1999 Accord LX Coupe
2.3L F23A1
5 sp manual
196K miles


Background: When driving my car and up-shifting into any gear the idle hangs for a split second after pressing in the clutch and sometimes even goes up 100 rpms or so. Unless I wait for the RPMs to come down its hard to drive the car smoothly/enjoyably, and it just becomes really annoying when driving up hills when I need to avoid loosing speed by shifting quick. If I have to accelerate to high RPMS (red-line) in traffic, while shifting I literally have to slowly count "one-one-thousand" before letting the clutch out before the RPMS drop to where they're supposed to be and I can re-engage the clutch. I've had several manual cars and never had a problem like this.

That said, nothing else is wrong with the car. No surging idle, high idle, etc.

I have a long 100 mile commute every day and this idle/shifting thing has been driving me mad.

I've read almost every thread I could find about this subject and have tried nearly every solution that applies to 6th Gen Accords including:
-Adjusting the clutch pedal in both directions
-Adjusting the clutch pedal sensors in both directions (thinking they might have some affect on the idle hanging when the ECU doesn't know the position of the clutch)
-Removing and thoroughly cleaning the IACV
-Checking for Vacuum leaks on vacuum hoses and the upper/lower intake gaskets
-Cleaning the Throttle Body and associated sensors
-etc... etc..

Most of the time when driving my idle hangs when shifting; however at totally random times when driving the idle will drop quickly like it should and I can drive normally with no issues. It only acts normal for a little while before going back to the normal hanging idle issue.

It seems as if this is either an ECU issue or an IACV issue, being it's at time intermittent nature and that nothing else is wrong. I'm leaning more towards the IACV being the culprit as it is original, but has anyone else ever had the problem and corrected it ?? I just want to check before dropping $100 on a new IACV

Anyone else have a suggestion or opinion on this?

 

That is an odd problem.

My first suggestion would be to get the codes read at a parts store just to be sure this isn't a sensor.

The next time you drive and you are cruising in the middle range of a gear (say 45 mph for fourth gear), just push in the clutch pedal. Does the rpm rise and stay there, or is this happening at the top of the rpm range?

Make sure that you have the pushrod to the clutch pedal adjusted properly, because you can wear out the clutch if not adjusted correctly.

 

I read this a little differently, and I predict there won't be any error codes. But check anyway.

Has it been doing this all along? Just bought the car so you don't know? Or did it just start?

My first suggestion is to do the procedure of setting the BASE idle, then forcing it to relearn. Look for a post by me in the DIY section about setting the base idle.

Finally, if it's done this forever, and/or the base idle doesn't change it, then it just might have to be that way. For a range of model-years, because of EPA regulations, they have to do that because a large amount of exhaust emmissions happen when you close the throttle suddenly. So they do various things (mechanical & software) to prevent the throttle from closing too quickly. And I agree it's a pain.

 

I have a code reader, no codes are set or are pending.

As far as pressing in the clutch while cruising, most of the time this is when the idle hangs and makes daily driving a chor. It is really obvious when at high RPMS though.

I tried adjusting the master cylinder rod both ways with several turns of the rod and it still does it. It is less obvious when the pedal is longer and clutch grabs closer to the top. I readjusted it again after taking videos and it is more derivable, but the underlying slow idle problem remains.

I've had the car almost year and its always has done this. I'll try setting the base idle over the next few days however I have my doubts, it idles right around 750-800 when warm. From what I read, that is what its supposed to be at?

It seems as though right after pressing in the clutch the idle hangs for a split second, then it drops normally. Check out the first video to see it at high RPM.

Check out these videos and you can see:

 

Terminology: BASE idle is with the IACV unplugged. It's low, like 550rpm or so.

If the base idle is set wrong, the ECU will compensate by opening or closing the IACV until the operating idle is 750. So you THINK the idle is OK. But it's less effective at handling situations like dropping to idle from engine-braking, or suddenly dropping when you close the throttle to shift.

 

Ah, I see what you're saying.

I'm trying to adjust the base idle now. I have the car warmed up and I unplugged the IACV to check what the base idle is, but the idle is now bouncing a small amount with the IACV unplugged, it doesn't stay steady enough that I can see what the base idle is set to.

Is it normal for the idle to bounce around with the IACV unplugged?
There are no vacuum leaks that I can find.

 

Here is what I'm talking about with the IACV unplugged:

Is it normal for this to happen with the IACV unplugged? The idle appears high, should I adjust the base idle while its bouncing a little or is the bouncing a sign of another problem? (yes I had my lights on to record this, however it was doing this before I turned on the lights to begin with).

 

That's too high. Looking at the daylight pictures, the first lightweight line is 500, next is a shorter lightweight line at 750, then the heavy line a 1000. Base idle at 550 should seem like it's about to stall.

Have you found the idle-air adjust screw? Above the bore of the throttle body?

Otherwise, I'd say vacuum leak in spite of what you said. Spray some throttlebody cleaner at various places to see if the RPM changes. For example air leaking in through the throttle pivot shaft.

 

Yes I found the screw I was just waiting to see about the jumping idle with the IACV unplugged before messing with it. Mine had white silicone smeared over it so it wouldn't come loose, not sure if that's factory.

Anyway, I had time today, I sprayed the whole intake system down including upper/lower gaskets, all hose connections, all seams/gasket areas (including back by the IACV) and the idle still bounced around with the IACV unplugged (with it plugged in, the idle is fine), saw no change.



I tried adjusting the screw anyway, no matter what position the screw was in, the idle didn't change really. I had it in as far as it would go, as well as out all the way literally (I was holding the screw in my hand), with no major change in idle. With the screw turned all the way in the idle jump thing was narrowed down , only about 100 rpm jumps up and down. Not sure if it matters, but I cleaned off the tip of the screw and sprayed carb cleaner in the hole to clear anything out.



Not sure whats going on here, it seems like there is a vacuum leak but again I can't find any. And when I put my finger over the bypass hole in the TB it stalls just as it should.

I don't even understand why the idle jumps up and down with the IACV unplugged since nothing is regulating the Idle. I'd imagine it would be a steady high or low idle, whatever it may be.. even with a leak.

Can another sensor be playing a roll here such as the MAP?

 

I'm pretty sure when the IACV is unplugged, it should close. So with it unplugged you still have suction at that little port? Cover the port and it stalls? That sounds like air is getting through the IACV when it shouldn't.

I'd try a junkyard IACV because mostly they get dirty but not broken. Has to be 1998-2002. Older years was a different version of the IACV, not compatible.