Hot idle speed oscillating

Have a question or want to start a discussion? Post it! No Registration Necessary.  Now with pictures!

Threaded View
My 1990 Mitsubishi Eclipse (DSM 1G 2L non-turbo) is starts and idles =
cold fine, and runs fine warmed up.  When warm and I put it in neutral =
(AT) the idle oscillates between 1800 and 2200 with about 1/2 second =
cycle time.

MAS was the first guess -- so I put a rebuilt mass airflow sensor assy =
in and same problem.  Checked all the connections and vacuum - all OK. =20

Also took the throttle assy apart and cleaned all - helped the running =
but still oscillating.   Cleaned Idle air motor, checked voltages and =
resistance for all sensors and also voltages coming from the computer, =
and all OK.

About out of ideas -- anyone have some thoughts on how to start =
troubleshooting this?

Thanks in advance.

Bob


Re: Hot idle speed oscillating

From sci.electronics.repair

EXCELLENT info.   J understood the problem right off the bat and that's =
exactly what it was.  The guy's a genius!


Hold on, fuck all this. If you can't find it in the codes, it can only =
be a few things.=20

Dig up, the first check is to keep it running until it does it. If it =
won't do it out of gear get some really good wheel chocks. When the idle =
hits a high point, unplug the IAC (or it's equivalent). If the hunting =
stops there are one of two things possible, but one solution might do =
it.=20

More than likely the IA path is partly clogged, especially where the =
pintle goes down. This skews the actual IACS from the true reading, =
indicating to the computer the amount of IA is higher than it is.=20

The IAC responds based on presets in the computer so therefore it will =
cut the IAC postiton to where it does not give enough IA. Then the =
computer goes into "stumble" mode, thinking (rightly) that it is about =
to stall because the idle speed has dropped too much. It will then open =
up the IAC to allow more air. The computer will always look for the =
lowest idle attainable. This sustains the oscillation. (hunting)

Now, if you rebuild the throttle body, it will be cleaned and hopefully =
be alright but then there is a possibility that the seat for the pintle =
valve has been eroded too much. You usually cannot machine this out =
because you cannot usually readjust the IACS to compensate.=20

But if you see a bunch of black shit in there of course clean it =
somehow. Throttle bodies really never need much of a rebuild other than =
that, if it's worn you might have to machione down a bore so the IAC =
drops lower into the hole. (same shit happened to carbs that were run a =
long time without an air filter, but the solutions were different)

However, if it is actually worn like that there is a workaround. There =
is a tang on the throttle linkage right there on GMs, it may be hard to =
find. On an American car it is a tang and you just bend it, on other =
cars you might have to shim it, or even find the actual adjustment. This =
will keep the throttle from closing too much. You MAY have to do =
something about the TPS if it is not adjustable.=20

The drawback to not replacing the throttle body for this problem is that =
the IAC pintle might get thrown down into the seat so hard the it will =
get stuck there, and then you will have a stalling ocndition and a =
constant check engine light.=20

So if you fuck with the idle position of the throttle, do it only as =
much as you need. (both of these solutions are against "code" but legal =
if it pases emissions, and it should)

If you don't do something though, it will eventually start stalling at =
red lights. Alot of people told me "well it was doing that idle thing =
for years and now it stalls". What, did you think it was going to fix =
itself ?=20

But it is not sensors. It is probably not the IAC itself, it is that =
things have changed for the IAC. The parameters of the intake cuircuit.

J

Site Timeline