04 Grand Prix won't start

[GTPsurvivor]

Starter is getting proper voltage and signal from the fuse box because it will crank as much as desired when I jump the crank relay sockets . And I did do the dirty S terminal test before I realized I can test it straight from the fuse box . 

The Pk system is still active . Theft alarm went off when it I didn't turn the door key all the way to the right to open it . But none of the ATS lights come on nor the message on the orange display . 

[Schurkey]

20 hours ago, GTPsurvivor said:

Starter is getting proper voltage and signal from the fuse box because it will crank as much as desired when I jump the crank relay sockets .

I guess you're saying that the relay socket has proper power on the connector for 30, and the output of the relay has conductivity to the "S" terminal on the connector for 87.  The problem is therefore ahead of the crank relay, on the control side of the relay wiring.  Could be on the power side, or on the ground side.  Obtain a service manual for your vehicle, look at the schematic for the starter circuit, trace the wires 'til you find the problem.  Specifically, will a grounded test light illuminate if touched to the connector of the relay socket for 85 when the key is turned to "Crank"?  Will a powered test light illuminate when touched to the connector for 86?

 

Substitute "Solenoid "S" terminal" for "Fuel Pump"; and ignore the "from stock plug" in the following diagram.

 

Edited September 19 by Schurkey

[GTPsurvivor]

That's correct . I don't have a test light but getting 12 volts to 30 and 85 and 12 ground with positive lead to battery,  to the other two with the key on.  

When I jump 30 to 87 the starter cranks until I turn the switch off . Otherwise doesn't do anything .

No spark or injection either way . Whatever system it is doesn't want anything firing up. Either that or it doesn't know any  better . 

 

[Schurkey]

6 hours ago, GTPsurvivor said:

I don't have a test light but getting 12 volts to 30 and 85 and 12 ground with positive lead to battery,  to the other two with the key on.

WHAT are you testing with?  A voltmeter?  WHICH "other two"?  86 and 87?  87A?

When I jump 30 to 87 the starter cranks until I turn the switch off . Otherwise doesn't do anything .

30 to 87 SHOULD make the starter crank until you remove the jumper wire.  It could be that the ignition switch controls power to 30, so that would make sense.  Do you have continuity from 85 to 86 on the relay?  There's a coil of wire inside, I don't know the resistance of that coil, but if it's open, you have a defective relay.

No spark or injection either way . Whatever system it is doesn't want anything firing up. Either that or it doesn't know any  better . 

On the vehicles I'm familiar with, no spark/no injection is a sign of (among other things) a failed crank sensor or sensor harness.  Connect a scan tool, look for RPM when cranking.  Having a REAL service manual would be a great help here.

 

 

[GTPsurvivor]

Yes multimeter testing only . 85 to 86 shows minimal resistance .  30 and 85 are getting 12 positive with key on . I can read RPM on the dash , what should I look for ? 

In other news I tested CANBUS at pins 6&14 and zero resistance , indicating an open connection somewhere but don't know where . If I hadn't lost my breaker bar I would have pulled the crank pulley off by now . But I was hoping electrical tests would pinpoint the problem . 

[Schurkey]

On 9/23/2024 at 3:19 PM, GTPsurvivor said:

85 to 86 shows minimal resistance .  30 and 85 are getting 12 positive with key on . I can read RPM on the dash , what should I look for ?

The vehicle electrical system is trying to activate the relay, since there's power to 85.  If there's ground on 86, and all the relay sockets in the relay box are tight on the relay terminals, the relay should at minimum, click when the key is turned to "Crank".

You know that jumping 30 to 87 causes the starter to crank, so the only untested part of that circuit is the contacts in the relay.  Have you tried another relay?

 

On 9/23/2024 at 3:19 PM, GTPsurvivor said:

In other news I tested CANBUS at pins 6&14 and zero resistance , indicating an open connection somewhere but don't know where .

ZERO resistance means you have continuity, not an open.

Infinite resistance means you have an open.

 

[GTPsurvivor]

I've swapped the relay and tested all of them anyway, verifying minimal resistance in 86-86 and listening to the relay click open when applying 12 volts to 30 and ground to 87. 

I inferred from videos that 0 hi/low canbus resistance was open, around 60 was normal, because there are two parallel resistors of 120ohms, and thus if it was 120 it was a short.  Maybe I had it backwards. But it's supposed to be 60. 

Testing all the ICM connections using wire piercing, while forced cranking if applicable. 

A and B ECM controls I got nothing.

