I have a 94 civic ex.
When my car cold starts, it idles fine.
When my car warms up, it idles fine.
When i drive for 20+ min, and then stop at a red light, the idle goes up to 2.5k... and sits there.
If i turn the car off, and turn it back on, it idles at 2.5k...
Does anyone have any ideas of what would cause this?
Thanks...

 

fast idle thermo valve.

 

I'd agree, the EACV that also works to control the idle on that year, doesn't have the ability to raise the idle that much.

 

I took the fast idle thermo valve out, and tightened the insides back up, put it all back together and it did not solve the problem. I then went to the junkyard and got another one, made sure the insides were tight on it, and threw that one in, didn't solve my problem. I am now getting engine code #5, MAP sensor. I got 2 different ones from the junkyard, and although it changed the idle, it didnt fix it. I just went out and spent 75$ on a new MAP sensor, and this made it run worse than it has before.... what am I doing wrong, I am negatively progressing!! Is there a time period in which the ECU must learn the new MAP sensor, although its OBD1 so i doubt that..

 

If you reset the ECU after the install ( and you should disconnect the neg. battery cable anyway when installing something like that, just to be safe) I'm pretty sure it's not an issue w/ the ECU adjusting.

Have you checked for vacuum leaks? Has the car been like this since you got it? If you got it like that, there's the possibility that the idle set screw was adjusted. From the factory it should have some epoxy or something over it, so you might be able to tell if it's been messed with.

Also, when you come to a stop (w/ engine warm) and it idles at 2,500rpm, does it go there as soon as you put the clutch in, or does it idle lower and then go up after you've come to a stop?

 

Yes, i disconnected neg. terminal when doing this. I unscrewed the sensor, and it seemed to me that there was plenty of vacuum pressure there.. how would i go about testing that though? i do not hear any air leaks anywhere near the sensor, the lines, or the charcoal canister.. The car started acting this way a couple days after i replaced the clutch... The idle goes up to 2500 as soon as i let off the clutch, and while i am driving, it feels like its at 2500 too, i do not have to give it gas when i start moving in 1st...... my mechanic said that the #5 engine light, MAP sensor usually does mean vac leak if the sensor is still good...

 

For it to be idling that high, (if not the FITV, which you've ruled out) I'd think it would have to either be a a vacuum leak or the throttle plate is actually sticking open - I'm not saying that's likely, but it's pretty high, it must be getting significantly more air than it's supposed to. Not necessarily through a vac. line itself, just a leak somewhere letting air into the manifold or even between the manifold and head.
I've yet to have to do this, but you can spray starter fluid around the vac lines and manifold, if the revs go up when you spray a certain area, you've found the leak.

I'm assuming your Electronic Air Control Valve is constantly closed (or as much as it can close) trying to lower the idle, so it's not making adjustments and making the idle jump or fluctuate like often happens.

 

The throttle plate is def not stuck. After this was happening, I took the tb off of the intake and i used carb cleaner to clean all the black residue out.. it was really dirty. I then got a new seal and put it back on. It is not sticky at all. I am guessing i have a vacuum leak, so im gonna go hunting for it. Why cant it just be something simple???

 

I just went around all the vac lines with starter fluid and no effect.... I am gonna get a different FITV and change that cuz i am out of ideas...

 

While you're at it, I'd remove and clean the EACV, just in case. I'm fairly confident if wouldn't raise the idle that much, even if fully open all the time, not on that year at least, but it's easy and cheap to clean it out w/ carb cleaner... maybe there's even an issue w/ it's seal to the manifold, but unlikely and I'm just throwing stuff out there.

In the past when I've removed either the FITV or EACV, I've both bled the system w/ the rad. cap at 1/2 way (or can just leave cap off) as well as no bleeding whatsoever, with similar results. If you refill the rad. you need to bleed things of course, but in my experience w/ this particular car it doesn't seem necessary after dealing with FITV or EACV - I run water wetter though so... point being I'm not thinking air trapped in or around the FITV would be the problem.