Snark Shark

My manual 86.0 US car sounds a bit funny when I blip the throttle. The revs oscillate a few times before settling down to idle. It oscillates about once per second. Drops down to about normal idle speed and then bounces back up to about half the rpm that it revved to the previous time, as if the idle stabilizer is "overshooting" the correct idle RPM. Is this normal?

NC928S4

Normal no. A failing ISV usually causes engine to die at idle. A high idle could be many things but more likely a vacuum leak. Worf has an excellent write up on testing components for causing problems. Check this out.

Mrmerlin

Well the ISV in your model is a wear out item so there is a good chance if it has over 40K miles its worn if its closer to 75K miles then it may be time to replace it.

The way to check it is to remove it and use a pick to move the armature,
what happens is the brushes eat into the armature and make holes that the brushes grab.
thus making the armature stick.

These pictures are from an 1986.5 928S ISV

NOTE the 1987 and newer S4 versions work on a magnet instead of brushes on an armature

NOTE the different wear of the 3 brushes.

NOTE the armature in this picture was repaired with silver solder to restore a smooth surface.

It has an additional 30K miles on it from the repair, It would have been better to replace this sooner than later .

NOTE if the ISV gets stuck then the computer may be damaged

Snark Shark

OK. I never had it stall on me, so I'll have to look for vacuum leaks, and probably replace the ISV anyway once I do some top end refresh, since this is a 160k car. Got my fuel lines done, so that is next on the agenda.

Imo000

Check the MAF sensor and if it's sitting properly and there are no leaks around it.

Snark Shark

Now that you mention it, I'm a little puzzled by the MAF. It's strange how it seems to fit just fine when rotated to any angle. And it was immediately loose once I loosened the airbox nuts. Is there some sort of trick to getting it aligned properly?

Snark Shark

So I finished up taking the whole organ pipe manifold off, and replaced all the questionable vacuum hoses, rubber sleeves, and the manifold gaskets, capped off a vacuum port for the EGR system that the PO apparently neglected when he did the air pump delete, cleaned the ISV with WD40 (it was very dirty), and put it back together, and the behavior seems exactly the same. This time, I notice that when it oscillates, it's actually undershooting correct idle speed, and almost stalling. I think it was doing this before because it sounds the same.

Got the MAF seated properly.

Checked the fuel damper vacuum lines, and they are odor free.

I tested the ISV with a 9v battery. Seems to open and shut instantly. The seal is not perfect when it's shut, but it does take some effort to blow into it and get any air flow out the other end. Normal?

I made sure my ISV was shut, reinstalled it, and blew into the vacuum port for the brake booster. My breath seems to easily come out of the "U" where the MAF goes. It takes much less effort than blowing thru the closed ISV. I put a one quart paint can lid over the "U" opening (fits almost perfectly), and then it did take some effort to blow and hold the lid down. Could there be something wrong with my throttle body, causing a metered-air leak? I don't seem to have the typical symptoms of an UN-metered leak: low idle, hesitation, bluish exhaust.

Mrmerlin

did you install a new ISV or are you using the old one?
did you understand how the photos I posted on how the ISV will wear out can cause the issues your having .
For your engine to run properly you will need to have every part in top condition,
that includes,
a fresh MAF and intake sealing O ring,
a new ISV,
a fresh crank position sensor,
a new TPS and short harness to it,
a fresh temp 2 sensor and new AMP connector,
along with all of the rubber bits.

NOTE Since you have indicated the engine has 160K miles,
its time for all of these parts listed to be replaced

Snark Shark

I understand the car is older than the lifetime of all of those parts, but isn't the inability of the throttle body to hold a vacuum a more alarming problem?

Imo000

Get some starter fluid and gently spray with small bursts around the intake. There still could be a major leak somewhere.

Snark Shark

I'm a little reluctant to try the starter fluid test because the area of interest is all under the intake manifold, in a confined space where there's several electric switches. Might have to get a smoke machine.

I think replacing the ISV is starting to make more sense to me now. When I blew into the vacuum port, the amount of airflow coming out of the air intake seemed way too high for the leak in the ISV to account for it, but reminding myself that the S3 throttle body has an idle adjustment nut that permanently opens a little valve, the combined effect or that, plus an ISV leak might account for it.

Still would like to know if a brand new ISV normally doesn't close quite all the way, just to satisfy my curiosity. I've read that some types of ISV do normally allow some air to pass in the closed position.

Another interesting thing, I just ran the car for a while to get it up to temperature, and the problem seems much less severe when warm.

FredR

The S4 ISV passes enough air in the voltage off position to hold a stable idle when warm and then closes the unit further if the idle is higher than the set point. Thus it can maintain an idle if the ISV fails.

Whether or not your model does something similar I do not know.

Rgds

Fred

Imo000

Try setting the base idle and see what happens. Maybe it's set too high and the ECU cuts the fuel to try lowering it and this causes a constant oscillation.

Snark Shark

Car has developed an 1100 RPM idle. Tightened the idle adjust all the way, and it continued to idle at about 700. Counted the turns and put it back where it was, for now.

Driver's side cam cover failed the smoke test. Hard to pinpoint with the plenums on. Probably multiple tiny leaks there. I had previously taken the other cam cover off, inpsected it and replaced the bolt o-rings, but the gaskets all looked like new, so I decided to leave the driver's side alone. Oh well, guess I'm doing a reseal job on that one too.