I'm going to solve the misfire and cold running issue

[soul4t]

I read an interesting bit on the wiki page about the ICV and thought I'd test it out.

I turned the car on with the ICV disconnected, it started running much better and held idle above 1000rpm's...more like what it should run like, no splutter. From what I understand, when the ICV is disconnected, it's valve is open a little way and lets a stable amount of air into the engine.

When I reconnected it, the car ran as it used to.

I then tested the voltage at the connector to the ICV...it showed just under the battery running voltage...13.96v From what I understand, the ECU should stop using the ICV when the TPS indicates that the throttle plate is open. I know that my TPS works...it's almost new....and it's set properly. When I tested the voltage and opened the throttle plate, expecting the voltage to drop. It remained the same. Is this correct?

So, to me, as far as I know, the voltage to the ICV closes the valve completely. If the valve was closed completely, wouldn't the car stall? I know, the throttle plate lets a very small amount of air into the engine, but surely thats not enough to keep the engine running?

Any ideas? I'd appreciate any input!

Cheers!

[soul4t]

Oh Yea I almost forgot, I tested that the ECU knows when the ICV isn't connected...it logged a fault code (04). Suffice to say, I've cleared the code and re-checked.

[Billwill]

I read an interesting bit on the wiki page about the ICV and thought I'd test it out.

I turned the car on with the ICV disconnected, it started running much better and held idle above 1000rpm's...more like what it should run like, no splutter. From what I understand, when the ICV is disconnected, it's valve is open a little way and lets a stable amount of air into the engine.

When I reconnected it, the car ran as it used to.

I then tested the voltage at the connector to the ICV...it showed just under the battery running voltage...13.96v From what I understand, the ECU should stop using the ICV when the TPS indicates that the throttle plate is open. I know that my TPS works...it's almost new....and it's set properly. When I tested the voltage and opened the throttle plate, expecting the voltage to drop. It remained the same. Is this correct?

So, to me, as far as I know, the voltage to the ICV closes the valve completely. If the valve was closed completely, wouldn't the car stall? I know, the throttle plate lets a very small amount of air into the engine, but surely thats not enough to keep the engine running?

Any ideas? I'd appreciate any input!

Cheers!

It sounds like the ICV is getting the wrong voltage from the ECU.
With power disconnected the ICV should snap shut with only a small gap available...early ICVs had an adjustment screw whereby you could adjust the gap...later ICVs are not adjustable.

If you put +12 volts directly onto the ICV it will snap open all the way.

My understanding of how the ICV works is that the ECU supplies a varying pulse of voltage on the ground side using Pulse Width Modulation which enables precise opening and closing values to move the ICV valve anywhere.

If you put a voltmeter across the ICV point you will not get much of a reading as it is pulsed DC unless the ground has been turned ON all the time which is when you should read +12 volts.

When you took your voltage reading did you test from Chassis to the + point of the ICV...in this case you expect to see +12 volts all the time.

If you measured across the two pins of the ICV you should get a varying voltage depending on engine speed....it is difficult to predict the voltage reading as it is a PWM pulsed DC but there should be some change...you need an oscilloscope to see what is actually happening. If you read a constant +12 volts I would be concerned that your one pin is solid ground instead of pulsing ie. that pin is shorting to ground

Let me read up a bit more on the ICV action to see if what I am saying is correct or total rubbish and I will get back to you.

[soul4t]

Thanks for your input mate! I measured across the two pins of the icv connector plug...if the ecu was supplying constant +12v and that does indeed open the Icv then it would cause a lean condition off the tps closed position!
I truely appreciate your input and if you do follow up your thoughts, we may well have an avenue!

Thanks mate!

[Billwill]

Thanks for your input mate! I measured across the two pins of the icv connector plug...if the ecu was supplying constant +12v and that does indeed open the Icv then it would cause a lean condition off the tps closed position!
I truely appreciate your input and if you do follow up your thoughts, we may well have an avenue!

Thanks mate!

I would suggest you test the ICV in the following way:

Remove the ICV completely...plug the hole in the bottom of the intake manifold with a cork and likewise plug the hose that you removed from the ICV with a cork as well.

Leave the electrical leads connected to the ICV and watch what it tries to do as you turn on ignition, start the engine, accelerate etc. You should see the valve opening and closing by a varying amount.

And make sure you fit the ICV back in the correct way with the arrow pointing in the correct direction.

[soul4t]

Cool, That sounds like a plan....

