Ive been having problems lately with a surging idle. It first threw a code 44 lean exaust then a 21 tps related. So i tested the tps, and set it at .54 volts. I noticed a difference of about .02 volts between the chassis ground and the ecm ground(top wire on tps plug) so im thinking i have a bad ecm ground. My question is where is the ecm physically grounded, and can i possibly temp fix-rig by grounding the ground wire on the tps to the chassis while still connected to the ecm?

 

I had a similar issue. I tied all the grounds together at the PCM (or ECM, in my case). The one you're most interested in is terminal B6. If you're having that much of a voltage drop with the engine off, I'd guess it could get worse with heat/vibration.

A good ground to the frame is better than a bad ground to the PCM, but it would be better to fix the ground wire from the TPS to the PCM.

 

I have a feeling the ecm isnt grounded properly, i am on my third one and my second set of injectors, also my second distributer and my 3rd or 4th ign module. So does anyone know exactly where the ecm physically grounds? Can i ground it somewhere else?

 

You can ground it anywhere to the frame/firewall which is clean & tight or if you're really **** about it, run the grounds to the battery (-).

 

Actually that is incorrect. the grounds that he is looking for are located at the back of the right & left cylinder heads with a braided ground strap going from the right cylinder head to the firewall.

 

The ECM/PCM grounds to the heads? That would make for a slightly better ground path, but doesn't seem "normal".

 

You mean to say that you didn't know that and yes It is perfectly normal, even the car frame is grounded to them. Just think about where the ground for the battery is located, a small wire to the frame and a larger wire to the engine block.

 

Well...I knew that several sensors on the engine were grounded to the block/heads and that the block was grounded to I believe the firewall and that the battery (-) is grounded to the frame. I understand how the ground path is set up electronically, but I still think it seems strange that GM didn't ground most everything to the firewall and call it done. It would make for a decent enough ground as far as voltage drops are concerned. Maybe the distributor creates too much noise which would be picked up by the firewall? This could explain why they did it as you say.

 

If you will take a look at some of the schematics you will see that almost all the grounds for the various circuits from the ECM are all tied into one grounding point and that is the right and left cylinder head. There are other circuits not tied into the ECM that have grounding points else where in the car and they go to the frame.

 

Huh...why clue as to why GM did it like this? I assume there was some reliability issues with grounding to the firewall and/or chassis. This is certainly more difficult a product to build than if the grounds were at the firewall.

If it's nothing electrical, then it stands to reason that you could ground the ECM/PCM & associated sensors to the frame...

 

Actually it is not just GM that was using this difficult product design as you call it. All automotive manufacturers used it. Here is a schematic design for the grounds going to the cylinder heads. This is from a GM service manual for a 1989 camaro and therefore the wiring will be slightly different from your car which is a 1991 camaro RS, PhLaXuS. However the principle is the same.

 

page 2

 

OK, so i can see the bolt on the back of the pass side head. How do you manage to get to these without tearing the car half apart?
Is it possible to create a new grounding system, maybe run a 4g ground directly to the computer or i am i just thinking bass ackwards. I do have an optima 4 post battery, and ive been struggling with issues that seem to be related to a bad ground for well over a year now(mainly engine control)any suggestions would be appreciated.

 

when you figure it out let me know... I am having the same problems. I'll let you know if i figure mine out ever.

 

I still see no reason why you cannot ground the ECM/PCM directly to the frame or at least back to the battery (-) if you're feeling **** about it. GM may have decided (probably for good reasons) to ground elsewhere, but with all the info I've seen, it should work just the same if tied to a good ground on the frame.

 

Well off hand I would have to say that if you want to run a seperate ground for every item that goes to those two (see attached pictures from 4 & 5 posts up) ground points you could. I would also have to say that since the main ground point from the battery is to the engine and not the frame it makes good sense for GM and other car manufacturers to ground as they did.

 

Well...you are totally correct there and that is the info I've been looking for during this entire thread: "...the main ground point from the battery is to the engine..." Now it makes to me why they chose to ground all the sensors there. It makes for more accurate readings from the sensors over the years when corrosion would screw with a lot of connections.

 

If i am reading this right I can run a wire straight from my ECM to the TPS and if this makes the idle issues stop and the codes don't come up anymore, that means the existing ground is bad somewhere from the TPS to ECM?

 

Yes, you could do that if you wished to do so. However let me point out that in doing so you would eliminate the ground for the MAT and Coolant sensors. They are tied into the ground wire for the TPS.

 

so maybe the ground is just bad from the TPS to whereever it splits? Or it could still be the TPS that is bad.