scooty

The idle sticks at around 1500 rpm before returning to normal when I blip the gas. Is this normal? Please watch the video. The car is a 5-spd, in neutral and the motor is completely warmed up. The AC is switched off.

It also sticks at around 1200 rpm while coasting in neutral and will not drop down to normal until the car is completely stationary. This started to happen after I removed and cleaned the throttle body and IAC valve.

ezone

Do a relearn?

mikey1

you probably have an intake air leak, take it apart and recheck your work

scooty

Did you mean a vacuum leak?

mikey1

i mean you probably have an air leak somewhere PAST the throttle body, this will allow your engine to suck in excess air, causing a rise in idle speed

scooty

Ok, so this all started when i did the throttle body cleaning DIY in the forgotten tune-up mods thread. The only place I can think of for a vacuum leak would be the one vacuum line that connects to the back of the intake manifold and controls the evap system, I think. I checked the line, looks good.

I removed the throttle body again and can't think of any other place that vacuum could leak from. I also cleaned the Idle Air Valve again and reassembled everything and did the idle-relearn as per the forgotten tune-up mods thread. It's worse now!!

I took another video (see link below), As you can see the idle now hunts between 1000-2000 rpm before settling down. The car was at proper operating temperature and each time I rev the motor I let go of the gas, the car does the hunting rpm thing all by itself.

[autostream]http://autostream.com/ibcivicforums/?page_type=firebirdplayerthumbnail&framepage=1362& transactionid=1356577060-67865134&posted_by=_www.civicforums.com&youtube_vi deo_id=qPGU3nguQR4[/autostream]

Also, while driving and then coasting with the clutch engaged, the idle will seek 2000 rpm and will not return to normal unless the car comes to a complete stop at which point it immediately settles down to normal unless of course I blip the throttle again.

So it seems like the ECU knows knows the car is in motion and is for some reason commanding the Idle Air Valve to hold 2000 rpm until the car is stationary. This is completely baffling me.

I wish that thread had a disclaimer warning that it can potentially seriously mess up the idle. I really wish I hadn't screwed with the throttle body cleaning to begin with.

Any help would be appreciated, is there any test for the Idle Air valve? There are no trouble codes at all.

lazlong

this

scooty

Already did the re-learn, twice now. It did not help much.

scooty

Lol, nobody knows?

Update: The ECU seems to have gone absolutely batshit crazy. The idle fluctuates between 1000 and 2000 rpm very rhythmically while coasting in neutral. It did this for an entire mile while coasting down a long hill. As soon as the car comes to a stop the idle settles down perfectly to normal.

Also the ECU appears to be commanding the idle to fluctuate even when the car is in gear! I had the car crawling along in first gear with the clutch out and not using the gas. The car would speed up by itself to 2000 rpm and then slow back down to 1000 rpm, repeatedly over and over again!

You would think if this were a vacuum leak the fluctuation would be a little more random, but its very predictable.

No engine codes so far and I did the idle re-learn again (three times now). I'm running out of ideas

ezone

The action of surging is a typical symptom of a vacuum leak (or incorrect base idle speed setting). I explain this in post #2 of this thread:

https://www.civicforums.com/forums/2...e-problem.html


https://www.civicforums.com/forums/3...ting-idle.html

Read through both of those threads. Ignore the codes, but read the information. Maybe you can gain some ideas here.


Do you know HOW to "look for" vacuum leaks?

mikey1

not really.....

it still does sound like an intake/vaccuum leak to me,

could be an air leak between the intake manifold and the head

scooty

I hooked up a vacuum gauge to the port at the back of the intake manifold where the evap canister purge vacuum line normally connects to.

I'm getting a very steady 20.5" of vacuum at idle. Does this eliminate the possibility of a vacuum leak?

I also started up the motor with the IACV valve un-plugged. The idle was higher than normal, about 1300 rpm, but it did not hunt up and down between 1000 and 2000, so this tells me the ECU is commanding that behavior from the IACV valve. Although with the valve unplugged, the idle exhibited a rapid up down fluctuation between a vary tiny range of 1200-1300 rpm every once in a while before settling steady at 1300.

How could the intake manifold start leaking when I haven't ever disturbed its connection to the head. Is this common issue with D17s?

ezone

No.
The design of the injection system is such that a vacuum leak is compensated for by the computer adding additional fuel.

You will not find a vacuum leak this way.

scooty

What's the best way to test for a vacuum leak then? I don't like spraying carb cleaner all over the place as the solvents can damage electrical components.

ezone

That's one of THE BEST ways.
Use controlled spraying though.

