Is this normal?

I have been chasing these gremlins for a week or two now and the situation is getting better but is not totally resolved. Replaced a 75 amp alternator with a 90 amp one, put on a new belt, realigned the belt by spacing the alternator pulley out a bit so it runs really straight. The new belt has been on a few days now so I'll take up any slack and tighten things down.

Car idles great during the day once it is warmed up. It idles a tad high (around 950-1000 rpms) but that is because we had to bump it up because as soon as I put the headlights on rpms drop south by more than 200 rpms making the car vibrate a lot.

Problem seems exacerbated when driving and coming to a stop when the rpms drop really low and then pick up. If the car sits at idle for any length of time with the lights on rpms steadily drop to under 800 rpms and the car starts vibrating. switch off the lights and the rpms pick up and everything smooths out.

Is the alternator still on the way out and not producing enough juice? It seems to be charging fine other than this. What is the correct charging voltage I should be seeing to assess if it is up to scratch?

If the alternator checks out okay, what is the solution? More grounds? Thicker charging wire from the alternator to the battery? These are 93 lights and had to be spliced into the existing harness. Any possibility this is causing a power leak somewhere? Would relays anywhere help?

I'm stumped on this but I need to find a solution because the car is a dream to drive during the day but at night it becomes a PITA if I have to stop anywhere.

Paul.

You might have a bad battery. With engine running, alternator should put out around 13.9-14.2V. Measure that directly at the alternator terminal please. Voltage at the battery when the engine is running should be very similar to the alternator V reading. Now turn on the headlight and see if alternator voltage and battery voltage is affected. Measure battery voltage with engine off, with headlight turned on and turned off (key in ignition but don't crank the engine)

Also measure what amperage draw you have on battery with everything off and key out of the ignition. Should be very small, less than .002Amp

Thank you Andre!

I had wondered about the battery, but I wasn't sure if it would affect things if the car was running. I'll check all these and get some figures up.

Paul.

With no idle control it would not be unreasonable for the headlight to pull enough power to cause your engine to drop by a few hundred rpms. I'm assuming that you do have idle control in order to compensate for electrical draw. In that case you could have a short in the head-light circuit. That would cause it to pull extra current and overload the alternator and, in turn, the engine. Or, the alternator itself could have an internal short and cause the same thing. Or it could be that your ISC is out of range so that it can't open the throttle enough to compensate for the alternator load.

Yeah, I still have idle control from the ISC, just no FIAV. When you say the ISC could be out of range, what would cause that? Incorrect BISS screw adjustment? I noticed the car ran better today and the idle is now higher. I am wondering if what Andre said above may be true but that instead of the battery being bad it just wasn't charging fully before because of the other issues with the alternator.

RPMs do still drop about 200 when I switch the headlights on though, so I think I'll take your advice and check those wires as well.

You make still have an ISC but it doesn't sound like it's working.

What the idle speed in the ECU set to? If it's not 950-1k then you have to have driven the ISC out of range (all the way closed) before the idle rises. The ISC won't do squat until the RPMs drop below the programmed idle.
Have you checked that the IPS is working? The ECU has to know that the throttle is closed before it will manage the idle.
I find missing throttle body ground straps on >50% of the car I see and on a DSM or GVR4 there used to ground the IPS so it can work.

when you put the steering on full lock does the ECU attempt to bump up the idle speed or do revs drop also?

Quoting stevep:
You make still have an ISC but it doesn't sound like it's working.

What the idle speed in the ECU set to? If it's not 950-1k then you have to have driven the ISC out of range (all the way closed) before the idle rises. The ISC won't do squat until the RPMs drop below the programmed idle.
Have you checked that the IPS is working? The ECU has to know that the throttle is closed before it will manage the idle.
I find missing throttle body ground straps on >50% of the car I see and on a DSM or GVR4 there used to ground the IPS so it can work.



Steve,

Thanks for taking the time to look at this, your input is very much appreciated but I get a bit confused by the different terminology that gets used. I'll try to go through this based upon my understanding of what you are asking.

As far as I know, the ECU has only been flashed for 560cc injectors and that wouldn't necessitate a higher idle so I am assuming it was set for stock but I don't know and I'm not sure how I can verify this without a chip reader.

