Ed Scherer

Car: '90 S4


Current Problem:

• When cold, seems to start and run fine for a few minutes.
• After a few minutes (and maybe even before it's actually "warm"), starts to idle poorly; smooth sometimes, but sometimes kind of rough, with occasional RPM dives down to maybe 500 RPM, and then a surge to recover. If put into gear, RPMs drop to 500 or so and it might stall without opening the throttle if one of the "RPM dives" happens.
• Revving to 1000 RPM and trying to hold it there, it's still uneven, with periodic RPM dives much the same as during idle. Revving even higher (e.g., 2000 RPM), it seems to smooth out pretty much.

Additional Observations:

• Fuel pressure at idle is pretty steady at about 3.3 bar (48 PSI).
• ARM1 air/fuel meter is generally oscillating from full lean to full rich during idle (normal, I believe). I couldn't really detect anything significant on the meter when the idle problems occur (probably not surprising given the very transient nature of the problem).

History:

• About a month ago (before I did a lot of maintenance on it), it was showing some signs of trouble: not idling smoothly, almost but not quite stalling on occasion. As I recall, it ran pretty OK until warmed up, and then would experience problems after warmed up. Not quite as bad as it is now.
• Subsequently, I did a bunch of maintenance (probably unrelated to the problem I'm posting about here, but just in case...):
  • Released tension (I don't think there was any) on front shaft clamp at the flex plate, measured crank end play (measured 0.18 mm), reclamped, added PKlamp, and marked with paint.
  • New TB/WP (PKsn'r, and Standke-modified WP), including new cam gears;
  • New front main seal and oil pump O-ring (while I was in there anyway);
  • Replaced MAF with JDS Porsche rebuilt MAF, as the original MAF was testing out (by Louie Ott) as being marginal;
  • Deleted air pump and all associated brackets and plumbing (and yes, I remembered to cap the vacuum hose that originally went to the air diverter valve);
  • New distributor caps and rotors;
  • Fixed some corrosion problems on #2 plug wire at distributor cap end;
  • Replaced all four fuel hoses in engine bay;
  • New passenger side fuel rail (to fix fuel leak);
  • Removed and resealed water bridge with new gaskets and O-ring and also replaced and/or re-clamped some hoses to fix two or three coolant leaks;
  • Removed and reinstalled oil filler neck with new gasket (another "WYAIT");
  • Replaced bad sealing boot on oil level sender connection at oil pan (made an improvised new boot using heater hose, OO faucet washers, and epoxy);
  • A/C evac/flush/charge (needed to correct a problem with too much oil in the system);
  • Replaced rotting alternator cooling hose;
  • Probably a few other things I'm forgetting.

Recent Fix Attempts and Diagnostic Tests:

• Suspecting the ISV, I tried some of the ISV cleaning/lubing remedies: PB Blaster / brake cleaner (per Morning cocktail for the ISV), but injected directly into the ISV hose (per post #5 in the thread CConfused! ISV WD-40 treatment for rough idle); I also later just tried spraying some WD-40 into the temporarily disconnected driver's side (U.S.) breather hose. While this didn't fix the problem, it seems to have improved things a little: it doesn't stall when left alone, but will surge a little and correct itself without stalling (still low idle RPMs, though).
• Remembering that I was running a rebuilt MAF, I got a loaner MAF (thanks, andy-gts!) and tried it. No difference.
• I tried listening... and thought I heard an air leak under or around the intake. Maybe it's a vacuum problem! So, this morning, I pressurized the intake (using a gadget pretty much like the one in Dwayne's post Testing for intake leaks). It's leaking: from 2 PSI to 0 PSI takes about 20 seconds. Using a stethoscope, I tracked down the leak to somewhere around #23 in the following diagram. I haven't yet removed or further inspected anything in that vicinity (I had to leave for work).
• Disconnecting/reconnecting (waiting a few minutes in between) the battery ground strap, thus resetting ECU brains, etc.; didn't make any significant difference.
• Disconnecting the O2 sensor and then disconnecting/reconnecting (waiting a few minutes in between) the battery ground strap didn't make any significant difference.
• Trying a loaner LH; didn't make any significant difference.
• Electronic diagnosis (still in progress): one significant finding is that WOT on TPS doesn't work.

Additional Theories Not Yet Pursued:

• Need to check all the main signals used by LH and EZK. As of 18-Aug-2010, awaiting additional diagnostic tools.

Questions at This Point:

1. How big is a "2 PSI to 0 PSI in 20 seconds" leak? Is it likely to be the source of the problem? (I'm hoping so).
2. Is it possible to R&R that Y connector #23 without removing the intake? How does it fit into the MAF boot and what kind of seal is used?
3. Any other diagnostic procedures I should be doing at this point, or is this leak the obvious thing to fix before looking for any other reason for the idle/surge problem?

Ed Scherer

John Speake suggested (off of Rennlist) that I try disconnecting the O2 sensor so that the LH runs without it and then see what happens. I'll give that a try ASAP (this evening or tomorrow morning) and report back.

This seems like a good suggestion, as I was thinking that the heat was just opening up the vacuum leak further; I hadn't considered that it's probably more significant that warming up (and thus the O2 sensor signal becoming trustworthy) triggers the LH going closed loop, using the O2 sensor signal. So, it's a whole different ball game cold vs. warm, with lots more stuff going on when warm.

Bill Ball

So, you really, really don't want to replace that ailing ISV, eh, Ed?

I've seen other 90s that don't work with the Spanner too.

