Troubleshooting Idle Control Valve (high idle)

[jpwerks]

Worn throttle body
take out and check up and down play
_________________
81' single cab 1.8T Syncro (converted)
85 Vanagon 1.8T
55 chevy
2008 330i BMW
2000 jetta 1.8T
2004 Golf 2.0

[dlalbert58]

Problem: Idle after warm up is about 2000 rpm.
Troubleshooting attempts:
1. Checked throttle position switch. It did not always switch at idle, worked OK at wide open throttle. Readjusted per Bentley and now switches at both idle and WOT. This did not help with the problem.

2. Checked idle control valve. Vibrated when power was turned on. Removed and cleaned with carb cleaner, solvent really dirty, but now good and clean. Problem remains.

3. Tried unplugging the ICV connector while running. Idle drops back to under 1000 rpm and car drives well. Checking function of ICV indicates that it is no longer adjusting idle speed for A/C and power steering loads. After shutting off and restarting, high idle returns. ICV now responds to A/C and PS loads. Tried unplugging ICV with ignition on but engine not running, ICV vibrates initially, but after disconnect and reconnect no longer vibrates. Apparently, once power is removed from the ICV, it does not recover without restarting the engine. Not sure if this is by design or not.

Is this an ICU problem? Anything else I should check first?

[borninabus]

if your throttle body is original, it is probably worn out.

check the throttle shaft for play in situ by grabbing the vertical spring portion and moving it axially: front/back, left right. when they are really worn the play can easily be seen. keep in mind there should be NONE so even the tiniest amount of movement is bad.

i used to run my idle screw about 1/4-1/2 turn from closed to compensate and it seemed to keep my ICV in check. any more turns out and the symptoms you describe would result.
_________________
88 Van WBX, A/T - 13 JSW TDI 6M/T - 2012 Touareg TDI Sport

[dlalbert58]

I did more troubleshooting as detailed in the Digifant Training manual. Checked continuity of all the cabling from the ISCU to the various components - all checked out OK. Measured 5V at the TPS, ICV resistance measured 4 ohms and the resistance of the temperature sensor was 1200 ohms (the engine was warm to the touch, so this is a good value)

. Couldn't find anything out of spec. I'm thinking I have to look for vacuum leaks.
Of course, after reconnecting everything it is now idling as it should. I am assuming (hoping) that had a dirty contact somewhere that was fixed by the unplugging/plugging of everything. Time will tell.
I did find one thing that was odd. On the ICU harness, right behind the connector, someone installed a 1900 ohm resistor and jumper wire between the hall effect lead and the ground lead. It doesn't appear to cause any problems and must have been installed many years ago. Anyone have an idea why this would be done?

[RawUmber]

What year is your Vanagon?

VDO's "D" subcode units will disable the Idle Stabilization Valve (ISV) until restart if it experiences an over/under load. This is probably what you are witnessing when you replug the ISV.

When functioning, VDO's idle system can only add air. Even at its lowest duty cycle. If your van has a fast idle with the ISV disconnected from the harness, it will idle even faster when the ISV is connected/running due to the system adding its minimum quantity of air. An approach to resolving that is to lower your base idle so that the idle system can positively affect the idle with its supplemental air.

The resistor from hall sensor output to ground is curious. I can't imagine a spec hall sender needing it. Your harness also has a chopped smaller green wire. Like the resistor, I don't know if this is a PO mod or an engineering change. Maybe one of the forum's harness gurus has an idea. If the wire carries any voltage, it's probably best to tape/shrink it to prevent a short.

[dlalbert58]

The ISCU is a D version, so that explains the disabling of the valve after disconnect.
The idle speed with the valve disconnected is under 800 rpm, so running a bit slow. With everything working correctly, is is about 900.
I checked the cut green wire and it is not connected to anything (ground or otherwise), just floating. The "good" green wire from the hall sensor appears to be a newer wire. Maybe the original, clipped wire had broken or was damaged and was replaced with this newer wire. Whoever did the work used the proper contacts and crimp tools, it wasn't a hack job.

[RawUmber]

The size shift on the green wire at the splice makes me wonder if the larger run is coaxial, with the resistor to ground being tied to the coaxial shield for mitigating noise.

Your van's year might help someone w/ harness knowledge narrow down what's going on there.

