r berq

98 Civic EX 4-door Automatic. D16Y8. 307K miles.

Read carefully, so you don’t just tell me to clean the IACV.

When the engine is started cold, idle goes to 1800 RPM and in a few seconds settles at about 1400 RPM which is OK. But when the engine warms up the idle remains at 1400 which is way too high for a warm engine. It stays at 1400 no matter how far I drive.

HOWEVER, if I shut off the engine and restart it hot (anywhere from 2 seconds to perhaps 30 minutes after shutoff), the idle quickly settles to 700-800 RPM, right where it should be, and stays good no matter how far I drive. In other words, whatever the engine temperature is when the engine is started, the computer sets idle speed appropriately for that temperature and then never changes it as the engine gets hotter.

IACV was thoroughly cleaned, no change. IACV was then replaced with a new authentic Honda IACV. No change. IACV is working properly, because idle is adjusted properly when turning on/off lights, AC, shifting between neutral and drive, etc. In all cases the idle is stable -- that is, it may be too fast or it may be just right, but it doesn't fluctuate up and down.

Any ideas what’s going on here?

P.S. This has been getting gradually worse over the last 100K miles. The only thing I can think of that is gradually deteriorating is the catalytic converter. Does feedback from the cat cause the computer to adjust idle speed? There is no CEL and no codes stored.

Nave43

Clean and adjust your FITV (Fast idle thermo valve) . Its what adjusts idle once the car is to op temp.

\/ DIY \/
https://honda-tech.com/acura-integra...valve-1564019/

:edit:
Scratch that your car probably doesn't have one... Its a year or two off.

r berq

Right. My car doesn't have one.

White_EG1

Try the idle relearn procedure. By unplugging the IACV connector when the car is running, adjust the idle screw. Turn the car off, and unplug neg battery cable then reinstall.

r berq

Tried it. No change. As I said, the idle is fine IF the car is started hot, so the computer does know how to idle.

Former User

Clean the throttle body (TB) and adjust the throttle cable to have some slack. Remove the entire TB to clean it. Spray with brake or carburetor cleaner to remove the sticky gunk that prevents the throttle plate from fully closing. Buy a new TB gasket. The gasket only costs $2.

r berq

That would make sense if the idle speed was ALWAYS too high, but that's not the case. When the engine is restarted hot, idle speed is fine. Seems like it must be something electronic.

Jimi Hondrix

Weird. If it was a faulty cat I would suspect a constant rough idle and a CEL. If the cat was partially clogged maybe it could cause this but, seems to be an electronic issue as you stated. Have you tried cleaning all the ground connections and afterwards resetting ECU?

Another thought is possibly a faulty temp sending switch or break in the temp sensor wiring.

TomCat39

Just for fun I would still try RonJ's suggestion first to see if it changes anything.

After that, I would swap the ECU as I think it's having problems switching to engine temp and it is what is controlling the idle via temp readings.

It seems the temp sensor is working but the ECU isn't adjusting after the initial reading. Faulty ECU.

I'd still clean the throttle body as it's a cheap fix before going after the ECU which is much more expensive.

Former User

This^ erroneous assumption led you to an incorrect conclusion.

r berq

I swapped out the ECU several years ago for another problem, which was not fixed by the swap. I still have the original ECU, so I will try swapping it back. Good suggestion.

r berq

Fast forward several hours. I swapped the ECU for the original. Not fun in a cold garage, because the wires and connectors are pretty stiff. Alas, no change. Rats!

Former User

r berq

Good suggestion, but throttle body has been cleaned and made no difference. There is slack in the cable. When engine is hot and idling fast, and is shut off and restarted immediately without touching the accelerator, idle drops to the proper speed, which would not happen if the throttle plate was stuck open. I have also disconnected and plugged the vacuum line to the cruise control, to make sure that wasn't doing something funny.

Former User

What leads you to think this^ is impossible?

When did you clean the TB? How did you do it? Provide details.

r berq

Not impossible, but HIGHLY unlikely. Engine restart can be via the starter motor, or by turning the ignition off and back on while driving, or by turning the ignition off and back on while idling in neutral, and the delay between off and on can be one second or 30 minutes as long as the engine has not cooled down too much. Idle correction is immediate and is 100% consistent. A sticking throttle plate would not consistently free itself up EVERY time the engine is shut off and back on, it would be intermittent.