Anyway, you are suspecting a vacuum leak under the intake. Looks like you should bite the bullet and do an intake refresh if you haven't and proably toss in a new ISV too. It could be sticking when it gets hot.

Gary Knox

Ed,
I had a small amount of idle surging with my '89 S4 manual. Also, had heat at the AC outlet! Did a vacuum on the line going from the 4 way splitter into the cabin, and found when the switch was on, there was a massive leak. Checked the solenoid manifolds in the front of the console area, and found a leak in the orange line - I'm sure it is the air recycle diaphragm. Plugged that line at the solenoid manifolds with a golf tee in the rubber connector, and voila - no more surging, and GREAT AC. Since this is a track car, I'm not too concerned about not having the recirculaition option available.

Gary Knox

Ed Scherer

I was hoping for cooler weather (like Winter) before tackling an intake refresh. It's damn hot out there right now (about 100 °F and 50% RH in the garage) and I'm not too motivated to spend any more time out there than I already have.

Regarding ISV R&R, can it be done without removing the intake from the front with the water bridge and oil filler neck out of the way? I was looking right at the front of it a couple of weeks ago, but I don't know if there would have been enough room to R&R it. I'll be annoyed if I missed an opportunity.

Ed Scherer

Well, the vacuum leak appears to be in or around that #23 Y. It looks like it might be possible to replace it and inspect the related hoses withing pulling the intake. Or not? How does that Y connect to the MAF boot?

Alternatively, with enough duct tape...

Ed Scherer

Thanks for the comments, Gary.

I actually closed off the vaccum line going to the interior so any HVAC-related leaks wouldn't interfere with the intake pressure test. FWIW (and thanks for the reminder that I need to deal with this, too), I think I've got at least one leaky diaphram in my HVAC system (BTDT, so I'm not too worried about fixing that later and those tend to be pretty slow leaks anyway given the small hose diameters). But I should at least plug/cap anything on the interior that I find to be leaking, as you've done.

jon928se

Ed

A couple of observations from my recent experience with erratic idle - sames symptoms as yours.

The failing LH in my GT (eventually got to the point it would fail after 30 mins or so classic LH failure symptoms of injectors firing with ign on Now rebuilt by JDS) struggled to control idle. The rebuilt unit was better at idle control but not perfect - still the same symptoms as before/you have just not as bad .

Idle mixture was very rich 3.5-4.0%CO. (No Cats and no O2 sensor)
After the rebuilt LH was put in I got this adjusted back to 0.7% This seems to have cured most of the problem but I don't quite know why.

Mongo

You have to take the intake off to do it. Ed, it's really not that difficult and I pissed and moaned about my ISV issues for over a year before finally doing it in 101 degree weather out here in early September 2008. If you haven't done anything else in there yet, it would be a great time to replace the hoses under the intake, flappy solenoid and of course the knock sensors. Throw some Ford 19lb injectors from Summit Racing for only $235 in there too and you will be driving a new car.

Ed Scherer

Putting all that together along with the fact that I do have cats (aftermarket, in a 928 Specialists X pipe), that implies that a failing LH might be causing or contributing to this? (It's just that idle mixture for your car is via the manual adjustment vs. automatic via LH for mine, right?) My car has never done the "classic injectors going random" LH failure dance.


BTW, I do have a cheap air/fuel meter (ARM1) in my car. I'll try to remember to see what it's doing next time I do some more diagnosing.

Bill Ball

Good question...I think it's going to be hard to get to the two nuts that hold the ISV bracket on the bottom of the intake. The bracket is a band wrapped around the center of the ISV. Get a mirror and light in there and you might be able to see it.

It's remotely possible to get at that 3-way plastic fitting, but my recollection is it's jammed against the valley. I had to repalce it once and I think I gave up and removed the intake.

Ed Scherer

Yeah, well, I've been without the car for a month already and if at all possible, I'd like to defer the intake refresh a few months. Especially if it's really not needed right now (there's still no consensus on what's really going on!)

When I do it, I'll be refinishing (powder or whatever) the intake and the other parts typically done, so it'll take awhile.

BTW, good job on the "one-upsmanship" (exactly!) on the temperature. I'll be sure to note if it's really 102 °F when I get home.

Ed Scherer

OK, before going off on too many tangents right away, I'd like to focus for a moment on the one issue I know I have and that might be quickly repairable: the vacuum leak at that #23 Y that I mentioned in post #1.

1. Is that (from 2 PSI to 0 PSI in 20 seconds) a very bad leak? Enough to potentially be the sole cause of this idle/surge problem?
2. How does that Y connect to and seal with the MAF boot?
3. Can I R&R that Y without removing the intake? If so, how?

I might as well try to fix this right away and see what happens. I'd like to remove the Y and inspect it and the adjacent hoses as long as I know it can be done (without removing the intake). I don't want to try, only to break it because I don't know what I'm doing and make the problem worse.

Stromius

Ed - With the intake R&R, did you change the CPS? If not, could the CPS connector (under throttle wheel) be cracked and bad? Separate thread this week on similar symptoms (rough when warm) reminded me of an older thread where heat caused the CPS plug to have poor connection. Mine was cracked and didn't think of it as I had the intake off
YYMV..good luck!

Stromius

Thought the later intakes were too large for water bridge removal because 1 bolt is blocked.??

PSI difficult to measure even with HD guage contraption. I never got a good read and I have rock steady idle. Piston positioning likely needs to be just right.

Y connects as a barbed plastic Y insert into MAF boot. Yes, removal I belive with some removal of stuff around it.

Might just be worth removing the intake again and doing it proper.