I have a thought....I have a spare ICV...the test should work perfectly if I disconnect the one on the car...leave it there and connect the 2nd one...ahh, but then I don't know how the car will react...but it should still show varying movement on acceleration or idle cause of the TPS right?...or I'll just do what you said

[Billwill]

Cool, That sounds like a plan....

I have a thought....I have a spare ICV...the test should work perfectly if I disconnect the one on the car...leave it there and connect the 2nd one...ahh, but then I don't know how the car will react...but it should still show varying movement on acceleration or idle cause of the TPS right?...or I'll just do what you said

That is a good idea to leave the existing one connected as it will have the same effect of plug the holes with cork. In fact it is better because the existing ICV will have the slight gap in the valve to let air through. Put the power plug onto the spare ICV and see what happens on the spare ICV

I actually needed to clean my ICV as it has been idling poorly lately so I have just done so now.

I took the opportunity to check the following:

Resistance of the ICV is about 8 Ohms.

There are two wires going to the ICV..the Red/White to ground (with ignition OFF) measures about 8 Mega Ohms ie. not really going anywhere....this is the Positive coming from main relay.
The Yellow/White wire measured to ground (with ignition OFF) is about 8 kilo Ohms ie. not a true ground.

I did not bother using cork to block the holes...dont think it matters too much.

Switch ignition ON...ICV stays closed.
Start the engine... ICV jumps open and flutters around...car quickly stalls. Voltage across ICV is about 12.5 volts...measuring using Red/White to ground or Red/White to Yellow/White
Re-start engine and keep blipping the throttle...ICV moves around trying to do something...mostly flutters around the half-open position.

I do believe that my theory that PWM is used on the negative wire ie. the Yellow/White wire is correct...hence this wire does not go to true ground when ignition is OFF. This also ties in with a Bosch manual I have: "Gasoline-Engine Management: Motronic Systems" which explains that PWM is used to operate various valves in the system although they do not specificaly mention the ICV.

Therefore if your ICV is not working as it should then you need to check that neither of the two wires on it are shorting to ground and the Yellow/White wire is going all the way to the ECU.

[soul4t]

Ahh, nice one! Thanks for checking out the extra bits for me!
I'll try this test when I'm next off...probably saturday! I look forward to it...

I'll update as soon as!

I did test the ICV to pin 4 of the ECU...but the diagrams show the other ICV cable goes to pin 22, but my ECU plug has no.22 empty!

Thanks Mate! I don't feel I can comment about the PWM as I don't know enough about it!

[Billwill]

I find it a bit confusing where the two wires actually go.....the circuit diagrams that I have (USA) show that the Yellow/White wire goes to pin 29(must be Motronic 1.7) and the Red/White wire goes to pin 54.

The ECU pinouts for the Motronic 1.3 which is basically the same as the 1.1 shows the White/Yellow wire on pin 4 which I think is correct...the Red/White wire goes to +12 volts somewhere.

As long as you are getting your +12 volts on the Red/White wire when the engine is running I would not worry about it any further...the White/Yellow (White with Yellow stripe!) wire is the one that goes to the ECU somewhere and this is the one that must not be at ground when the engine is OFF but should rapidly pulse to ground when the engine is running.

I tried Brianmoooores' trick of fitting a small light bulb across the ICV to see if there is any flicker but this would be too fast to see....just as a voltmeter on DC is too slow to re-act to the pulses. A voltmeter on AC may show some varying voltage though.

[soul4t]

Thanks Mate!

I'll check it out! Rain's stopped play this morning, but tomorrow I remain hopefull!

[soul4t]

Ok,

I still need to do some testing, checking where those wires go, as you've explained, if it's not raining, I may be able to do this tomorrow morning...

Both my ICV's measure 9 Ohms.

I did see what happened to the ICV:

I plugged in the other ICV, leaving the current one in place. When the engine was cranking, the Valve was fully open, until the car started, then it closed shut. It remained shut for the duration of warm up. Nothing changed when blipping the throttle either when cold or warm.

When the car reached the tip of the blue area on the temp gauge, it started to hunt, rhythmicly not too fast. I think this might be due to the ECU starting the injection economy thing when above a certain rpm it cuts the fuel to the injectors when the TPS senses closed position, until it reaches a certain lower rpm, then it kicks in again. Interestingly when I removed the FPR vac hose from the throttle body, the hunting was slightly quicker...higher fuel pressure...quicker reaction? Either way...this was interesting so I thought I'd include it!

The same engine behavior occured (engine speed and warm up hunting etc) when trying this test with the ICV unplugged, but not plugged into any other ICV.

