forum.audiv8.com Foren English Audi V8 Forum Restoration Project

Hey everyone. I bought a 1991 Audi V8 5-speed 3.6L about a month ago. I've been working on it trying to make it run. It hasn't run for 5 years and has been worked on by tons of people trying to fix it in that time. With this website and the wiring diagrams I have made a lot of headway but I need a little help still.

Initially the car would not trigger the ECU. I found the power to the ECU from fuse 29 to be broken somewhere. By jumping it, I am able to get the ECU alive and read codes off of it. I also had a dead fuel pump which I am replacing with a Walbro GSS341. The fuel pump relay does work and has power but is not triggered by the ECU. The timing sensors are both okay and I do have spark. The engine runs on starting fluid. It is not triggering the injectors though. The injectors DO work, tested by jumping them. Also bypassed the fuel pump relay to supply power to the injectors and still nothing.

I only have one code 2121 for the Idle Switch. When I try to do the output test to trigger the injectors, giving full throttle does not activate the test even though the correct code for injector one is blinking on the CEL.

Can the idle switch stop the engine from starting? And what can I do to fix it?

Thank you everyone for the help you've already provided with this forum. Hopefully I can get this cool and rare car back on the road.

One of my V8 projects baffled me for a while.

I managed to get the car running initially, then it simply wouldn't start again.
The final result was the ECU injection control, and which I think was damaged by either water ingress due to a bad windscreen replacement, or a complete meltdown of an Alarm immobiliser which caused a lot of damage.

Whether /if the idle switch directly stops the engine from starting, I can't remember. But the output tests ran fine for me, but the injectors simply weren't being triggered.

Thank you for the reply,

With the idle switch not functioning... aka the WOT switch also not functioning, I can initiate the output test where it is ready to test and blinks the code for injector 1, but I can't trigger it to perform the test. I could probably trick it into doing the test with a jump lead or something right?

Are there any known faults that keep just the injectors from firing besides roasted drivers? It is using a used replacement ECU.

Jumped the WOT switch for the output test and the ECU was able to fire all eight injectors and trigger the idle witch and charcoal canister.

I did find discontinuity between the ECU and the fuel pump relay where it is meant to trigger the relay.

What for god's sake is WOT?

Idle switch does sometimes do a lot to fail the start, NO signal to the injectors seems to me a bit to nasty - but not inpossible.

Did I understand you right? you used a other ECU and faild further? or do you mean you would try a other ECU?

At the moment I would say, if any responsible sensor fails there shouldn't be fire nor a signal to the injectors. Maybe I'm wrong..... I think not

Oh... do you check the second fuel-relay?

Hello, haha,

WOT is Wide Open Throttle. According to the sjmautotechnik instructions, full throttle triggers each injector test.

I was able to test the injectors successfully by unplugging the Idle Switch and jumping the correct pin to ground. The ECU is able to fire all eight injectors.

What I meant was, the previous owner replaced the ECU once already as an attempt to solve this issue.

Is there a good way to test whether the injectors are firing during cranking or not? Could this be a problem with the fuel pressure regulator? I know for a fact that fuel is getting through the rail and out the return pipe but I can't say how much pressure it is under.

What is this second fuel pump relay? It is not on the wiring diagrams I have been using.

Thanks for the reply!!!!

Problem solved. No electricity to the fuel injectors due to fuel pump not firing. Not sure why the ECU doesn't trigger the fuel pump so I rigged a toggle switch to do it. Runs good now but idles at 2000 rpm. Probably has to do with the idle switch and it needing to relearn how to run properly.

On to the brakes. The line between the T connection in the rear and the whatever unit is in line to the right rear wheel is rusted through and leaking. Replacing that and bleeding. Clutch works and transmission goes in gear and moves.

In Antwort auf:

---

On to the brakes. The line between the T connection in the rear and the whatever unit is in line to the right rear wheel is rusted through and leaking

---

The "whatever unit" is the break force limiter for the rear RH calliper

All brake lines have been replaced and are working. I have a strange problem with the engine running though.

It only seems to run on 5 cylinders. When I first start it up, it runs on all eight but after a minute the driver bank of cylinders starts to die one by one. First cylinder 8 then 7 then 6 and occasionally 5 will also stop working. If I pull the plug wire for any of these while it is running, there is no change.

I have tested and found there to be spark on all cylinders. Using a dummy spark plug.

When I listen to the injectors with a stethoscope while the engine runs, all eight are working. It seems almost like the driver side fuel rail is not pressurized enough.

Any thoughts on this? Fuel Pressure Regulator? The original fuel pump was dead so it now has a Walbro GSS341 which should be able to make plenty of pressure.

What have you done to make it run after a month?

I replaced the fuel pump and installed a switch to activate the fuel pump relay without interrupting the ECU's control of the relay. This was for testing purposes but the system works how it is supposed to still.

If the other Bank is working fine I would try to exchange parts between both banks. Starting with the Ignition system ( Distributor with arm, cables, Spark Plugs (even if there is a spark, it does not necessary mean that the spark is strong enough)), then the injectors. At some point the fault should travel to the other side.