[tencentlife]

That doesn't look like any Vanagon ISCU connector I've ever seen, and RawUmber's right that the fat green wire sure looks like the OXS coax, with a tie-in to ground the shielding.

The O2 is shielded in the normal wire harness because it runs for a few feet in the same bundle and parallel to the pulsing injector power wires, which do have measureable current and would induce noise in the very-low power O2 signal core sufficient to foul the signal.

I'm guessing the clipped green wire is the rpm signal, ohm that back to the coil to find out.

The whole thing looks like someone reengineered the ISCU connections, it has the same basic tandem DIN relay layout. If the wire colors match the diagram's then someone just installed them in a different connector body, and maybe they changed to a shielded wire thinking the rpm signal was being washed out. The ISV power is pulsing at frequency in the same bundle, so maybe not that crazy an idea. Couldn't hurt anyway.

I don't have confirmation of this occurring, but credible people have warned not to disconnect the ICV while it's running due to the chance of damaging the driver Tx. When you want to test its effect on idle speed (and as stated, it can only add intake air and increase speed), just pinch shut the ribbed rubber J-tube, no chance of influencing the control electronics that way.
_________________
Shop for unique and useful Vanagon accessories at the Vanistan shop:
https://intrepidoverland.com/vanistan/
also available at VanCafe.com!

Please don't PM here, I will not reply.

Experience is death to doctrine.

[dlalbert58]

The van is 1986. I'm the original owner but after 38 years, I can't possibly recall every service performed on it. Originally it was serviced by VW dealers. After the dealers became unwilling to work on the car, I've had independent VW shops work on it. Now that I am retired, I am doing most of the smaller jobs myself, leaving more complex stuff to the repair shop.
I did check the clipped green wire and it is not connected to the distributor connector. It is not going to ground either, seems to be floating. The new green wire has continuity to the distributor connector but the green wire at the distributor is original, so there must be a splice somewhere. I'll have to trace the wire between ISCU and distributor to find where that splice is.
I should also check the O2 cable to make sure that is OK and not been co-opted somehow. The sensor needs replacing anyways, it has been there for a very long time.
Edit: I rechecked the wiring diagram and it shows that the green wire from the ISCU is connected to the distributor green wire at the ECU connector contact. First place I'll look is the ECU connector to see what the harness looks like there.

[crazyvwvanman]

I have seen that wire in COAX before, at the ICU socket.
And a quick check in the gallery came up with a photo by UberAudi I think it was.

My suspicion is that early 86 difference is involved.

Mark

[dlalbert58]

It is a relief to find out that this harness is not an unknown variant. crazyvwvanman's post led me to https://www.thesamba.com/vw/forum/viewtopic.php?t=301674&highlight=.
I verified good continuity from the ECU connector to the hall sensor and O2 sensor. I did not physically trace any wires as everything is inside gray plastic sheathing. I do have the coax splice shown in the referenced thread, although I believe this is between the ISCU and hall sensor, not the ECU and O2 sensor.
The van is running fine now, I suspect there was a dirty contact and all my unplugging and plugging connections corrected it.

[4Gears4Tires]

Uh, I don't think you have just one dirty contact. I think everything in your engine bay is dirty. Also, it appears your transmission bolt is loose.

Not a bad idea to clean everything and double check everything in your engine bay.
_________________
'87 Syncro Ferric Oxyhydroxide Superleggera Edition
'85 Westy Sciuridae Domus Edition

[crazyvwvanman]

Just a follow up to the earlier discusion of a coax wire seen at the ICU wiring socket in a few early Digifant.
Today I followed a link to an early Vanagon Injection systems Pro Training document.
The Digifant section included some diagrams.
Below is Page 131 from that document.
It clearly shows a coax wire to pin 17 of the ICU.
The coax is missing from the Bentley early 86 diagram but it should have been included.
(Page 130 shows why the coax was thought to be needed that early wiring version.)

[RawUmber]

Good catch. Nice to see that coaxial run to the ISCU documented somewhere. Notable that the ISCU was getting engine speed from the coil rather than the hall sender.

Another followup... by way of PM, dlalbert58 clarified that he had measured 1900 ohm of resistance between the hall effect coil signal lead and the ground lead, rather than seeing an actual resistor. So no resistor. The harness is wired like the above diagram: ground connects directly to coax shield.