The ISC could be an issue. No-one seems to know why but I purchased two brand new ISCs from the States and they would not work in my JDM throttle body. They were too short and would not close at all. I had a constant 1500 rpm idle. I went back to the JDM ISC and got idle back into the 700-1000 rpm range but it is a 2nd hand ISC and it may be worth investing in a new one. I can order one when I next go to the dealer. If the ISC isn't the problem it will be worth having a spare anyway.

What is the IPS Steve? Is that what I see some people refer to as the throttle closed switch? If so this could be an issue. I have an aftermarket S90 throttle body. The entire idle control assembly with the FIAV and ISC bolts on but I have a feeling that I no longer have a throttle closed switch. Will the ECU really not activate the TPS without this and is there anyway around this if not? If it is as simple as that, I guess I'm going to have to fabricate a bracket and switch assembly. Obviously without the throttle closed switch my throttle body itself is not grounded. If my ISC isn't being activated I doubt that will make any difference, but is the throttle closed switch the only item on the throttle body that requires a ground and could my solution be as simple as re-grounding the whole throttle body assembly?

It would help dramatically if you or anyone could explain any of the above. The car is so nice to drive at every other speed , I'd be willing to do whatever is required to sort out this issue.

Quoting fuel:
when you put the steering on full lock does the ECU attempt to bump up the idle speed or do revs drop also?



I haven't noticed any issues on full lock (I'll have to check), but I do have issues I noticed with the A/C turned on. That hadn't been an issue before because it was cold here so I had the A/C switched off. I'm thinking that my ISC isn't working as Steve suggested and that this throttle closed switch could be the culprit. I assumed my ISC was fine because despite blocking off my FIAV, I still got a higher idle when the engine was cold. Is that function of the ISC activated under a different circuit (coolant sensor)? If so it would make sense that my issue is simply that the ECU is not getting told that my throttle is closed and that it needs to activate the ISC.

But if that is the case wouldn't I stall every time ... or not necessarily?

EDIT: Confirmed! No throttle closed switch. Working on a fix for that!

^^ Nice catch. With the throttle_closed_switch not working the ISC doesn't try to maintain idle rpm 'cause it expects you to override it with throttle input. It's in dashpot mode so it just tries to "smooth" the effects of your right foot. The switch has to trigger for it to take over idle control.

Thanks Roger!

With that in mind, it doesn't sound as if it is doing too bad a job of maintaining idle given it has so much stacked against it. gives me hope that if I get this switch back in operation the idle should be pretty damn good.

BUT is there any other place I could locate the switch other than the throttle plate. I know DSMLink can 'simulate' this switch somehow (possibly by taking a reference from the TPS?) I'm just wondering if machining or welding something onto the throttle body is the only solution?

put it on the throttle pedal inside the car

John, seriously, that would work fine?

I was thinking about doing that by modifying a clutch cutout switch or something but wasn't sure if it would be as good. I guess as long as you don't rest your foot on the throttle at junctions it should work right?

have curtis get you an audi 5000 turbo throttle microswitch from the yard.

fab bracket so pedal trips it when fully closed

done deal

So you physically do not have an idle switch? Easy. Use a 2G DSM tps instead of a 1G TPS. The 2G TPS is 4 wires instead of 3, the 4th wire is for idle switch. Done.

John, Andre thank you both.

Andre, so if I grab a 2G TPS my 1G equivalent harness plugs straight into it and I just cut the 4th wire, cut the orinal closed throttle switch off my existing harness and find a connector to plug the two into each other?

Quoting grocery_getter:
So you physically do not have an idle switch? Easy. Use a 2G DSM tps instead of a 1G TPS. The 2G TPS is 4 wires instead of 3, the 4th wire is for idle switch. Done.



Think he runs the 90 style t-body, and the 2g tps won;t bolt to that. /ubbthreads/images/graemlins/dunno.gif

Need to check that, actually! It is an aftermarket throttle body that accepts the OEM idle control assembly. I seem to remember it is actually the 91-92 assembly that bolts in without modification, whilst the 1990 assembly required some 'massaging' to fit. It could just be that I might get lucky. It would be a nice easy fix if I could do this!

