The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

Please Support Kiggly Racing
Please Support Morrison Fabrication

General 1990 ISC motor – reverse polarity for new part?

This site may earn a commission from merchant
affiliate links, including eBay, Amazon, and others.

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

We're on Boost

15+ Year Contributor
2,113
851
Aug 25, 2007
Seattle area, Washington
When plugging in a new Airtex or Standard Motor Products ISC to my 1990 harness, it runs backwards.
The pintle extends when you turn the key on, and retracts when you turn the key off. That’s backwards.
My original 1990 ISC does it right.
The pintle retracts when you turn the key on, and extends when you turn the key off.

I’ve looked through lots of ISC posts and haven’t found anybody else reporting this.
Seems I will have to reverse the polarity of one of the field coils to reverse the direction of the new motor, unless there is some other explanation.
But everybody says the wiring is the same for 90 vs 91 and later ISC’s.

Does anybody have an idea what’s going on? It just seems wrong.

The original 1990 ISC tests at 30 ohms on all 4 coil pins. I pulled it out of the car after a few episodes of galloping idle, high idle, and finally at the end it was making a loud clicking noise like a fuel pump whenever the ignition key is on. But it quit making that noise so that now I can do testing on it without any drama.

I’ve looked at the ISC drivers on the ECU board and they don’t show any physical damage, in fact they look brand new like fresh from the factory. LOL. 1990.
The ECU board looks clean and nice from what I can see on top, capacitors were replaced by ECMtuning when they did the whole EPROM conversion and so-forth in 2015.

ECU is a 1990, with ECMlink V3.
Logs show ISC steps at 120 when I first turn the key on (engine cold), which then goes to 90 when I turn the key off. That I believe is normal. That 30 steps makes the pintle move about .040 to .060 inches, as near as I can measure it (about 1 to 1.5mm).
All 3 of the ISC’s are at about the same initial position when disconnected – with the pintle tip at about 1 and 11/32 inches out from the flange face.
All 3 of the ISC’s move about the same amount – just the direction is opposite!

The 2 new ISC’s I bought from Rock Auto for about $80 each and they look beautiful.
They are listed here:
http://www.rockauto.com/en/catalog/...,fuel+&+air,idle+air+control+(iac)+valve,6072
 
91 iac in 90, looking for the ecu pinouts between 90/91

No idea why replacement 1990 ISCs would be wired incorrectly and require a rewire, though.


Wow, I think that thread is saying the same thing I'm saying, except I haven't tried it yet (swapping the wires). Thanks for posting!

In post #5 he says "The lower row left and right side need to be switched" (at the plug) which, no diagram, but I think he is saying the same thing I want to try.

The connector pins which are numbered 1 through 6, he is probably saying to switch #4 and #6 with each other. That reverses the polarity of one of the field coils.

Stepper motors are either bipolar (4 wire) or unipolar (6 wires or 5 wires). The bipolar ones for sure will go the reverse direction if you swap the wires to just one of the coils (not both).
I think it works the same for unipolar steppers, and unipolar is what we have on these cars.
So, #1 could be switched with #3
or #4 could be switched with #6.
Either should work.

I'll change that last bit of harness that I have to extend anyway, because the 1990 harness isn't long enough.
 
Last edited:
:thumb:

The way the factory would have it would be swapping pin 4 (white, on a '90 would run to ECU pin 67) and pin 6 (black, on a '90 would run to ECU pin 68).
 
:thumb:

The way the factory would have it would be swapping pin 4 (white, on a '90 would run to ECU pin 67) and pin 6 (black, on a '90 would run to ECU pin 68).


Ok I'll do it that way. Thanks!

Here's a nice pic of the original 1990 ISC that was in my car. This car is an early 90, built in September of 1989. Pic is 4240 x 2832 pixels so it will go very big, if it loads up here fully.
Aww, it came in at 1600 x 1069 pixels, still not too bad.
You must be logged in to view this image or video.
 
Here’s a pic of my 9 inch extensions to the harness wires.
Now the plug can reach the ISC position easily.

The plug here is kind of upside down. The upper 3 wires in the pic are actually wires 4, 5, and 6.

I used 18 gauge and 16 gauge GXL engine compartment wire,
parallel connectors (crimp) item number 214305 from Del City,
heat shrink tubing sizes ¼”, 3/16”, and 1/8”
a Channel Lock No 909 crimping tool (great tool for generic crimps)
and a heat gun.

You must be logged in to view this image or video.


...... and here's a video of the pintle retracting (ignition on) and extending (ignition off) like it should, which was the whole point of swapping wires 4 and 6.

You must be logged in to view this image or video.