Alternatives:
Brake cleaner. (Must be flammable type. Brake cleaner evaporates completely leaving no residue to harm later on.)
Propane.
Alcohol.
Sometimes I can only see a reaction on a scanner datalist if a change is minute.
Got a scanner that gives an engine datalist?

And it doesn't take much of a vacuum leak to make the idle surge.


Other options:
Smoke.
This guy is a nut, but this can be pretty effective:

Ultrasonic listening device (bionic ears).....
Or a length of ordinary tubing held to your ear, used as a stethoscope.



None of this will find INTERNAL leaks such as PCV or IAC or any other means of air uncontrollably bypassing the throttle plate.

Did you read through the links in my other post?

scooty

Yes I did. I guess I will use brake cleaner and spray around looking for vacuum leaks. Places to check that I can think of:

1) Brake booster hose, all the way to the booster
2) Evap purge canister hose
3) PCV hose
4) Cruise control vacuum line
5) Base of throttle body to intake joint
6) Intake manifold to head joint. Really not looking forward to jacking up the car and crawling underneath a running eninge to get to the bottom of the intake manifold.

Am I missing any other place to look? Does brake cleaner mess up paint?

ezone

I use it to clean grease off of factory painted panels sometimes.
It can strip wax off but normally doesn't hurt the paint.

You shouldn't be spraying near important exterior paint though?
Test a small area first?




Carb and throttle cleaner WILL ruin paint.

You can pinch most of those hoses off with pliers and listen for any change.
You probably don't need to crawl under the car.

scooty

Update: It is coooold outside!!

I can comfortably say I don't have a vacuum leak. I sprayed down all the areas I mentioned above with brake cleaner and noticed no change in idle. Spraying in the throttle body killed the motor immediately as expected. Oh and I could not find a cruise control vacuum line so I'm assuming its electronic.

I also plugged the IAC valve port in the throttle body with my fingers and this killed the motor immediately so I guess that eliminated the possibility of a massive vacuum leak.

So I guess this leaves only three things:

1) Bleed the cooling system some more since I lost and replaced some coolant when I removed the throttle body

2) Replace the IAC valve.

3) Bad throttle position sensor so buy a used throttle body from Ebay?

I don't have any CEL light indicated. Is it still worthwhile to get codes scanned? I'm assuming there will be a bunch of codes stored from when I ran the engine with the IAC valve disconnected.

ezone

Air bubble is not the problem.
The IAC is not controlled by coolant temp directly.
Only the PCM controls it based on its temp sensor reading.
EDIT: There are more criteria the PCM uses to control the IAC, but its temperature input comes from the computers' coolant temperature sensor.

Tap on the IAC valve with something like the handle of a screwdriver.

It could just be stuck open (right now it sounds like this is the most likely thing possible since blocking its passage killed the engine).

IF tapping brings the idle speed down then either clean the IAC or replace.

(The exact same goo that builds up around the throttle blade collects in the IAC and makes it stick. You disturbed the goo by cleaning earlier.)

Testing is MUCH cheaper if it will keep you from wasting money.



More:
You said:
I also started up the motor with the IACV valve un-plugged. The idle was higher than normal, about 1300 rpm, but it did not hunt up and down between 1000 and 2000, so this tells me the ECU is commanding that behavior from the IACV valve.

The PCM was not controlling the IAC 100% even when it was plugged in. It was TRYING to, but couldn't.
It THINKS it was though. EDIT: It could open the passage further, but couldn't close it.
Then the PCM was controlling the fuel delivery--- causing the surge.

I'm pretty sure I explained how this works in one of those other threads I linked earlier.

ezone

Again: It really sounds to me like your IAC is stuck open and can't close its passage.

mikey1

have you ever done a compression test?

this is a long shot but it still could be a head gasket or even a mechanical failure of some sort, like a leaking or cracked valve

clinton_hedrick

Mine had a similar problem, but I threw a code, it was 4*** something i think (Air intake system leak). I couldnt find any vacuum leaks, cleaned the throttle body, and reset my ecu, but the light just kept going on and off. Sometimes it would idle rough, sometimes it did just fine. It eventually went away on its own, and my car has been running fine for 6 months now. It had me stumped too.

scooty

So i tapped on the IAC valve with a wrench while the idle was doing its up and down thing and it made no difference.