Edit: I think the EVO III TPS is also 4 wires. In which case I'm pretty sure this will work.

Good luck Paul!

Quoting cheekychimp:


As far as I know, the ECU has only been flashed for 560cc injectors and that wouldn't necessitate a higher idle so I am assuming it was set for stock but I don't know and I'm not sure how I can verify this without a chip reader.



Is the chip labeled in any way? Who made it, Jeff?

Quoting cheekychimp:
The ISC could be an issue. No-one seems to know why but I purchased two brand new ISCs from the States and they would not work in my JDM throttle body. They were too short and would not close at all. I had a constant 1500 rpm idle. I went back to the JDM ISC and got idle back into the 700-1000 rpm range but it is a 2nd hand ISC and it may be worth investing in a new one. I can order one when I next go to the dealer. If the ISC isn't the problem it will be worth having a spare anyway.



Can't help much here but if your comments in other threads on your harness having the pins incorrectly swapped and your comment about not having an IPS It's hard to conclude there really was a problem with the ISC. How are you attaching the new style ISC to the car harness? I got and continue to get the impression your car is early style so it would have the early ISC with a pigtail. I ask because I've seen people just follow wire colors making adapters and that's winds up connecting some of the coils reverse polarity so it steps in the wrong direction.

Quoting cheekychimp:
What is the IPS Steve? Is that what I see some people refer to as the throttle closed switch? If so this could be an issue. I have an aftermarket S90 throttle body. The entire idle control assembly with the FIAV and ISC bolts on but I have a feeling that I no longer have a throttle closed switch.



Noted below that your missing the Idle Position Switch = throttle closed switch.

Quoting cheekychimp:
Will the ECU really not activate the TPS without this and is there anyway around this if not? If it is as simple as that, I guess I'm going to have to fabricate a bracket and switch assembly. Obviously without the throttle closed switch my throttle body itself is not grounded. If my ISC isn't being activated I doubt that will make any difference, but is the throttle closed switch the only item on the throttle body that requires a ground and could my solution be as simple as re-grounding the whole throttle body assembly?



Correct, if the ECU doesn't see the signal get grounded by the IPS it doesn't know that the throttle is closed or that it should manage the idle. Yes, the IPS is the only thing on the intake manifold that uses the ground. As suggested below, if you can use a 2G TPS then use it and follow a modified 2G TPS adjustment procedure to get the switch to close when the throttle is closed and open when the throttle is opened slightly. Keep in mind this signal is one of the two that swaps around between 90 and 91. If you have a 90 stype ECU it should runs to pin 6 and if you have a 91+ style to pin 14.

Quoting stevep:
As suggested below, if you can use a 2G TPS then use it and follow a modified 2G TPS adjustment procedure to get the switch to close when the throttle is closed and open when the throttle is opened slightly. Keep in mind this signal is one of the two that swaps around between 90 and 91. If you have a 90 type ECU it should runs to pin 6 and if you have a 91+ style to pin 14.



Oh how could I be so stupid!

You have no idea how much I wish I had realized the significance of those two pins before now. The things you have said together with what Phil has told me in the JDM ECU thread are really starting to make a huge amount of sense now.

When I had idle issues initially and intended replacing the ISC I also thought with the car being a 1990, I would need pigtails to install a new one. I even had a member cut me a spare pigtail off a 1990 harness and send it to me. I was really surprised when we came to actually replace the ISC to find that the harness actually plugged straight into the ISC. Now it makes complete sense because Phil has been suggesting that only pre-facelift 88-89 JDM cars match the USDM 1990 DSM wiring whilst both my 1990 JDM cars are probably identical in configuration to the USDM 91 GVR4s.

Having been blindsided by the Fluidampr Pulley timing mark being off which I thought was the real reason my other car ran like crap, I think I have been creating my own problems with that car all along without realizing it because I swapped pins 6 and 14 when I clearly didn't need to.

Still good news if that is the issue. It should be an easy fix at the end of the day, I can run an EVO III/ 2G TPS in this car to fix the idle and I should be able to swap my wires back and hopefully get Link running in the other car now.