1984 Ur-quattro 20vt, 336whp 380 torques, 6-speed

[Highlander]

This car is frigging killing me!!!

I HATE loosing and, this thing so far has me stumped.

CO @13% at idle........WTF

Verified so far:

Fuel pressure
Ecm power and grounds
Tried 2 different 02 sensors
Sweep test of TPS
Mechanical timing verified
3 MASS airfllow sensors!!
Coolant temp sensor
Air temp sensor
Closed throttle switch
Vacuum leaks
wastegate diaphram
By pass valve not installed at this time
2 sets of injectors with proper software
Fuel trim readings with vag com
Vacuum line connected to motronic
Maf reading at idle

I just don't get it.........maybe something in the harness.

It seems like a bad MAF to me...........I just have that feeling

We plugged in VEMS with base map.......major difference!!!

Now system too lean 0.5% CO.......... 1500 PPM HC's (lean misfire) ......CO2@ 9%!!!!!!!!!!!!

Mark is sending us a proper file for VEMS............wish us luck.


WHAt AM I MISSING!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!



ARGH!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Craig

[dbldmnd]

Sounds like you now have to dig into the harness and buzz out all the wires. Verify continuity for the signal under test and then verify no continuity to any other wires/signals. Tedious, but that is where I would go next.

[ralleyquattro]

Unless completely necessary I am not opening up this harness.. it would mean pulling it all out and it's original ABY to boot... since the engine ran much much better with VEMS this is the way I'll be going.

[Highlander]

It's strange,

I know VEMS is speed density based system, and doesn't use MAF ........

It just ran so much better with VEMS.............maybe we should check MAF voltage
at the Motronic, and double check Maf wiring?.

Na........let's try VEMS first as much as I hate to, I'm worried we'll be tuning VEMS around
a bigger base engine problem.


Onwards and upwards!!!

Cheers,

Craig

[dbldmnd]

I didn't mean to remove the harness, but to work at the ECU connector and verify continuity/no-continuity at the other end of the wire(s). For each pin in the ECU, verify it isn't shorted to any other pin. For each input to the ECU, verify there is only continuity to that signal and not shorted(connected) to any other ECU pin. That's the type of testing I was thinking of.

Tough problem, but you guys are so close.

[Highlander]

I didn't mean to remove the harness, but to work at the ECU connector and verify continuity/no-continuity at the other end of the wire(s). For each pin in the ECU, verify it isn't shorted to any other pin. For each input to the ECU, verify there is only continuity to that signal and not shorted(connected) to any other ECU pin. That's the type of testing I was thinking of.

Tough problem, but you guys are so close.

That may happen again.........I'm obviously missing something.

I bet it's something stupid at that. :cheers:


Cheers,

Craig

[1wheelpeel]

Sorry to hear this isn't resolved yet. I wish I could offer a decent guess to the problem but I have absolutely zero motronic experience. You said fuel trims were acceptable and the mechanical timing was ok. What sort of timing advance was it running under those conditions? I can't help but think this all stems from a physical change Martin has made to the powertrain and not an OEM component.

Thats one strange problem.

good luck guys, John.

[Highlander]

John ,

I can't honestly remember ignition timing.......I would say @10 degrees BTDC....but I'm not sure.

It was a long, hot day yesterday.


Cheers,

Craig

[ralleyquattro]

Not to mention frustrating :frustrated:

[quattro87]

Doesn't sound fun, but how hot could it be? Hit 104 F here yesterday. :( Hope you guys get it worked out. It's ashame to have that nice of a car not running up to it's potential. Hang in there!!!

[ralleyquattro]

I keep telling Craig, all it's good for is pictures and looking good, she's not getting washed, not getting stipes, not getting door rings, not getting waxed, not getting pampered untill she behaves.
We'll see who wins... otherwise it's timeout for the rest of the summer...

[Katman]

Hmm..

Assuming that you don't have a bad *input* to the ECU, and knowing that a different ECU does the same thiing..

Are the injectors all batch fire, or individual or bank fire?

If they don't all fire together, maybe get you scope on the different injector outputs, and see if they all have the same signal and pulse width?

Did you try pulling the injector connectors off, one at a time to see what happens?

