven stepper motor or servo, so that's good to know, thank you!

Here are the photos as promised earlier with apologies for the links.

First is the engine itself.
https://photos.app.goo.gl/3SFiqTxpSArWULsU9

Next is the weather seal over the TCM along with the pass-thru for the harness.
https://photos.app.goo.gl/hdXkUCS1eG3T6nUS6

This is the harness that comes from the ECM and goes to the TCM. This is the only place that I would suspect wire breakage, but with the tape adhesive and dirt melted in to the bundle it is hard to check the individual wires.
https://photos.app.goo.gl/FU1KgcajeeMTCkKL7


As to the other questions, it was a drive-thru wash and no sun roof so there shouldn't be any water penetration from the top of the A. When I pulled the weather seal from the TCM everything was dry and I didn't see any evidence of water or excess corrosion on the TCM case or in the connector. I'll check the ECM connectors this weekend. I did check the fuse and relays boxes but nothing suspicious there either.

[fdryer]

Disconnect battery negative first before disconnecting modules (ECM, TCM, BCM, ABS, etc). This ensures a safe power down for erasing memory in bcm and ecm. Factory parameters remain intact and this can help as a hard reset when examining connectors. Look for ground splice packs - a gathering of negative wires from nearby electronics terminated into a small rectangular terminal strip covered by a 2" x 2" gray plastic cover with a 10(?) gauge ring terminal bolted to a chassis stud. At least one should be on the drivers side near the cowling/hood hinge point. Others are scattered everywhere so electronics have two grounds - wired ground to chassis and electronics ground thru cabling/connectors. The two main grounds are battery to chassis and chassis to engine block. Hopefully any hands on work corrects this mystery reduced power mode, geek tech for limp home mode.

According to service manual descriptions, reduced power mode occurs when one of four feedback position sensors between two in the pedal and two in the throttle actuator. Each dual position sensor feeds back to the ecm. If one sensor gives a signal that doesnt fall in the programmed range, the ecm determines this issue as a problem and defaults to reduced power mode and turns on the icon of the engine with a large down arrow. Reduced power mode restricts power to around 15%, just enough to limp off a highway or local road without dying in the middle of traffic. I don't recall reduced power mode from a TCM error. You might want to check throttle and pedal connections.

[squidflakes]

Originally Posted by fdryer

You might want to check throttle and pedal connections.

A broken ground seems like its a pretty likely cause and that's where I'm starting this weekend. I hadn't thought to check the pedal and throttle sensors, so I'll add that to the list.

I'd been meaning to get a better scan tool and this seemed like an excellent time to justify that purchase. Its no Tech2, but it can pull the manufacturer specific codes and can also do a network scan and that brought up some interesting stuff.

I'm now seeing the following codes from the ECM:
U0121 - Lost comms Braking System Control Module
U0101 - Lost comms TCM
U0140 - Lost comms with Body Control Module
P1632 - Theft Deterrent Fuel Disable Signal Received

I'm also getting these from the BCM:
U2106 - Lost comms TCM
U2105 - CAN bus error with ECM
U1016 - Lost comms Powertrain Control Module

All of this makes me think there has to be an loose ground or a shorted connection somewhere in the harness. Though it seems odd that the modules can't connect to one another but they are responding to the scanner.

[squidflakes]

I put the TCM problems on the back burner for a bit and did some other maintenance on the car. Brakes, fluids, plugs, and coil packs.

After some additional testing it looks like all of the grounds are connected correctly and the throttle position sensors are good but I'm still getting a CAN bus failure between all of the modules. Some additional reading started pointing me toward the BCM as the culprit as it seems to act as a hub between the other modules on the bus. In addition to this, the instrument cluster is no longer working so I'm attempting to find a new module and a shop that will actually reprogram it.

On top of all of that, after replacing the plugs and the coil packs the car will no longer start. I assumed that the replacement coils were bad and ordered another set. The new set had the same issues.

I pulled the plugs from the 2-4-6 bank, put them in the new coil pack and wired them to ground. All three had visible spark.

I pulled the plugs from the 1-3-5 bank, did as above but no spark. I tried the coil pack I thought was bad and still no spark.

Is there anything else I can test? Did I get two bad packs in a row or is there a larger issue?

[fdryer]

U0140 - Lost comms with Body Control Module
P1632 - Theft Deterrent Fuel Disable Signal Received

1-Is the check engine indicator on or off?
2-Is the wrench indicator on or off?

I have VCX NANO and a pressure control solenoid error code (Pnnnn) triggered the wrench indicator that comes and goes. The check engine indicator never turns on for the xmission error since the wrench indicator is decoded as a pressure control solenoid issue. I don't know if U codes triggers the CEL or wrench indicator.

I must have missed the previous post with error codes but what stands out may be the P1632 and U0140 codes unless they come and go. I'm not sure if those codes can be erased or battery power removed for a minute to delete them or if they're stored in historical memory (fixed for GM purposes). The bcm basically runs all other electronics not run by the ecm (engine EFI system) or tcm (electronics controlling xmission shifts). The bcm has Passlock security to detect ign switch rotation, three wire Passlock sensor, and communicates to the ecm a go or no-go signal to allow or disable injector operation. Passlock does not affect spark. Disabling injector operation is one method to immobilize a vehicle using existing electronics and programming for virtually every vehicle with EFI systems, remotes with or without keyless operation. The bcm also acts as the gateway for error codes emanating from itself, ecm, bcm, airbag and abs; sending error codes thru the OBD II port. Each module has its own self-test powerup sequence in the time it takes to turn on ignition before starting. The bcm OBD II port is used by GM and some aftermarket scantools to program/alter some embedded firmware. No tuning for anyone interested in super or turbo charging.

If those and other error codes comes and goes, there may be an intermittent issue with the bcm. Unfortunately, only new, blank BCMs can be programmed as they need odometer mileage and VIN programmed. Passlock security is learned in one 10-minute session using the appropriate scantool or three 10-minute sessions without a scantool.

The mystery is that one cylinder bank has spark but the other bank doesn't, suggesting; wiring harness between ecm and ignition coil pack or partial failure of the ecm (never reported here). Having spark, to me, the ecm is partially(?) operating for some unknown reason but more diagnosing is needed. When plugs were removed, did they wreak of raw fuel or not. Several starting attempts without the engine firing up should result in either a flooded engine (spark shorting thru liquid fuel instead of arcing across plug gaps to ignite the air/fuel mixture) or injectors aren't firing. There's also a question of whether fuel and pressure is generated at the fuel rails. Depressing the fuel test valve after turning on ignition should result in a spray of fuel (cover test valve with a rag).

If necessary, I can post wiring for the ignition coil packs. There should be wiring diagrams posted eons ago but the search feature here leaves a lot to be desired.

[squidflakes]

The plugs all smelled of fuel when I pulled them and there was pressure at the fuel rail prior to the test, though I had pulled the fuel pump fuse during the test to prevent any sort of fun pyrotechnics show.

The codes have been pretty consistent. I've erased the codes, disconnected the battery entirely, then reconnected it after a few days and had the same codes return.

If you could post that wiring diagram, I'd really appreciate it. I've got Alldata but their entries for Saturns are terrible.