I know, in some ways this doesn't show exactly what may be happening, as the idle was mainly high, the ICV would normally get the message to shut, but it does seem to correlate with the engine behaviour when it's plugged in.

So I need to find out what those wires are connected to and ensure the Yellow/white wire isn't grounding! When checking the ECU wires I did find continuity from one of the pins to pin 4 on the ECU, but it could still be grounding somewhere...but then would it open fully on startup?

I'll keep you posted!

[soul4t]

Right then!

I pulled back the protective bit from around the ICV plug to expose the wires.

The Yellow/white wire has no continuity with ground with ignition off...it measured 840,000,000 Ohms (I think!).
With the engine running it fluctuates between 29 and 27 Ohms.

The Red/white wire shows no continuity with anything with ignition off.
With the ignition on it has continuity with Battery positive and shows 12.20v when measured to ground.
With the engine running it measures 13.85v accross ground.

The ICV pins are supplying a constant +13v with the engine running around 12v with the ignition on and nothing with the ignition off.

I tested that I have continuity from the Yellow/white wire to pin 4 on the ECU harness plug.
I also tested that there is continuity between the Red/white wire and one of the pin 87's on the white relay next to the firewall along from the Diag port...I think it's the main relay...DME?

With ignition off, ICV remains at default position...small air gap, with ignition on, ICV remains in this position, with engine cranking, ICV opens wide, when running, ICV shuts tight.

I saw on another thread that someone had tried another ECU and when there was no improvement, they opened up the ECU's and both had fried circuits...I took a look at the ECU and I'd need to pry up some smallish metal clamps to get into it...as well as unscrewing a few torx screws...didn't fancy it, has anyone done it without causing any problems? As it seems that the ECU is sending the wrong message! The CPS has been changed, checked so that can be ruled out...surely the ECU is the last thing to check!

Cheers!

[Speedtouch]

Opening up the ECU for inspection is fairly straightforward, and if you take care, you are unlikely to damage anything.

[soul4t]

Ahhh, @Speedtouch...I was just reading some other threads and was thinking of PM'ing you this question!

Thanks! I'll take a look...probably tomorrow now as I need to go to work!

[mcbonio]

You don't need to remove the torx screws, just pry up the metal tabs along the edges. The ECU unit as a whole is more robust than you think. You won't break it opening it.!

[soul4t]

Ah thanks mcbonio!

I'll keep you posted as usual!

[Billwill]

Yeah worth opening up the ECU and having a look inside to see if there is obvious damage.

Are you 100% sure the TPS signal is working and is getting all the way to the ECU?

[soul4t]

Right,

I've opened up the ECU, got to both sides of the board and it looks brand new! No signs of damage...good news I suppose!

I double checked the TPS, T1 to Pin 52 on ECU, T2 to ground and T3 to Pin 53 on ECU...all good...I also checked that with the throttle plate closed, the TPS senses closed...continuity T1 and T2.

I also sprayed brake cleaner around all the intake gaskets, hoses and mating surfaces! No change...I removed the FPR vac hose to TB and a rise in rpm...then sprayed at the hole and a raise too...then a slight stumble.

When it's cold the exhaust smells rich...then warm, not so.

So what on earth might be happening! I suppose it's still possible that the ECU is bad, but it's unlikely isn't it?

[mcbonio]

Still not sorted this Soul? I have exactly the same issue and symptoms as you with my m30, I've also done all the checks you have too.

When I start mine up there there's an almighty gasp of air from the engine and then for a minute there is a greyish fuel rich smoke until it warms slightly then it idles fine. Does yours do this? I'm suspecting leaky cold start valve or injectors seeping fuel over night then being partially burned off in the exhaust?

[soul4t]

Hello mate,
Hmm, interesting...by almightly gasp of air, do you mean rpms rise, then fall smoothly to a low rpm?
I have the greyish smoke out the back, but not loads. Mine gradually becomes a more stable idle as it warms up...it doesn't hesitate when I touch the throttle when it's past the blue on the temp guage, it does hesistate before it gets to the end of the blue mark. I don't know enough about your engine management system...is it supposed to idle high for the warm up? What does the cold start valve do?

[mcbonio]

I'm not actually sure my late m30 engine has one as its motronic 1.3, I think only motronic 1.1 engines did. They basically inject more fuel into the engine when it's cold.

They can cause cold idle and rich running problems. They either get gunked up and don't open or stop closing all together.

What year is your car again?

