choinga

So, after completely gutting everything in my car that had to do with idle (because I was rich after my Vitesse upgrade...John re-flashing my chip proved to be the ultimate cure, thankfully) and everything's been working beautifully for several months...a few days ago it started idle surging when starting cold.

I'll try to post a short video of it later, but it is what it is...start the car and the idle just oscillates from about 900 to 500...very synchopated. Once I drive the car a couple hundred yards...it doesn't even have to be warmed up...it goes away?

Any clues on what this possibly could be?

In the last 3 months I've got all new vacuum lines (venturi delete), brand new TPS and ISV and new plugs & wires.

John_AZ

Check the BIG air intate boot clamps and other BIG clamps. It sounds like a random vacuum leak. Clean mating surfaces.

choinga

I've got the Lindsey hard pipe kit that comes with the very nice ratchet down clamps and the silicon sleeves...just double checked them and they look good and are tight as can be.

The thing that is so bizarre is how quickly it goes away...well before the car has had time to even warm itself up...literally, by the time I pull out my driveway and get to the end of the street at a stop sign that's not 200 yards away...it's not doing it anymore...

MM951

Ah, so reflashing the chipboard was what it took to cure your rich problem? Was it a fluke in the tune or something?

I'm slightly rich at idle/low load and have gone through everything to figure it out....

John_AZ

Poor starting/erratic +++++ behavior.
Clean all the electrical connections to the ISV/ICV, AFM, reference sensors---anything with a plug on the engine electrical system. Use a quality electronic spray cleaner. Dry with compressed air. Check cables for damage.

Condensed/edited from Technical Service Bulletin 8706--924S/944/944S & 944 Turbo

John_AZ

Did you replace the O2 sensor? If you did, did you use the pinch wire connectors? (no solder).

choinga

MM951...yes, I have the big injectors (83#) and apparently John tweaked the idle tuning...I would run in the high 12's at idle when the car was warm...now it's nice and stable at 14.7.

I replaced the O2 sensor with the stock one (not universal so no splitting) when I installed my wideband about 5 months ago. So, I doubt that's the problem. Could I just unplug it before I start when it's cold and see if does that?

I'll try cleaning some of the electrical stuff - but damn, this is weird how fast it fixes itself...

choinga

I put a short video of what my car is doing up - It's dark in my garage so you can't see the cluster as well as I thought you would be able to - but you get the gist. You can hear how at the end even after rev'ing it a bit it starts to subside and as I mentioned, if I just drive it a couple hundred yards, it's completely gone at idle. HELP!

http://www.youtube.com/watch?v=2WcN4jg9-N0

John_AZ

I listened to the vid.
No solid suggestions.

Have you checked the fuel pressure at the rail? Fuel filter clogged or fuel pump check valve sticking?
Fuel damper or regulator diaphram cracked? Do you see fuel drops when you pull the vacuum lines on either?

GL
John_AZ

choinga

Thanks John...I sure hope not...both the fuel pump and filter are brand new...as in like last month. I had a funky check valve and replaced all of it while I was in there. My damper is brand new also. Fuel pressure is pretty constant at around 44PSI at idle so I don't think fuel delivery is an issue.

UPDATE: So, I went out and started cold again...pulling the vacuum line of the FPR evens it out immediately...it will surge occasionally, but that immediately stops the oscillating surge. After the car gets warmed up, it seems like the AFR is 'hunting' more than usual...takes a bit longer to creep up to 14.7 at idle than it has in the past. I guess I never noticed, but is it normal for the AFR's to go rich briefly when shifting?

Other thing that's interesting is that my car seems to be drinking oil. This last time I started it up I got the engine oil light and I pulled the dipstick and was down a quart...I just changed the oil about ~400 miles ago. I get a decent amount of white smoke when the car starts up cold (it's not HG smoke, I know what that looks like...it's not billowy smoke...kinda like what it looks like when it's real cold outside...even when it's not that cold outside, if you know what I mean...) until it gets up to temp then it's normal...no engine temp issues and the headgasket is new about 2000 miles ago... I do get a lot of black'ish moisture out the tailpipe when idling cold as well...so maybe something is leaking somewhere...

It's just bizzare...it's a mess until it gets warm and then everything seems to calm down...of course it would be this way...I my inspection expires this month...so why wouldn't the car go ape**** on me when I have 2 weeks left to get it done...

choinga

One other video - from the engine bay...

The fuel pressure isn't clear in the video, but it's bouncing around at ~44PSI.

The thing you'll notice is that when I pull the vacuum hose off the FPR it smooths out (if I leave it off for awhile, an occasional surge) and when it goes back on you get that syncopated surge again. It wasn't near as pronounced I don't think because the car was 'sorta' warm...temp gauge was all the way down but it hadn't been sitting over night or anything...just a few hours...

http://www.youtube.com/watch?v=96obzrjTr8Y

fast951

Determine if your engine is running rich or lean first (I'm guessing lean).

When you disconnect the vacuum line from the FPR, the fuel pressure increases, the injectors will dump more fuel. This tells me that your car is running lean at cold start and the extra fuel makes it happy.
There is a slight possibility (not likely) that the engine is running rich, causing the RPM to hunt. Disconnect a vacuum line other than the line to the FPR, note what happens.

A few things to check (not in any order):

- Vacuum leak - Unmetered air is entering the system after the MAF causing a lean condition.
- Engine Temp sensor - If it's not sending the "cold" signal to the DME, the DME will not richen the mixture. Or if it's sending "too cold" of a signal the DME will over richen.
- TPS - if when cold, it's not sending the Idle signal, the car will run badly
- ICV - If it's sticking a bit when cold, then freeing up. (Tap on it when it acting up and/or disconnect it).

The time it takes the problem to clear, may not be related to engine temperature, but it is the time it takes the O2 sensor to heat up before the DME goes into closed loop. You can verify this by monitoring the AFR.

choinga

Thanks John. Wideband at cold idle definitly does not show lean...in it's normal place around 13ish until it gets warm. I'll have to re-check another line when the car is cold again. Any one in particular - maybe the one off the BOV?

The three things you mentioned are all brand new...new as in brand new box new, so I hope it's not any of those. Should disconnecting the ICV before starting the car when it's cold isolate that as an issue?

I'll triple check wires and connectors to make sure everything is good...this one has got me...

Crazy Eddie

Did you ever find a solution to this issue ??
I am experiencing a similar problem
I changed out the:
Temp sensor
ISV and the TPS
Still getting the up and down RPM"s
Almost like it wants to stall, but at the last nano-second, the RPM shoots up then down ( Hunting )
After a few minutes, the car runs like it should, but even after a few minutes the car is not to temperature yet ....
Very Strange.... This all happened out of the blue! Car was running great ( except of the oil pan gasket which still needs to be changed...)
Totally Stumped !!!
Any tips would be welcome
TIA
Ed

Ashman

Resurrecting this thread. Did anyone here actually come up with a definitive answer for the idle surge at startup problem?

I'm having this same exact issue with my N.A. 1986 944. I don't remember this happening until it had been sitting for a while this winter. Now, when I start it up it almost dies, then surges back to life over and over until I rev it up a bit for a couple of minutes.

I guess I'll start with the easiest solution first - new temp sensor and go from there?