C crank pos signal to ECM. check

D Fuel inj signal to ECM. check

F cam signal check.

G 18x crank signal check, got voltage fluctuation cranking

H crank 3x signal check

J Cam signal check

K, L, M ground wires check

N and P hot wires 12v check

So all of them checked out except A, which shouldn't come on until 400 rpms, and thus didn't. I expected a value from B for ECM bypass of ignition timing control but got nothing. 

 

[Black92GS]

9 minutes ago, GTPsurvivor said:

 

I inferred from videos that 0 hi/low canbus resistance was open, around 60 was normal, because there are two parallel resistors of 120ohms, and thus if it was 120 it was a short.  Maybe I had it backwards. But it's supposed to be 60. 

 

 

The 2 120 ohm resistors in parallel will give you a ~60 ohms reading between the DLC pins if the circuit is fully intact and both terminating resistors are in play.  ~120 ohms between those pins indicates that one of the terminating resistors is no longer present (ie. a break somewhere along the line).  A complete open circuit from a meter perspective will show as infinite resistance (ex. OL display on the meter. True 0 resistance indicates full continuity.  Between the canbus pins of the DLC, a 0 resistance indicates a dead short between the 2 lines. 

[Black92GS]

On 9/6/2024 at 1:19 PM, GTPsurvivor said:

Summary:  CEL and ATS lights are not coming on. 

No ignition even when bypassing the crank relay, which is the only way to crank the starter  . 

Wait...how did I miss this?   The lack of a CEL when the key is turned to run is a good indicator that something is astray.  Check to make sure that the PCM is being powered up when the key is turned to run. Also, check to ensure that all of the ECM/PCM fuses are intact. The lack of the voltage on the bypass wire at the ICM is likely related to this as well, since I believe that voltage comes from the PCM. The ICM can control the spark timing without the ECM...but it would need the bypass voltage present in order to do that. 

Also, I believe that all 3rd Gen Ws use PCM controlled starters (ie. You can release the key from the crank position, but the engine will continue cranking until it starts).  If the PCM is offline, it won't crank with the key either.

Edited September 26 by Black92GS

[GTPsurvivor]

19 hours ago, Black92GS said:

Wait...how did I miss this?

That's why I posted that bit so somebody would pick up on it . Only the ABS and  traction control lights , but those have been on for years . Neither works . 

Is there an easy  way to test if ECM or PCM are offline ? Then I could go further .  They don't get power from the. ICM that's all I know so far . 

 

[Black92GS]

31 minutes ago, GTPsurvivor said:

That's why I posted that bit so somebody would pick up on it . Only the ABS and  traction control lights , but those have been on for years . Neither works . 

Is there an easy  way to test if ECM or PCM are offline ? Then I could go further .  They don't get power from the. ICM that's all I know so far . 

 

Easiest is to see if a scanner can talk to it.  The ICM should be able to generate spark without PCM involvement...but it likely requires the bypass voltage to be present to be able to, which is generated by the PCM. 

Give all of the PCM fuses and such a check while you're at it as well.  Something as simple as a blown PCM fuse could cause all of these issues. 

[GTPsurvivor]

Both PCM fuses are fine as are all other fuses.  DLC does connect to scanner but no codes except since battery has been disconnected about 13 times since it last was running.  It threw a map sensor code while artificially cranking, but that's not going to prevent ignition, just cause it to run shittily. DLC hi/low cam is 0 ohm resistance. Yeah the lack of bypass voltage was the most interesting test on the ICM connectors, because the rest of the wires tested normally. 

[Black92GS]

9 hours ago, GTPsurvivor said:

Both PCM fuses are fine as are all other fuses.  DLC does connect to scanner but no codes except since battery has been disconnected about 13 times since it last was running.  It threw a map sensor code while artificially cranking, but that's not going to prevent ignition, just cause it to run shittily. DLC hi/low cam is 0 ohm resistance. Yeah the lack of bypass voltage was the most interesting test on the ICM connectors, because the rest of the wires tested normally. 

Being able to communicate with the PCM is good. The CAN resistance is suspect and indicative of there being a short, assuming the testing is being done correctly. I believe it’s supposed to be done with the battery disconnected. I can’t recall exactly how those vehicles work, but I believe the ignition switch position is read by the BCM, and then the appropriate command is sent over to the PCM. I’m not sure how the PCM and BCM communicate with each other though. The lack of a CEL could simply be a a bad bulb in the cluster…or it could be something more.  
 