My guess is the distributor arm or the distributor cap. I have seen the craziest faults happening with these parts. sometimes they just stop working properly when they are warm.

I would rule out the fuel system, because as far as I know there is no difference between the two banks. Same regulator and so on. And that the Ignition valves open up to short just on one side... I don´t think so. But, nothing is impossible.

You're right about the ignition system. With distributors, there is never certainty. The one thing I would note is that the distributors do not provide spark to just the bank they are located on. Some of the wires cross over... With my issue that means one of the malfunctioning cylinders gets spark from the driver side distributor and the other two get it from the passenger side.

I would also agree that if the fuel system is working on one side, then it should work on the other side just the same.

Thanks for the input!

I think you mean injection valves alias injectors, not ignition valves

At the rest: fully agreement.

Of course, I must have been sleeping

I was not aware anymore that the ignition cables cross over. This makes a fault in the ignition system a lot less likely.

Then I would really have a look at the fuel system. Are these the cylinders the farthest away from the entry of the fuel line?
I have heard from ( very powerfull) engines breaking because the cylinders on the end of the rail where leaning because they did not get enough fuel anymore.

Maybe also something is wrong inside the fuel system that prevents the fuel from flowing easy ( can be something as easy as a swollen rubber line). The pump itself should have more than enough pressure to run the engine in idle.
Thinking about this, these are all faults that would appear under full power more than under idle....

If you restart the engine instantly after the cylinders where gone, what happens then? And, what happens if you rev the car up?
Also check the electrical connection to the injector valves. Maybe you have some sort of verdigris or so that prevents the valves from opening the whole time. Also check the side towards the ECU. As soon as the engine gets warmer, the opening time gets shortened, and then they just do not let enough fuel pass for an ignition.

If I try to rev the engine up, I have to modulate the throttle to let it climb in RPMs. If I give it full throttle, the engine will die completely. It revs much easier when cold. The injector valves are paired from left to right according to the wiring diagram and confirmed by electrical tests with the multimeter. I would assume then that the opening time is adequete since it seems to be enough for the other side to fire. I have cleaned the contacts on the valves.

The entry fuel line, pressure regulator, and return line are all located on the passenger side fue rail, the functioning side. There are two braided lines which then cross to the driver side rail. I haven't really been able to determine a clog anywhere on those lines.

Once I have an extra set of hands, I can do some more tests. Doesn't help that everything is covered in snow now.

Camshaft timing is okay?

I would check the throttle valve and the respective sensor at the front below the idle valve, especially the
connector.

Cylinder 5 is is firing, so Cam timing on that bank must be correct. No reason to think it wouldn't be anyway.

The Idle Valve functions as supposed to when triggered during the output test. Could that sensor really cause this? I have confirmed continuity with a multimeter between those pins on the plug and the pins on the ECU. I do still get an occasional error code for that sensor although I haven't been able to read them recently.

Below the idle valve, check the connector of the throttle valve sensor (Drosselklappenpotentiometer)

You think that the ECU does not know how much Fuel it is supposed to inject.

I know of the weirdest problems resulting from a corroded connector on that sensor, from idle problems to an engine shutting down to Rodeo rides when triggering the kickdown to transmission emergency mode - it can be the cause for so many problems...

It don't read the rest of this thread... but YOUR description look to me similar like my "Luftmassenmesserproblem"


Maybe sometime I'm in mood to tell about...

I have ruled out an ECU communication problem because:
The right bank of cylinders fires correctly and each injector on that side is paired electrically to a cylinder on the left bank. Therefor all injector valves should be opening for the correct amount of time.

I have also confirmed that all eight valves are opening by using a mechanics stethoscope to listen to the actuation.

In Antwort auf:

---

and each injector on that side is paired electrically to a cylinder on the left bank.

---

Who did tell this lie to you?

Have a look at the attachement..

and this one...

2

BUT this one means that what you mean.

This is a bit stupid... the valves open and close not at the same time, why should the injektors do it?

The wiring diagrams and my digital multimeter. That's how it works.

That is the wiring diagram I have been working with. 1991 Model.

Many early injection systems run a batch fire or semi sequential fire. The injectors are paired according to the firing order to fuel consecutively firing cylinders. It does work and at high RPM it makes no difference whatsoever since the injectors are almost always open.

I've done a few more tests.

With the engine running at idle, I disconnected the spark plug connector of each cylinder not firing, put a screw driver in the end and saw it make solid arcs to the valve cover. Plugging it back in and still no firing.

I swapped around the injector connector from cylinder 5 which is working to cylinder 6 was not working and this made no difference. I already swapped injectors around and the problem does not follow.

I connected a fuel pressure gauge to the primary fuel inlet from the pump and read steady 58 psi with the regulator vacuum line disconnected and engine running.

My next test is to connect the fuel pressure gauge to just the driver side fuel rail and see if there is a difference.

I'd gauge the compression ratio of all of the eight pots.