bayrrej

-- Refurbished Intake (S4 intake added by PO -- still w/ all S3 electronics) replacing recommended gaskets
-- Added Bosch 4 hole injectors
-- Replaced MAF w/ rebuilt from Ott's Engineering

Upon starting, would idle around 900, would then totally stall around 2000 rpm just once but EVERY time after starting (warm or cold), then idle would gradually increase to 1200 -- would stay at 1200 if started warm.
To fix that removed intake and replaced ISV (w/ same 3 prong S3 version) the old one failed completely on testing.
Upon starting now, the idle immediately leaps to 2200 and is rock solid there even after warm up.
Smoke test is negative...
Linkage appears functional and unbound
Does not have cat, but headers, does have O2
Any ideas?

davek9

I had a bad LH do that, output transistors for ISV shot, mine would start out around 900, when warm go to 1,500.
If you're ISV is new, see if u can get a loaner ECU.

Edit: sorry just read this is not an S3,

Good luck, could be # of things, do u have a shark tuner?

Have u pressured tested intake for leaks?

bayrrej

That's the conundrum -- everything BUT the actual intake is S3 so the LH and ECU are still S3.
It does smell like electronics to me esp. since smoke test turned up nothing...
I'll see if I can get my hands on an ECU...

FredR

Not that I think it would explain your conundrum but do you know for a fact that you have 24lb injectors installed? If someone was so S4 focussed maybe they did not know they have different size injectors. If this was the case it would be silly lean and surely would not run but...?

It is beyond me as to why anyone would put the S4 manifold on your motor given there is no mechanism to modulate the flappy - a first from my perspective and a rather bizarre thing to do me thinks but I try not to pull my nose up at something just because I do not understand or cannot make sense of it.

Presumably the throttle switch modus operandi makes no difference for the two models given they are on/off contacts.

Are the inlet runner ports configured the same on both engine types to permit a spot on retrofit like this? Obviously a fast idle is usually associated with an uncontrolled leak in the inlet tract. As I understand the S3 has a different type of ISV but does this mate into S4 inlet tract ISV hoses without issues?

Has the motor previously been known to run with this manifold fitted and thus a new problem or is it a project that never ran and was purchased in such condition?

Apologies for questions rather than answers but it kind of seems that we need to establish we are all singing off the same hymm sheet here and know what hymm we are supposed to be singing.

Rgds

Fred

bayrrej

1st, this setup has apparently been run for the better part of its life having been done by the first owner prior to 2003.
A flappy controller has been added and is apparently functional.
I replaced the injectors w/ recommended bosch 4 hole injectors spec'd to replace the existing injectors.
I ran a smoke test with no apparent leaks...
Prior to my mucking about, it ran just fine, so the setup, however bastardized, does work.
Maybe the PPO just liked the look better? Who knows?
At this point, it appears to be the Battle Hymn...
Thanks for the reply.

FredR

OK- that clarifies a few points-

If you ran a smoke test and saw no leaks then the only other thing I can think of at the moment is a throttle that is not returning fully to its seat.

Have you checked that the idle switch contact is making/breaking correctly? Also if the throttle body is that from the S4 then there is a set screw throttle stop that under normal circumstances should not be touched. If for any reason that is out of position it will stop the butterfly from closing and the engine will idle higher - unlikely but do not discount anything until the cause is ascertained.

hwyengr

Don't tear anything apart until you try a different LH. If the only thing you've changed since the last post is the ISV, I'd bet your LH idle circuit is fried and sending a "full open" signal to the ISV. Your old one was frozen in place, so that's what limited the idle to 1200.

It could be another leak, too, but eliminate the LH possibility before potentially making more leaks by disassembling.

bayrrej

That sounds logical, as the ISV WAS the only thing I changed.
I'll give it a shot -- thanks for the response...

davek9

That all depends on How you did the Smoke testing.

My suggestion was to "Pressure Test" the intake from the MAF input (remove MAF) and insert Pressure Tester and see if it holds pressure and doesn't leak.

IIRC to replace the ISV on an S4 requires removing the entire Intake (unlike an S3) so anything that can leak is suspect now, including the mismatched S3 heads to S4 intake (oval vers round).

GlenL

Gotta be a vacuum leak or other source of "false air." It won't run that fast just with a mixture or timing problem. I'm not sure the ISV can boost it that high.

I always disliked the expression "false air," for what that's worth. "Intake leak?"

I'm surprised the smoke test didn't find anything. Can you describe how that was done?

My guess: leaking intake gasket.

hwyengr

It wouldn't be surprising if the ISV can get it that high. I was reading the NHTSA report a while back about the Audi 5000 acceleration incidents in the mid '80s. A wide open ISV on that 5-cyl zinged up to 2000 rpm and could get the car up to 35 mph if in gear.

I still think you should work down the fault tree, checking a variable while leaving the gaskets and hoses in their current state. If that doesn't fix it, then start checking for leaks.

bayrrej

Smoke tested w/ one of the more legit-looking paint-can kits from the web. That part seemed to work fine.
But if there were a bunch of small (gasket-type) leaks it might be harder to pinpoint.

A pressure test sounds like a good, least invasive, next step.
Interesting on the potential oval to round mismatch... The gasket was for the S4 intake and didn't show any obvious mismatch...

Thanks for all the input and suggestions.

Petza914

Since you have a rock-solid idle, even though it's a high one, If you have a vacuum leak somewhere, spraying some starting fluid around the intake manifold and other areas will create a momentary rise in idle if it's being sucked in by the leak. Might be an easy test to do first. I would also check to make sure that the throttle cable isn't binding by manually reaching in and trying to close it all the way.

My car being a '79 k-Jet is a comletely different injection setup than you have, but when I upgraded to a powdercoated Euro Spider Intake, while everything was apart, I cleaned the throttle plate and the intake merged collector that all the legs run to. Upon reassembly, I too had an elevated idle and it would hang during shifts. Nothing else changed, but what I found is that the throttle plate wasn't closing all the way when I would accelerate slowly and then push in the clutch to shift. If I accelerated hard, the RPMs would drop a lot further during the shift. What happened is that in cleaning off the build-up on the throttle plate and where it closes I had created better flow past the plate at idle so although the plate was returning to the same position it had before, it wasn't as "closed" because the gunk was gone. I created a simple mod and just added a light helper spring to the throttle plate cam on the outside where the cable attaches and things run perfectly again. That's why I say, check the mechanical components as well as the electronic ones. Here's a pic of the helper spring I added.