[FIXED]2004 Audi Quattro 1.8 (AWB) No ground to control side of fuel pump relay

I have no ground from ECM to control side of relay. Key off, relay unplugged and backprobed at ecm i have 3.5 volts. With the relay plugged in I read battery voltage. If I take a test light to ground it it only pulls down to 9.6 volts.
I connect ohm meter to known good ground and checked the wire for short to ground. Ohm meter shows OL. (Or on the Verus ^^^^) I ohm the wire and shows .56 ohms.
Checked powers and grounds at ecm all ok.

I also used a test light from pin 86 to 85 and with key on test light won’t light.
Did the same with known good relay and test light lights.

It’s my cousins car and when she was leaving my house it just died at the stop sign at the corner of my house.

I suspect ECM but I don’t understand how it just all of a sudden went bad.

Thanks for any help!!!!!!

EDIT: engine code is (AMB)

Also, if I jump relay I can get it to communicate because this relay powers up other fuses also.

Only code in ECM is low bat voltage because she killed the battery trying to restart it

does your scan tool have bi directional control to turn the fuel pump relay on?? if it does id backprobe the pin at the pcm with a testlight hooked to battery positive and see if the test light will light while turning the pump on with the scanner... let us know what happens

There is no communication because this relay powers more than just the fuel pump. If I jump the relay I get comm but then what’s the point lol.

Odd thing happened. I could never get her battery to take a full charge. So I put my battery in it out of my truck. (No known problems with my battery) but in her car, just cranking it over almost killed it? Very odd.

didnt see that. my bad. maybe just a bad relay??

sounds like a bad ground cable or positive cable cranking over that slow..

Yes once I found that the battery wasent charging and it almost killed mine I obviously moved from thinking it could be a bad ECM.

Early on I made sure the relay was good.

I’m going to research all the ground locations and start by checking the ones that are more likely to have bad connection.

The ECM ground wires did measure good.

.010 volts with key on

Try a jumper cable from your battery negative to your engine , engine cradle, and to body and see if any make a difference if it runs when your hooked to 1 or the other you know which ground to look at (only use ground side cable no need to hook up positive) as I recall there's a main ground wire on passenger front side of engine cradle that I've had similar issues with and using jumper cable was a easy way to identify that

Sent from my SM-G928P using Tapatalk

Key off, relay unplugged and backprobed at ecm i have 3.5 volts. With the relay plugged in I read battery voltage.

This is a bias voltage used on these systems and is normal to see when the relay is unplugged. You should see this same 3.5 on that relay control pin. Do you? If you do, this confirms circuit integrity.

With the relay plugged in I read battery voltage. If I take a test light to ground it it only pulls down to 9.6 volts

Is this still being measured at the PCM? If so, disregard my first suggestion. Your wiring is fine.
As for the 9.6 v part, your test light bulb has too much resistance. I bet you are using an LED style.

I connect ohm meter to known good ground and checked the wire for short to ground.

No reason to do this test if you are seeing 12v at that PCM connector. Also a short to ground on this wire would keep the relay energized all the time.

No I’m using a regular test light. The same test light I use to energize other relays before.

I did get the car to run with a new battery. But while I was test driving it it just died and wouldn’t restart. When it doesn’t start I have no check engine light at key on.

It seems to be very intermittent. Occasionally when the key is in the on position the check engine will come on and will start.
I’ve checked power and grounds on the ECM in both instances and come up with the same results. As one other person here suggested, I will use a jumper cable to simulate body, engine and chassis ground to see if anything happens.
I would suspect if it was a ground problem I would see that on the grounds on the ECM when the problem is occurring.

I had to end up unplugging the damn coolant fans cause when there is no comm they run high speed as a default I assume from not seeing coolant temp. Annoying as hell and a quick battery drainer lol

Thanks scannerdanner, I pride myself on not being a parts changer. That’s why I’ve been spending so much time on this.

ScannerDanner wrote:

Key off, relay unplugged and backprobed at ecm i have 3.5 volts. With the relay plugged in I read battery voltage.

This is a bias voltage used on these systems and is normal to see when the relay is unplugged. You should see this same 3.5 on that relay control pin. Do you? If you do, this confirms circuit integrity.

With the relay plugged in I read battery voltage. If I take a test light to ground it it only pulls down to 9.6 volts

Is this still being measured at the PCM? If so, disregard my first suggestion. Your wiring is fine.
As for the 9.6 v part, your test light bulb has too much resistance. I bet you are using an LED style.