I cleaned the TB on the car with rags and carb cleaner. My mechanic removed the TB from the car and cleaned that and the IACV. I didn't watch him so I can't provide details on how he did it, but he generally does things right so probably he did that right, too.

Former User

Remove the air intake system. When the engine has the high idle problem, use your finger to block the hole inside the TB leading to the IACV. Does the idle speed drop?

If not, use the palm of your hand to block the top of the TB. Does the engine stall?

r berq

Yes, it drops.

Former User

Possibilities:

• Idle speed needs adjustment as per service manual
• Air in cooling system
• Dirty, sticking IACV
• Bad IACV
• Bad ECT sensor
• Bad or uncalibrated TPS

r berq

Ha! Where did you find that neat list? Or are you just real smart and made it yourself?
Anyway, last item on the list is the TPS. I checked it today and it is definitely out of spec -- shows 1100 ohms at closed throttle, and should be between 500 and 900. So I'm guessing the computer never "sees" the throttle as closed and therefore never sees a need to decrease the idle. Thanks for your help, Ron.

The bad news is, I intentionally tweaked the TPS to (successfully) resolve a transmission flare between 2nd and 3rd gear. That's a common problem -- don't remember if I discussed it here on this forum -- essentially the TPS adjustment makes it appear to the computer that I am accelerating harder than I really am, and the computer adjusts the shift points enough so it doesn't flare. So I need to live with the idle problem, or go back to the transmission problem, and funny idle is the lesser of two evils. Unless I can somehow fake out the TPS circuit to fix both the idle problem AND the transmission flare problem....

Former User

Try recalibrating the TPS voltage to spec to see if the idle issue is eliminated. If so, do 3 consecutive ATF drain-refill cycles in an attempt to address the flare issue. Use Castrol Import or Honda ATF.

r berq

I did 5, count 'em, 5 ATF drain cycles over the course of a couple weeks. I figure that gave me about 97% new Honda fluid. It didn't help in the least with the flare. I can't find the link now, but there is a thread on another forum that has been going on for years discussing this flare problem with 98 Civics. Symptoms are that the car shifts fine when cold, shifts fine under hard acceleration, but flares under gentle acceleration when hot. One guy fixed it with a new linear solenoid, but that didn't work for others. One guy fixed it by hacking his ECT sensor to show hotter than real coolant, but of course that caused starting problems. I fixed it by hacking my TPS adjustment to simulate hard(er) acceleration. At this point I will just suffer with the bad idle, since the shutdown/restart after warmup pretty much takes away the problem. Unless I can think of a simple way to tinker with the TPS -- I could short it out at the low-resistance end -- but a TPS doesn't look like it can be disassembled easily.

Former User

Hmmm...I've seen many Civic auto trans issues fixed or improved by ATF drain-refill cycles. A bad or miscalibrated TPS can also cause auto shifting problems.

I'm suspect of the TPS in your case for both the idle and transmission issues.

r berq

Idle, I think you nailed it. Transmission, not so sure. The transmission flare existed for years with the original TPS. That TPS failed causing dramatic transmission shift problems. I replaced it with an aftermarket no-name piece of crap that caused little hitches at highway speed whenever I or the cruise control moved the throttle small amounts, and still had the flare issue. I replaced it again, with a Dorman brand TPS, which works fine but still has the same flare problem.

I suppose it is possible there is a transmission problem that the TPS tweak compensates for, by increasing the shift-point RPM so fluid pressure is higher. But I'm really beyond my depth with that speculation -- and if that were so, then why would the tranny shift cleanly before warmup? Interestingly, I had a similar flare with my 1984 Nissan Sentra automatic, again under light acceleration but never when accelerating hard -- and I don't believe the Nissan's tranny was electronically controlled. But that was my wife's car, so I didn't obsess over it like I do with mine.

Jimi Hondrix

My daughter's '98 Civic EX, AT, 145k miles ... seems to do the same exact thing described above. I did a 3x3 drain and fill with Castrol maxlife ATF and added 6 ounces of Lubegard Red a month ago. Smoother afterwards but, still flares under gentle acceleration when hot.
I don't think I will mess with the TPS though.