[soul4t]

Mine's 1990...
If it's motronic 1.3 the blue temp sender is one to check...if you check your ECU the 3 row plug is motronic 1.3 2 row plug is 1.1 AFAIK.
what year is yours?....it might say on realoem.com
..

[mcbonio]

I have the greyish smoke out the back, but not loads. Mine gradually becomes a more stable idle as it warms up...it doesn't hesitate when I touch the throttle when it's past the blue on the temp guage, it does hesistate before it gets to the end of the blue mark.

This is exactly how mine behaves. I'd love to get to the bottom of it.!

[Speedtouch]

Many older cars are like this. Probably something to do with gaskets and metal surfaces that expand once warm, and contract when cold, letting in small air leaks.

You only need to introduce a small air leak into a fuel injection system for it to cause all manner of mayhem, and confuse the ECU.

[soul4t]

@Speedtouch...do you think that an exhaust gasket could cause such a confusion (cars without O2 sensor).

I've replaced the intake gaskets and the rocker cover...all vac hoses except 1 has been replaced...I'm pretty sure I've replaced the TB gasket too...I've sprayed loads of Brake cleaner round the gaskets too and the remaining vac hose that I haven't replaced, but have inspected. Even after doing all this I still want to be 100% sure it's not a vac leak. When I remove the FPR vac hose or the opposite side (that would normally go to the carbon canister I believe), the idle rises and it sounds more comfortable.

I'm going to do a video to show what's going on!

@mcbonio, do you have an ICV...it'll be in the video so if you don't know now, you might get it from the video...if you know and you do have one...have you done the test that Billwill suggested?

I think I'm going to do the ICV test again, but plugging the holes to simulate exactly the level of air it's letting in...expecting to see the ICV open to compensate and let air in.

To be continued!

[parkin10]

Many older cars are like this. Probably something to do with gaskets and metal surfaces that expand once warm, and contract when cold, letting in small air leaks.

You only need to introduce a small air leak into a fuel injection system for it to cause all manner of mayhem, and confuse the ECU.

i`d go with speedtouch on this
it took me ages to realize i had tiny air leaks on the vacuum hoses were they go into the throttle body on my m20 engine
its still a little bit flat at low revs when its cold

[soul4t]

Hmm...yea, maybe I should just change that last vac hose then! Cheers!

[mcbonio]

Hey Soul,
Yeah my m30 does have an ICV and its working fine. I did however strip all the intake system down again tonight do go over the hoses one more time and clean out the ICV hoses and AFM.

I popped the little bung off the adjustment screw hole and just out of curiosity turned the screw and found it was about a quarter turn from being wound in all the way.! I had previously thought it hadn't been tampered with as the bung was still on. Somebody has monkeyed with it and put the bung back in.

I fired the car up and began to lean out the mix by unwinding it a half turn at a time until the engine no longer was bogging and the revs comfortably and accurately returned to idle speed after opening the throttle a bit.

I took it for a spin, the car is much perkier and free revving, it feels like it's got 40bhp more I kid you not.!

I need to borrow a gas analyzer and do it properly but I may have found out what was wrong with mine, I suggest you check yours.

[soul4t]

Ah, nice one!

I don't think that mine has the adjustment screw...I'll double check though. Nice one for sorting it though!

[parkin10]

just remembered
when i was looking at the vac hoses i found that the servo did not hold a vacuum from the one way valve back
i got a bone yard servo to sort that
when i got the old servo off there were bits rattling around inside

[mcbonio]

Yours will have the screw Soul, I think it's underneath the AFM with a black bung.

[soul4t]

Ah, the CO idle adjustment screw on the AFM? If it is, it should only adjust the idle mixture...when mine was unscrewed alot of turns ago, lean, it was even worse...I then turned it clockwise to richen it up and it raised the idle speed...so I've compensated for something else not working, but it stops it stalling/vibrating when warm!
@parkin10 I'll have a good look at the servo...how can I check that valve?....I suppose I could disconnect it to block the hole manually, then I'd see if it's improved...but I'm pretty sure that my problem with idle is due to a lack of air because the ICV isn't opening at all...

[Speedtouch]

@Speedtouch...do you think that an exhaust gasket could cause such a confusion (cars without O2 sensor).

Quite possibly. Without the correct back-pressure, the car won't run right.

Pop the bonnet, start the engine from cold, and listen to the area around the exhaust manifold for any signs of blowing - you'll hear a chuffing sound, and maybe even see where the exhaust is escaping from.

[soul4t]

Ok cool,

I'll check these things tomorrow and most likely include a vid!

[mcbonio]

Get the vid up and lets have a goz.