I connect ohm meter to known good ground and checked the wire for short to ground.

No reason to do this test if you are seeing 12v at that PCM connector. Also a short to ground on this wire would keep the relay energized all the time.

Well this morning it randomly started communicating again and ran for a minute before it died and now it’s back to no comm.

Anyone want to take a look at these and see if I’m missing anything.

Thanks for any help!!

No I’m using a regular test light. The same test light I use to energize other relays before.

If you are absolutely 100% positive it is a ground side switched relay coil, then just ground that circuit with a jumper wire.

I did get the car to run with a new battery. But while I was test driving it it just died and wouldn’t restart. When it doesn’t start I have no check engine light at key on.

It seems to be very intermittent. Occasionally when the key is in the on position the check engine will come on and will start.
I’ve checked power and grounds on the ECM in both instances and come up with the same results. As one other person here suggested, I will use a jumper cable to simulate body, engine and chassis ground to see if anything happens.
I would suspect if it was a ground problem I would see that on the grounds on the ECM when the problem is occurring.

I had to end up unplugging the damn coolant fans cause when there is no comm they run high speed as a default I assume from not seeing coolant temp. Annoying as hell and a quick battery drainer lol

Thanks scannerdanner, I pride myself on not being a parts changer. That’s why I’ve been spending so much time on this.[/quote]

Don't focus on the symptoms, you know where your problem is. That relay is not being turned, you have a wiring problem to that relay or that relay is bad. You've said yourself that when you jump the load side contacts, everything works as designed on the car.
I am more concerned about that relay itself. The control side of a relay is pretty standard as far as current flow goes. They draw about 100ma. Use your ammeter as a jumper wire and give that control wire a ground and give us a current flow measurement. Just be 100% sure it is a ground side switched circuit so you do not fry the ECM driver.

When I wrote that I think when I jumped the relay that also happened to be one of the times the ECM was actually communicating.

I am positive the relay is a pull down type relay. Right now I just jumped it and I get power on the 3 fuses and the fuel pump runs. So the only thing the relay isn’t getting is the ground from ecm But it won’t start because of no comm.

(I added pics for this next part)
Checking the comm wire on pin 60, with key off its at .55V.
Key on and ECM plugged in the comm maxes at 3.99V
Key on ecm unplugged the comm maxes at 4.457V

With the ecm unplugged or plugged in I can communicate with every other module on the car. I painstakingly with through each on that the car has. (It takes forever to connect to each one)

I don’t want to keep coming back to the ecm but I keep getting lead there(or leading myself there)

But what do you think of those comm voltages?

From what it looks like to me there are 2 powers and 2 grounds for ecm

Power all the time on pin 62
Power with key in start or run on pin 3
Grounds on pins 1 and 2

Thanks for the replies Danner. You don’t know how much I appreciate this for my wife’s cousin.

Before we go any further. Are you saying that with the load side of the relay jumped, the car will still not run and you have no com to the ECM?
I thought you said otherwise earlier. Sorry, this is hard to do in this format sometimes.

I know I get confused to sometime in this format. Let me try to explain where I’m at at this point.

The car will randomly communicate, but most of the time it won’t. I know this because the check engine light doesn’t self check. On the very rare time it will start, I will see the check engine light self test.

If I put a jumper wire between batt ground and my t pin on pin 65 at the ecm, the fuel pump relay will turn on and fuel pump will run and 3 fuses that that relay powers will have power on both side.

So that lead me to the ecm not grounding the fuel pump relay, because of the no comm. when the ecm actually communicates everything is fine.

I believe that at the time I wrote the first post saying, if I jump the fuel pump relay the car starts. But I believe this was a coincidence and the ecm just happened to be communicating at that particular time.

Right now, if I ground the relay the pump runs and there is power on the fuses but it won’t start. And the check engine light isn’t coming on at this time.

So that’s when I started looking at the communication lines to see if anything was messing them up. (I posted pics of what I seen on the scope on my last post)
The only module I can’t communicate with on the bus is the ecm. Unless it happens to be one of those rare times the check engine light comes on. Then the ecm will communicate.

I am sorry the post was confusing. I was getting confused I think by the random no comm.

I hope this clears up things a little bit.

Nandor69 wrote: I know I get confused to sometime in this format. Let me try to explain where I’m at at this point.

The car will randomly communicate, but most of the time it won’t. I know this because the check engine light doesn’t self check. On the very rare time it will start, I will see the check engine light self test.