THat's all that comes to mind to this CIS guy at the moment.. :-(

[Quattro Krant]

I have nothing to offer here in the way of suggestions as we all know what I do when I get stumped :slap:

[1wheelpeel]

I keep thinking about this and its driving me nuts. How can an engine run so rich while in adaptive fuel control? I remember Craig saying the O2 was capable of an excellent waveform (off idle I'm sure). So 3 MAF sensors couldn't change it. A false air leak should have caused a lean condition. What gives? The computer has all the correct information it requires in order to tune itself but it can't. Like I said earlier, I have NO motronic experience so I'm not sure what live data is available to Craig. If the critical live data is acceptable, why bother chasing input wiring. We should have seen it on the scanner right? Outputs could be a different matter. You would swear that this is a mechanical fuel supply issue. Something the computer can't calculate or compensate for. I realize VEM's virtually corrected the problem but how? surely the MAF alone couldn't be the reason. What about altitude compensation? Martin said the motronic did throw codes but for what I can't remember. Half the time codes might not be the root of the problem but rather just the side affect. The chicken and the egg sort of deal. ARRRHHHHH Keep us posted guys. Good luck. :(

[Audiurq]

Switch to CIS, 20vt and get that working first?

[glibobbo21]

continuity will yeild nothing if the Vems works, ISV?

[ralleyquattro]

Steve touched on something that I have read before but forgot.

"Do you have any missing going on? I've seen some issues with the HO coils on stock AAN ECUs. I have a feeling that the coil dwell isn't quite right, but I really have no way to tell and no real way to fix at this time. Have you tried the stock AAN coil packs?"

[Highlander]

Steve Eiche (Mr Motronic) said something interesting

Time will tell.

Cheers,

Craig

[Rocket SLC]

Glad to hear the VEMS is in!

[Highlander]

Switch to CIS, 20vt and get that working first?

:slap: :slap: :slap: :slap: :slap: :slap: :slap: :slap: :slap: :slap: :slap:

[ralleyquattro]

Switch to CIS, 20vt and get that working first?

Maybe I can do what Al Solaroli did.. run 20vt with CIS.

[ralleyquattro]

So.. got the car running on VEMS finally after having to buy a used laptop with serial connection, for some reason the two cables I tried would communicate with the ECU but would not allow firmware update. Any reason why these ECUs are still serial and not usb?

Anyhoo, got a nice file from Marc to try out... but she still seams to run rich.
I am sure it's not the software but probably the same problem that prevented Motronic from running well.
Attached is the log of it running in closed loop with isv disconnected (otherwise rpm hunts around 1200).

Can someone tell me if this looks good.

Side note, it's running much much better than on Motronic, at least there is idle and it can be sustained.

I also tried to lean it out in the fuel table, went as low as 60 in the squares, the open loop lambda was happier, but then rpm was at 1500 even without ISV attached.

BTW the file needs the extension changed to .vemslog, currently xls

Oh yeah, I am running the Siemens 60lbs on 4 bar of fuel (stock AAN).

Cheers

[Marc]

Hi Martin,

I looked at your file, the log still looks rich to me, not sure if this was before or after you adjusted your fuel. Note that the revs will increase when you get the lambda closer to 1.0, because the engine is burning more efficiently and thus revs higher. You have ignition based idle control on, so thats going to help somewhat but there are limits to how well it can pull in your idle. Two ways to reduce it: 1) change your ignition timing around idle (try going down to 0 degrees or so). The idle control ignition mechanism will pull that up and down to the set limits (10 degrees) to help adjust your idle. if that still doesn't do it, you may need to dick with the idle screw to change the amount of air coming in through the throttle body.

[ralleyquattro]

Hi Marc,

Thanks for the responce, fuel pressure is stock 4bar.
The ISV was disconnected because it could not keep a steady idle.... is that because ignition based idle was ON? The throttle is closed shut i.e. the little stud is all the way out.

Thank you!

Now I know how Richard felt.

[ralleyquattro]

[rant]

On a related note I can't win for loosing, last night the water pump packed it in on the 90 quattro, since I am going tdi anyway that car is off the road, not going to spend a penny on that one... this morning had to drive the 80 quattro to work without a validation sticker, because of the postal strike I am still waiting for insurance paper, without which I can not get a sticker... to top it off one son has trepthroat and the other one had high feaver all night... I am so deflated!

[/rant]