I'm not sure why the valve is stuck if it is. While cleaning it, I separated the valve from the plastic portion that houses the connector, and cleaned the rotating valve assembly thoroughly with carb cleaner until it was completely shiny and spun freely.

The plastic part appears to consist of just the connector and two round magnets that I'm assuming form a field to rotate the valve into the correct position as per signal voltage from the PCM . I really don't understand how these go bad in the first place.

ezone

Oh. Now I think I got what happened.

*facepalm*
Did you have to remove 2 funny shaped (Torx?) screws to do this?
You weren't supposed to do that.

(random IAC pic from Google)

You pulled the field coil off and you then had a bare, naked valve like this:



There are slots under those screws for minor adjustment of the coil position, yes?
Loosen the 2 torx screws and rotate the field coil one way or another and see what happens to the idle.
If no improvement, rotate it the other way and recheck how it idles.

scooty

Yes this is exactly what i did. I did notice the slotted adjustment and have already tried it at both ends of the adjustment, it only makes a difference of a couple of hundred RPM but the idle still bounces wildly while coasting in neutral or blipping the throttle while stationary. I was having trouble with the idle before I ever took it apart, here is a brief timeline of what happened:

1) Everything is running fine, no problems what so ever, I read the forgotten tune up mods thread

2) Remove, clean and reinstall throttle body and IAC valve.

3) The idle is now messed up, it sticks at 1500 for a few seconds before returning to normal while blipping the throttle. It sticks at 1500 while coasting in neutral but returns to normal while stopped. Note that there is no hunting at this point

4) Couple thousand miles later I remove throttle body again, remove field coil assembly from IAC valve, clean everything really well and reassemble.

5) Now, I'm here. The idle hunts endlessly between 1000-2000 rpm while coasting in neutral but returns to normal while stopped. While blipping the throttle after its stopped, the RPMs will hunt between 1000-2000 RPM once, but then will return to normal.

ezone

Um, so where does this line fall into your list that started with 1) Everything is running fine, no problems what so ever,???

Man, I have so few issues with these, I rarely ever need to chase this kind of stuff on 7th gens. Older cars could be hell though.


Ok. I guess you need to unscrew everything you did.

Adjust the IAC doodad to give the lowest idle speed it can possibly give. (Is it physically possible to have gotten the coil on 180* off? Or the shaft of the valve rotated upside down? I'm guessing at anydamnthing now.)


Did you take apart ANYTHING else on the throttle body? You didn't take the blade off the shaft, or loosen the screws, did you?
Did you mess with the throttle stop screw (minimum air rate)? (That is preset and never to be tampered with!)
Is the throttle cable adjusted too tight?
Can you twist the throttle (gently force it) even more in the direction of "closed" and have the idle speed drop?

Is the TPS voltage at 0.49V with the throttle at rest?

New gaskets for the IAC and the throttle body? Several tiny vacuum leaks can add up to one big one, in the eyes of the computer.


Did I read that this engine has some aftermarket intake manifold on it, or was that some other thread?



From the Honda site:
Diagnostic Procedure:

1. Do the ECM/PCM idle learn procedure. (This won't do any good if there are real underlying problems that haven't been resolved.)
   
2. Check the idle speed.
3. Inspect/adjust the throttle cable.
4. Test the throttle body.

Also check for:

- Intake air leaks

Idle speed spec:
700 +/-50 rpm (USA)
Still sounds like the base engine speed is far above this spec., for whatever reason.....The PCM can only use the IAC to control the engine higher or lower than that base speed, and that action depends on everything else being "right". The PCM should be trying to lower the engine speed to hit its target of 700.
If you have it surging from 2000 to 1200 and back repeatedly, that must have something pretty far off to cause such an increase over the correct base speed....and yet still have the TPS sitting at CTP (closed throttle position voltage). This is almost always caused by a vacuum leak of some sort.

They can't take into account all the possible things that humans can do to a car though.

Did you mess with anything else? (I have not read the forgotten tune up thread.)

scooty

What I meant to say is I was having trouble with the idle before I disassembled the IAC valve apart. It goes like this, 1) No problems, 2) Removed and cleaned throttle body and IAC valve but did not disassemble the valve. 3) Problems with idle. 4) Removed throttle body again and this time disassembled IAC valve. 5) Even worse idle problems.

No this is not possible. The way the housing fits its not possible to have the valve in the wrong orientation.

No I did not.

No I never touched that screw.

No I already checked for this last night.

I will check this and report back.

I reused the old ones. They were soft, supple clean and looked pretty good. I checked for leaks around this area with brake cleaner, nothing.