If I put a jumper wire between batt ground and my t pin on pin 65 at the ecm, the fuel pump relay will turn on and fuel pump will run and 3 fuses that that relay powers will have power on both side.

So that lead me to the ecm not grounding the fuel pump relay, because of the no comm. when the ecm actually communicates everything is fine.

I believe that at the time I wrote the first post saying, if I jump the fuel pump relay the car starts. But I believe this was a coincidence and the ecm just happened to be communicating at that particular time.

Right now, if I ground the relay the pump runs and there is power on the fuses but it won’t start. And the check engine light isn’t coming on at this time.

So that’s when I started looking at the communication lines to see if anything was messing them up. (I posted pics of what I seen on the scope on my last post)
The only module I can’t communicate with on the bus is the ecm. Unless it happens to be one of those rare times the check engine light comes on. Then the ecm will communicate.

I am sorry the post was confusing. I was getting confused I think by the random no comm.

I hope this clears up things a little bit.

No problem at all my friend and that does clear thing up. First, have you seen this video?
Second, don't focus anymore on that relay and that it is not being turned on. That is an effect, not the cause.
When your check engine light is not lit and it is not starting or communicating. Do you have 5v available to your sensors under the hood?

ScannerDanner wrote:

Nandor69 wrote: I know I get confused to sometime in this format. Let me try to explain where I’m at at this point.

The car will randomly communicate, but most of the time it won’t. I know this because the check engine light doesn’t self check. On the very rare time it will start, I will see the check engine light self test.

If I put a jumper wire between batt ground and my t pin on pin 65 at the ecm, the fuel pump relay will turn on and fuel pump will run and 3 fuses that that relay powers will have power on both side.

So that lead me to the ecm not grounding the fuel pump relay, because of the no comm. when the ecm actually communicates everything is fine.

I believe that at the time I wrote the first post saying, if I jump the fuel pump relay the car starts. But I believe this was a coincidence and the ecm just happened to be communicating at that particular time.

Right now, if I ground the relay the pump runs and there is power on the fuses but it won’t start. And the check engine light isn’t coming on at this time.

So that’s when I started looking at the communication lines to see if anything was messing them up. (I posted pics of what I seen on the scope on my last post)
The only module I can’t communicate with on the bus is the ecm. Unless it happens to be one of those rare times the check engine light comes on. Then the ecm will communicate.

I am sorry the post was confusing. I was getting confused I think by the random no comm.

I hope this clears up things a little bit.

No problem at all my friend and that does clear thing up. First, have you seen this video?

Second, don't focus anymore on that relay and that it is not being turned on. That is an effect, not the cause.
When your check engine light is not lit and it is not starting or communicating. Do you have 5v available to your sensors under the hood?

Yes sir I have seen that video. And checked the grounds and that bolt. The grounds are good.

I was having a hard time confirming which wire exactly is the 5v ref because it seems none of the sensors share it (externally according to the diagram anyway) but I just pulled the mass airflow sensor connector off and with key on, I have 5v (4.98) on the white/green wire and 0v on the other 3 wires so I’m assuming that is my 5v ref wire.

I think we may be getting somewhere.

Remember even if all of those sensors have separate 5-volt reference wires, that they still share one central regulator inside of the computer maybe two at most.

Also, because of the possibility of more than one 5v reg. You need to verify 5v on another sensor. I would pick the TPS

ScannerDanner wrote: Remember even if all of those sensors have separate 5-volt reference wires, that they still share one central regulator inside of the computer maybe two at most.

Yes sir I know that is the case also from of course your 5v ref videos.

So it seems the color of the wire changes for the 5v ref. On the mass airflow it is the white and green wire that shows 4.98v

I wanted to confirm this so I went to the cam sensor. On the cam sensor I get 4.98v on the white and black wire and battery voltage on the green and grey wire. Both go to ecm.
The white and black wire is shared with the charge air pressure sensor.
I get 4.98v on the white and black wire and battery voltage on the grey and blue wire. Both go to ecm.
So I have determined that the white and black wire AND the green and white wire are my 5v ref wire.

And sense I am getting 5v on the ref and still not communicating I know it isn’t a shorted sensor causing the problem.

Please correct me if I am wrong in this direction.

ScannerDanner wrote: Also, because of the possibility of more than one 5v reg. You need to verify 5v on another sensor. I would pick the TPS

I just did this. And the 5v ref appears to be yet another wire. With key on I get 4.98v on a green and violet wire