Cold start issue after setting base idle (7A)

[RTurner992]

Well done.

Thank you Dave. Let’s hope this solves a lot of other issues! Revved it and no backlash! Trying not to get ahead of myself but once I’ve changed the oil, got it up to temp and given it a run I’ll be able to assess everything better. Thank you for your input!

[RTurner992]

You must be be pretty chuffed after all that research and hunting! I know the feeling

Unusual for a fuse to be totally missing. Blown is more likely.

I would guess it runs better than you’ve ever seen it run with that chain of events…?

I’m over the moon, sceptical, and a little exhausted, haha. As I said to Dave, I’m trying not to get too ahead of myself. I only ran it up for about a minute or two turn turned it off. I really want to change the oil before I go any further.

You know, I’m not surprised at all that the fuse was missing. The amount of horrible, dodgy wiring I discovered and had to rectify was shocking.

That brief minute or two idling seemed great. Again, as I said to Dave, I gave it a rev and the revs dropped nicely and controlled, and the timing belt didn’t slap. It would be incredible if a missing fuse and setting the base idle solved that and the hot start issue. I’ll be sure to update this and the other threads! Thank you for your input as well, really appreciate it.

[John.]

It sounds like you'll now be able to enjoy the car, you deserve to enjoy it !

[newsh]

Wow, a breakthrough, well done!
I've been following this and racking my brains from time to time, I was wondering yesterday if perhaps it was an air leak somewhere but the richness indicates a lack of air rather than too much..
I guess an open ISV is a leak past the throttle butterfly, but it wouldn't show on a pressure test..
Re the codes, is there a fuse in the diagnostic plug 12v supply? Would be strange if that one was missing too!

[RTurner992]

Wow, a breakthrough, well done!
I've been following this and racking my brains from time to time, I was wondering yesterday if perhaps it was an air leak somewhere but the richness indicates a lack of air rather than too much..
I guess an open ISV is a leak past the throttle butterfly, but it wouldn't show on a pressure test..
Re the codes, is there a fuse in the diagnostic plug 12v supply? Would be strange if that one was missing too!

Thanks man. Yes, I was thinking along the lines of an air leak, with everything pointing to it running rich!

Yes, I checked every fuse, and they're all good, and no others are missing either! What tripped me up was that I only have two auxiliary fuse slots (the beige ones at the front of the fuse box. I didn't give those much attention because, at first glance, I assumed they were spare fuse holders, which I've now discovered are the set of 4, further in, next to the relays. So I had S27 populated, but not S28!

I've got some other bulbs on the way today, so I'll knock up another test lead and give it another go. I also discovered that the two ports had previously been removed from the holder and slotted back in the wrong way around! I sorted that a while back, but still couldn't pull codes.

[Lago Blue]

Re: Unable to pull codes and "I need to determine whether the ISV is disabled in the base idle setup mode."


Could the problem with both the above be that the black DTC connector has no power? IIRC it is supplied via fuse #21.

Bridging the two connectors during the settings procedure may not do much if no power was present.

No VAG tool is required for pulling codes, just a paper clip, and an LED test light if you can't get the dash light to function.



It makes sense that the ISV would be made inoperative during base settings procedure; so one is not fighting it.

I like Bernie Benz's tip at 20v.org to set the idle via the screw just below spec, then one can see the ECM and ISV take over.

The page below may contain the confirmation you seek.

Audi Coupe Quattro with 20v Engine Serice Training p34 Screen Shot 2026-04-22 at 6.51.40 AM.png

[RTurner992]

Re: Unable to pull codes and "I need to determine whether the ISV is disabled in the base idle setup mode."


Could the problem with both the above be that the black DTC connector has no power? IIRC it is supplied via fuse #21.

Bridging the two connectors during the settings procedure may not do much if no power was present.

No VAG tool is required for pulling codes, just a paper clip, and an LED test light if you can't get the dash light to function.



It makes sense that the ISV would be made inoperative during base settings procedure; so one is not fighting it.

I like Bernie Benz's tip at 20v.org to set the idle via the screw just below spec, then one can see the ECM and ISV take over.

The page below may contain the confirmation you seek.

Audi Coupe Quattro with 20v Engine Serice Training p34 Screen Shot 2026-04-22 at 6.51.40 AM.png

Thanks for the info. I didn't think to check the service training manual. I still feel like testing again with the CO tester would be beneficial, as it clearly wasn't set before, and I've set it with measuring equipment, but with an inoperable ISV. I understand now that the ISV is set in a 'constant position', but it's still unclear whether that's going to be the same position as an ISV that isn't functional.

By test lead, I meant that previously I'd made a lead using a 2x2 Vag Com adapter, cut the 2x2 part off and soldered the LED and a switch to it, to effectively make a 'tool' that I can plug directly in. The LED does illuminate as well; it's just that I can't physically get it to start blinking. So there is power to the black connector, and you're right, it is powered via fuse 21. I have tried just a jumper wire and LED as a second test to rule out the lead I made, but I still get the same result, unfortunately.

[RTurner992]

Ok, so I've managed to finally pull some codes with my new LED and resistor.

2232 - G70 MAF
2242 - G74 CO Pot
2312 - G62 Coolant temp sensor
4431 - N71 ISV

So I've basically got codes for everything that I've been testing, so I'm going to assume these are all stored codes. I did try to clear these after the 0000 sequence at the end. I ran it again, but all the codes remained. I'm not going to worry about these for now, as the directive from the codes is to check the repair groups in the Bentley manual, which is all the tests I carried out previously, and they all passed. I'll get it up and running, actually drive it, then attempt to clear the codes later and see if anything comes back.

[RTurner992]

Just a quick update. I changed the oil and filter and got it back into base idle setup mode. Fortunately, I counted the revolutions of the CO pot when messing, so that was set where it was when I set it up before. All I had to do was tweak the bypass screw to put the idle at the right spot. The revs drop much more controlled than before, and I'm pretty sure the backlash in the timing belt has gone, although my hot start issue still remains. When I get it to start hot, it misfires and idles badly, and the backlash in the belt is present. It also feels like a different car when driving. It feels a lot healthier, and there's no sudden jerk when the throttle slams shut as you take your foot off the accelerator, as the ISV is now catching it as it's supposed to. There's still some from the rear diff mount, but I'd say about 90% of the shunt you get is gone. No surging idle after driving. From cold, it idles slightly higher to begin with, then settles to where I set it, as opposed to what it was doing before. Correct idle at startup, then higher idle after driving. So all in all, a big win! I've just got to solve this hot start issue now!

Thanks to everyone who's had any input!

[newsh]

Pulling the ECU fuses “should” clear the codes. Or to be sure disconnecting the battery.

[PeterS]

All I had to do was tweak the bypass screw to put the idle at the right spot.

If you end up with a wandering idle, check the oring on that bypass screw...don't ask me how I know after replacing the ISV and lots of troubleshooting.