Yes it does, It is a Weapon-R short CAI.

OK i think you misunderstood. The engine holds base idle speed very well at about 700 rpm. There are only two (2) scenarios under which the surging occurs:

1) Coasting in neutral. The surging stops and idle drops to base at 700 RPM as soon as the car stop moving.

2) When the car is stationary and the idle is steady at 700 rpm if the throttle is blipped past 3000 rpm, on its way back down to 700 rpm the idle will drop to 1000 rpm, surge back up to 2000 rpm and then settle at 700. This behavior can be seen in the YT video link i posted earlier.


No I didn't, I only removed the throttle body and the IAC valve. None of the other sensors were touched.

ezone

Ok, so it is just a CAI, not an actual manifold. No problem.

I guess I did misunderstand it then.

But I still think the base speed is too high. The PCM can pull the speed down to a normal range using the IAC valve, but it is likely near the very end of its controllable range.

I haven't seen a manual trans 7th gen car in weeks. I can't easily compare.....

But I will say a slightly elevated idle speed is probably programmed into the PCM as long as the PCM sees a VSS input greater than "0"....(I expect it on other makes, so it wouldn't be unusual here either) BUT it should not surge while it does this.

This brings me to ask you to drive it and check 2 different ways:
Does it surge if you push the clutch in to coast (shift lever still in a gear)?
Does it surge if you pop the shifter into neutral without touching the clutch?I saw it. A brief "hang time" of a couple hundred RPM for a second or two might be expected, but the high speed and surge is not.
On some other makes this would indicate to me that the base idle speed is too high. (I actually used the "hang time" to set base idle speed on the late 80s-mid 90s Mazda cars by ear, since you can change adjustments by a long way without seeing any change in idle RPM speed just sitting still, but blipping the throttle told a lot more about what was going on with them.)

Again, I don't have a real similar car to compare with.
I'm just going from an addled memory of 25+ years of various cars kinda blended together. I've forgotten more than most will ever know.




Did you disconnect the battery for several minutes or do anything to kill the PCM memory before you tried the idle relearn?

When you cleaned the throttle body, you changed how much air can get around the closed throttle plate. Now the PCM might still be compensating for how much air flow there USED to be before it got cleaned. PCM needs to forget what it knew before and learn something new.





I'm running out of oddball ideas. I still think there is probably something simple that has been overlooked.

I'd want to see what is going on in a datalist on one of my scanners.
I might come up with something completely different if I had the car in front of me.


It's too late to tell you "If it ain't broke, then don't f*** with it."

How about "If it ain't broke, keep on fixing it until it IS broke!"

HTH

scooty

I think I found the problem! It's the TPS sensor. I back probed the middle wire and got 0.48 V at idle and 4.69 V at full throttle. Normal voltage I believe is 0.45 to 4.5, so I'm 0.03 V off at idle and .19 V off at full throttle.

Also I'm not sure if this is normal or not but with the key in the II ('On) position I can hear a slight electrical buzz/humming noise from the IAC valve. I know it's the IAC because the noise goes away as soon as I unplug the 3P connector to the valve.

Ezone to answer the rest of you questions:

Yes to both questions. Also driving along in 1st under 2000 rpm the car is very jerky and I can feel the PCM try to modulate the throttle. A lot jerkier than what is normal for a manual car in 1st gear.

Yes I did. I pulled the PCM/FI fuse and I also disconnected the battery for about a minute.

LOL, I've been know to be guilty of both.

ezone

Nope, that isn't it. Maybe.

0.49v is perfect at closed throttle.
WOT voltage is fine too.
Don't touch it.
Yet.

I start worrying if I see one that ISN'T exactly .49v. Very rare to see any other voltage with closed throttle.
But if you can snap the throttle a few times and get a DIFFERENT reading every time at closed throttle, something may be going on. Maybe a flaky sensor, but that's pretty rare.

I have heard of carb sprays causing damage to TPS but never actually witnessed it.
But it might explain what has been going on.
Noise from the IAC is probably normal. It is an oscillating motor and is operating whenever the PCM is powered up.

Back to the TPS. Attach voltmeter again. Slowly begin to open the throttle by hand. Voltage should slowly and steadily increase as you open the throttle.

I really want to know what is going on in the datalist while this is going on. What is the TPS reading as seen by the computer while this is surging?
(Sorry, I seriously depend on being able to see what the PCM sees in order to figure out what is going on when I have to diag something. A capable scanner is my best friend.)