Idle control valve

Just as a follow up Dave and I met earlier and managed to confirm the issue. Centre pin is indeed battery voltage. Two outer pins are the feeds from the ECU and the info I posted earlier re 3 pin ICVs is incorrect. The two outer pins should get a corresponding voltage to rotate the valve open and closed.

I'm getting 4 to 8 volts on one pin but nothing on the other so it can't re-open. Voltages on Dave's were similar 7-9 at cold but 3-4 on the opposing pin.

Wires are all checked so I'm guessing an issue with the ECU outputs for that pin.

Thanks for the loan of the Sykes Pickavent reader Dave. Unfortunately it can't talk to my ECU - probably the result of the modifications....

Darren ....... You have not told us what the Black meter lead is attached to ..... and what voltage to earth is on the center pin of your connector.
Unfortunately I am going away for the weekend as of tomorrow, and there is a lot to discuss after these readings, and I have not got time to correspond before I go. But I cannot see that you could get any reasonable reading with the connector disconnected. as the coils are not loading your readings and therefore your readings are quite spurious and would not be there if there was the load of the coils connected.
The ECU will deliver an earth to either of the outer pins depending on the engine state. By measuring the outer pins off load, you are just getting a leakage of volts from the solid state switching transistors which have no power in them, but as the meter effectively has no load you cannot rely on those voltage readings.
Just a quick test, if you like ....... Remove the idle valve from the engine ..... from a 12v battery, connect the positive to the center pin, and the negative to each of the other pins in turn (not both together ), and you should hear the mechanical internals click each time you change the negative from one pin to the other.
John

.

Black was just to earth (chassis). I take your point but these aren't a fixed voltage as such, they're a high speed signal so it's just an approximation.

What I was looking for was any reading on the 3rd pin and whereas mine shows zero (or very low voltage at best), Dave's is in the 4v range.

That figures with what I'm seeing in that the ECU can close the valve, but can't open it again. Using a 9v battery I can flick the solenoid one way or the other so I know that it works.

[simontaylor]

Did you swap the ecu between the 2 cars?
Then you might find out if the fault is loom related or not.
Have you checked the continuity for the ICV to the ECU?

Black was just to earth (chassis).

What the ECU does, when required, is to earth either pins 1 or 3....... with reference to pin 3 which is 12v.
If you put your DVM across pin 1 or 3 and earth, what you are reading is a spurious voltage between two earths.
Did you have 12v on pin 2 whilst you were taking your readings ?

Using a 9v battery I can flick the solenoid one way or the other so I know that it works.

That's good ....... but don't know what the mechanism is doing. If it is blocking an airflow in one state,..... is it in fact doing so ? Is there a possibility that the mechanism is impeded in any way so that there is a leakage past the closed valve. (if that is what the valve does)

I've rewired pins 1 and 3 so there should be continuity there to the ECU.

Pin2 is 12v + at all times so doesn't change.

Swapping ECUs over would be the best test but is reliant on getting an MOT first so I can move mine.

Both should run on the opposing ECU but wouldn't run well due to my larger injectors.

Mechanism is good. It's quite simple when you look at it.

I have just joined USA Renix forum...one of the FAQ's says "don't use a digital multi meter on these systems as the high speed switching gives you misleading results"

As I think this is PWM (Pulse width modulated) ...I can see that this could be an issue with the idle control valve signals- the meter is probably reading peak signals when you need an RMS value- do you have access to an old fashioned multi meter that you could try...combined with Johns comments about measuring across the valve rather than with it disconnected?

I should say I'm a mechanical engineers by training so electrics is not my forte, but there is something odd afoot here.

Yes it's not ideal - osilloscope or similar would be better but the switching is in the msec range and it at least gives a broad indication in my view.

Essentially I have two pins that read broadly the same as Dave's car and one pin that reads nothing, so like for like it appears to be that pin that is causing the problem - which seems to tally with the results I'm seeing at the valve. Just my working theory at this stage though.

If I can get the MOT done then I can make better comparisons I reckon.

[simontaylor]

I see your point Darren, that you are crudely comparing like with like and getting different results. Lets hope that is not causing you to make any incorrect assumptions when fault finding.
Maybe MFaulks can test your ECU for ICV operations?
My local MOT testers station is very sympathetic to classic, race and modified cars. I am sure if you make an appointment to suit you when the car is warn....your local guys will help you pass, just might need to book it for a couple of weeks in advance.

I think I'll be ok Simon with the idle set manually. It's not ideal but I think it will do. At least it's summer so cold idle is easier than winter.

Yes broad assumption re the ICV but there's very little good info out there on the internet and most I've found is incorrect.

Reading Dave's engine manual there are indeed only two dynamic inputs to the ICV (both via the ECU) - temp (cold 1100rpm idle - hot 850ish rpm idle) and revs. So it only functions on those two inputs - revs being the primary which it tries to hold within range as per the temp.

There's no feedback from the ICV itself - it just opens and closes based on the two feeds from the ECU at pins 1 and 3. Those energise two coils to open and close. From what I can tell from reading the manual, the ECU plays a constant game of tug-of-war via the voltage signal. It oscillates both inputs very quickly but essentially the two outer pins always see some voltage in a steady idle state.

From what I can see I'm missing one....