Car: 1988 Honda CRX Si
Engine: JDM 1st-gen B16A

It happened after I did a valve adjustment. After I started it up it idled kinda low and then when I tried to drive it it stalled a few times. But once it was allowed to warm up it idled decently, though a little rough, except that the rpm's would hang after shifting momentarily and the idle would bounce up and down several times before settling in around 750. Under cruising the car ran perfectly smooth, no misses or anything.

I did another valve adjustment today and the same thing is happening... WTF is wrong? I've done several valve adjustments before on this car and others, I know what I'm doing and I'm using a Snap-On tool for it...

Oh, another point: At idle, the air/fuel reads in the 'full rich' zone about 3 LEDs from the end.

 

check your vacuume lines. maybe one is not connected

just a thought

 

They're all connected but I hear a hissing from the IM somewhere... gonna pull out a spray bottle and check that later.

I just replaced the IACV with a known-good one, and while it was cold the car ran great. Idled solidly at 1500rpm, you could rev it and it would stick at 1500 after revving (wouldn't happen before).

When it got a little warmer, tapping the gas resulted in a stall. Starting it up and letting the idle reach 750 rpm (it would bounce wildly between ~250rpm and 1000rpm) it would then sit comfortably at about 750 rpm even after revving.

THIS IS REALLY PISSING ME OFF!

Oh yeah, the fast idle valve stays open too, even after the car's warm, but unplugging it makes almost no difference

Edit: I made a small movie of how the air/fuel sits at rich - watch the gauge on the left

2.2MB, no sound


[Modified by raene, 3:17 PM 8/5/2002]

 

C'mon... anybody?!? I can't drive it if it always wants to stall on me

I'm starting to think it might be the idle speed - on my engine, unplugging the IACV/EACV (whatever it's called) causes the engine to die - is this normal?!? The Haynes manual for CRXs says the idle should drop but not die...

 

How tight did you do the adjustments? intake/exhaust

 

I did 0.006 on the intake and 0.008 on the exhaust, since I don't have a 0.007

*update* I set the idle as per Haynes: Warm the car up, unplug the IACV so it dies, start it up and hold the throttle at 1000rpm for a few seconds, then let off the gas so it idles itself with the CEL on. Then adjust the idle so it's around 500rpm. Well, my frickin' idle kept bouncing around between 500 and 750 rpm! It doesn't stay steady!

So I played with the idle for about an hour, resetting the ECU, adjusting the idle, etc... it just keeps doing the same thing over and over. Start it up, it revs to 1500, drops down to 750 and past, bounces around and almost stalls until you give it gas for a good 2 seconds, holding the throttle steady. Then it idles rock solid. But at some idle adjustments the RPM hangs after you let off the gas.

WTF?????


[Modified by raene, 1:33 PM 8/6/2002]

 

Are you sure you didn't adjust your valves too tight? Did you adjust them with the engine cold?

 

yeh he did... the valves are fine. He's quite competent at that kind of stuff.

 

I'm going to pick up a 0.007" feeler and triple-check it in the next few days... argh

 

 

C'mon guys... I've done this before several times... both times I did the adjustment I left the hood up and the car sitting overnight before doing the adjustment

Just noticed that holding the RPMs at a steady point results in some wavering of the needle as the air/fuel ratio bounces around in stoich. Could this all be a problem with the O2 sensor?!?


[Modified by raene, 10:14 PM 8/6/2002]

 

Friend of mine dropped a JDM B16A in his 92 hatch (Used to have a D16A).
He retained the stock wiring harness for his D16, and is using the D16 ECM.

His idle will act the same way yours does. Upon startup, it will wildly bounce between a "near stall" and 1000+ RPMs.

The person that helped him install his engine did not connect either the Fast Idle Thermo Valve, or the IAC to a coolant line. He is using the IAC from the B16A.

I hope to clear up hi problem after correctly routing the hoses.

I wanted to ask you if you used a multimeter to discearn the (difference) in signal sent to the ECM from both the JDM IACV, and the D16 IACV. I have not done this yet, I was hoping it was something you had already tried.

A side note: his hoses on the JDM engine were all plugged into "odd" ports for shipping. You may want verify the flow through this system using some clear tubing. I'm going to do that today to make sure we connect the correct inlets and exits.

Good Day.

 

can you please explain me how u changed youe IACV? how much is a new 1 from a dealer? i thnk i need to change mine too

 

check your vaccum line to your map sensor also check for your TB gasket..