I may have also had the bypass voltage thing backwards. There is a 5 volt reference signal coming from the PCM, but it is being pulled to ground by the ICM until the engine exceeds 400 RPM. The bypass no longer being pulled to ground is the signal to the PCM that it can now control the spark timing. The spark timing line should have a signal on it only when the PCM is controlling spark. If you tested with the ICM connected, that may explain why there was no voltage on the bypass wire. It also means you should have spark even without the presence of the PCM…assuming the ICM, Coils and its related sensors are functional. 
 

[GTPsurvivor]

The CEL was coming on regular because O2 sensors have been offline for a while . Then stopped coming on after battery disconnected because no DTC set . I may need to test CANBUS with battery offline . 

As I understand it, PCM runs on can Hi and BCM on can low so they may get their orders separately . 

This is going on over a month now , and I'm not much closer . All I've proven is that everything works except that it won't start due to some electrical problem , which I already supposed . 

[Schurkey]

On 9/23/2024 at 3:19 PM, GTPsurvivor said:

Yes multimeter testing only . 85 to 86 shows minimal resistance .  30 and 85 are getting 12 positive with key on .

Does the computer control the ground side (86) when the key is turned to "Crank"?  I've made the assumption that it controls the positive side (85).

 

[Black92GS]

On 9/28/2024 at 5:43 PM, GTPsurvivor said:

The CEL was coming on regular because O2 sensors have been offline for a while . Then stopped coming on after battery disconnected because no DTC set . I may need to test CANBUS with battery offline . 

As I understand it, PCM runs on can Hi and BCM on can low so they may get their orders separately . 

This is going on over a month now , and I'm not much closer . All I've proven is that everything works except that it won't start due to some electrical problem , which I already supposed . 

The CEL should always come on “Key On Engine Off”. Thats the bulb test, and the lack of a CEL at this stage can indicate the PCM being offline, corrupt, or possibly a major communications issue. 
 

If the CEL isn’t coming on KOEO….that issue needs to be  addressed first before anything else. 

[GTPsurvivor]

Alright so I reviewed some information and did the canbus test again.  Both sockets receive voltage which increases when  key is on.  Can hi-lo resistance continues to be  zero, meaning there's a short.  I read that 120ohm constitutes an open connection on one of them since one of the 120ohm resistors is offline. Both of them in parallel should  get about  60. I get 0, using same connectors as for the voltage test.

The CEL never comes on anymore, coincidentally with the start of this issue.  The ABS and traction control stay on, oil light comes on temporarily then disappears. 

All this would seem to  point to a short going to the ECM, which would be on hi canbus, but not sure how to test it.  I may have to decode a wiring diagram, if anyone has a good one that may be easy to understand for a novice. 

 

[GTPsurvivor]

On 10/1/2024 at 3:13 PM, Schurkey said:

 

I think that's right because 30 and 85 are getting +12.  So it appears to be getting proper voltage and the relay is properly working, and yet the starter doesn't engage without bypassing the relay from the 30-87 socket, and that only happens with key ON. Something else is telling the relay to not open, because if it did then at least the starter would engage. 

[Black92GS]

On 10/4/2024 at 3:02 PM, GTPsurvivor said:

Alright so I reviewed some information and did the canbus test again.  Both sockets receive voltage which increases when  key is on.  Can hi-lo resistance continues to be  zero, meaning there's a short.  I read that 120ohm constitutes an open connection on one of them since one of the 120ohm resistors is offline. Both of them in parallel should  get about  60. I get 0, using same connectors as for the voltage test.

The CEL never comes on anymore, coincidentally with the start of this issue.  The ABS and traction control stay on, oil light comes on temporarily then disappears. 

All this would seem to  point to a short going to the ECM, which would be on hi canbus, but not sure how to test it.  I may have to decode a wiring diagram, if anyone has a good one that may be easy to understand for a novice. 

 

Now we’re getting somewhere! One module shorting the can lines can take down the entire network. It’s also not a coincidence that the CEL stopped illuminating when this issue started. Whatever is causing the CEL to not illuminate is the cause of your no crank. 
 

Most GMs have 1 or more splice packs somewhere under the driver side dash, where the can lines from each module terminate and are linked with a comb. Unplugging the comb allows you to test the can lines to each individual module.  A diagram would be needed to know exactly what goes where. 
 

Failing that….you can also try disconnecting modules until communication is restored. Not sure why… but I’m suspecting the EBCM (ABS) to be the culprit. Unplug that one first.