 

Your friend is using the D16A6 computer?!? Ouch, that's gotta suck. I'm using PR3 M/T ECU at the moment.

I'm not sure how to connect my vacuum lines - ie which goes where. The '88 CRX Si I installed the engine into has two fast idle valves on the firewall whereas every other CRX I've seen only has one. But unplugging the vacuum lines makes no difference to the idle anyway, so who knows what they're really for

I checked the differences between the JDM IACV and the USDM D16A6 IACV according to my Haynes manual and couldn't find any problems. But last night the ECU threw a code 14 (IACV). After resetting it doesn't seem to have a problem but I'm going to investigate some more today.

LxAccord97, to change your IACV first disconnect the electrical connector. Then unbolt the IACV from the intake manifold (two bolts), then unclip the two coolant hoses and remove them from the bottom (easier to do when the IACV is unbolted from the intake manifold). Then install your new IACV. It's really easy, takes ~5 minutes.

pg_thao, I'll try a replacement MAP today - I've got two extras sitting in the garage.

 

I have heard that OBDII cars use three wires for the IACV, My friend's D series P28 uses 2.

His JDM B16 uses a FITV below the TB, and an IAC behind the IM. I connected each of these to coolant flow, but his idle still jumps between 1000+ RPM and about 400 RPM. His old D16 did not have a FITV, I am still considering disbling this valve.

Here's what we did for testing:
(Assuming B16(A1?) and D16 P28 ECM, with a turbo installed. I am unsure of his O2 sensor setup right now. )

TEST:
B16 IAC and FIT valves connected to coolant flow.
RESULT:
Idle still fluctiuates, aka "hunting Idle". (The ECM is hunting for the correct control over the IAC to correct idle.)

TEST:
Unclipped the electrical connection for his IACV.
RESULT:
The idle still fluctuated, but more rapidly, and to lesser extremes. (I would guess the only reason it didn't jump quite so high/low is because of the rapid rate at wich it fluctuated.)

TEST:
Replaced linear B16 IAC with coil-type D16 IAC. Bolt pattern did not match perfectly, only one bolt went in, but the ports fit well for testing. Connected the ECM to the IACV.
RESULT:
The Idle still demonstrates the "hunting idle" problem. Jumps up and down about every 1.5 seconds between 1000+ and 400 RPMs.

TEST:
Disconnected the ECM from the D16 IAC valve.
RESULT:
Same behavior as disconnecting the ECM from the B16 IAC valve. This time we let it sit for about 30 - 45 seconds until the engine died.

Where this leaves us:
I am still considering that wiring alterations to the Idle sensor/ECM may be a cure.
I also believe a series of tests disabling/altering the Fast Idle Thermo switch could be worthwhile, because his D-series engine did not have one.
As stated before, I am unsure where his O2 sensors are, or how many his P28 is supposed to use (1 or 2). I know there are NONE installed in his turbo exhaust manifold. Having a secondary O2 sensor installed where the primary (only) sensor should be, could be a cause of idle problems. (I say this because he is running rich at most loads in lower RPMS. Idle included.)

Happy troubleshooting.

 

Here's an idea I came up with over luch.

A simple way to test (disable) the Fast Idle Thermo Valve, is to:
1) disconnect the intake arm from the TB.
2) You will see two holes on the firewall side of the TB just prior to the Throttle plate. The lower one feeds the FIV. The upper one feeds the IACV.
3) Thouroughly clean AND dry around these holes prior to the plate.
4) Use some duct tape and tape over the lower hole that goes to the FIV. Be sure and use enough to get a good purchase, and don't let the tape touch the throttle plate.

I have not tried this method yet, so I'd like to disclaim any indications that this is an infallable testing method. Potentially the duct tape could come loose and be injested causing engine damage. I would hate for that to happen to someone.
____________________


Another thing I noticed about my friend's (turbo remember) throttle body, is that it was lined with a significant quantity of oil. I suspect that's from turbo blow-by?
I'd like to test, and then possibly clean his fast idle thermo valve, because it could have a quantity of oil in it's airways, contributing to this "hunting idle" problem.

If that is the case, I could easily see TB cleaner collected in the FITV causing the same problem on NA engines.

I'll test these ideas this week, and let you know.

Good Day.


[Modified by WOTTEG, 1:27 PM 8/12/2002]