I finally realized I could rotate the video 90 degrees to make it full screen, so it's full screen now LOL. A little easier on the eyes.
 
Last edited:
The wiring of the ISC coils between 90 and 91 changed at the ECU. On a 90 ISC coil C & D is reversed pins 4 and 6 are swapped relative to the ECU pins on the later cars. ISC pin 4 went to ECU pin 67 and ISC pin 6 to ECU pin 68 on a 90. I always assumed it was a difference in the internal wiring of the old ISC since I didn't notice a difference in the code or operation on my test bed when using a 90 ECU and 91+ ISC but I wasn't looking specifically for it.

We used to point it out to people swapping to 91+ TB that they needed to swap the wiring when they hooked up the 91+ ISC on a 90 car.
 
Steve, Brian, maybe you guys can give me an opinion about something that shows up pretty well in my video. I noticed this first by measuring the pintle position before and after the video. It wasn't the same. It wound up a little more extended at the end than it was initially.

Sure enough in the video you can see this happen.

It happens in the first on-off cycle in this series of 4 cycles. The YouTube player shows this really well because you can hover your mouse over the timer bar (with the player on pause) until you see the time you want, click it, and bam it snaps there, still on pause. Or for a 10 second jump you can use the hotkeys L and J.

So if you snap it to these times, you will see the 5 times that it is in extended position:

1 sec
11 sec
19 sec
25 sec
33 sec

At 11, 19, 25, and 33 sec, the positions are all identical. But in the intial position at 1 sec, it is slightly less extended. If you snap back and forth between 1 sec and 11 sec you will see the difference perfectly. You could run the video to 1 second, then hit the L key, then hit the J key, to jump it back and forth.

So I’m wondering, is this how it is supposed to act, or is this something weird?

This is actually the 2nd video I made of this. The first video which I made a few minutes earlier shows the same thing happening.
 
I am certainly no expert on electronic devices, but it looks okay to me. One thing I do notice on the video is that the "retract" at 0:01 actually extends first so it knows pretty well where it is and ultimately figures out where it needs to go.
 
I am certainly no expert on electronic devices, but it looks okay to me. One thing I do notice on the video is that the "retract" at 0:01 actually extends first so it knows pretty well where it is and ultimately figures out where it needs to go.

Yes it does start out (in that first retract) by extending a little, then retracting. I've watched this video frame by frame in a video editor, which is a little better than what we can do at 1/4 speed in YouTube, and sure enough, in this 30 frames per second video, it does spend about 5 frames worth of time going the wrong way there haha!
I am hoping the same thing you are saying, that maybe this is ok. I'm just going to go ahead and put it all back together and drive/log with it for a while, and see what happens.
 
On my setup with Magnus intake manifold and Boomba throttle body, the 91+ type ISC interferes with the stock battery shelf.
Since I’m using a Miata battery anyway I don’t need the whole width of the battery shelf, so I cut it down with a band saw. Took off one leg and left the other leg there with only 1 spot weld.

This shelf was built like a tank! It was a real battle with my band saw even with removing most of the doublers first by drilling out the spot welds. Looking for a new band saw blade now LOL

You must be logged in to view this image or video.
 
I got a nice marked up photo today from Boomba support showing the idle set screw on this Evo VIII-IX throttle body. This is what you use instead of a BISS. It does it the old-fashioned way, sets the point at which the butterfly stops just short of completely closed.
"Search google for this image" doesn't find any photos marked up like this, so for anybody wondering about it, here it is!

You must be logged in to view this image or video.
 
Last edited:
The new ISC is working out real well, but I did still have some high idle issues even with the new ISC, mostly while rolling around slow in parking situations.

To me it seemed like it was caused by too much ignition advance, as per this in the ECMtuning wiki called Idle Problems:

If you are running a configured idle speed that's higher than stock, it's possible to observe a slight oscillation in idle speed even when everything else is working properly. This is because, by default, the timing tables in the ECU significantly advance ignition timing as the engine speed rises above 750 rpm, and engine speed tends to increase as ignition timing is advanced. You'll want to reduce this positive feedback by retarding the ignition timing at idle and slightly above idle. It gives the ECU's fine idle control mechanism more authority.

My target idle is 1100 rpm. Looking at the numbers I had in my max-octane ignition advance table in ECMlink, the low load advance seemed awfully high. Then looking at my logs, it seemed like the problem was not just at 0.3 load factor, but also at 0.4 load factor. So I reduced those numbers by a lot, all the way from 1000 rpm through 2500 rpm, and put the same numbers into the min-octane ignition advance table as well.

This helped tremendously - made the car a lot more drivable at parking lot speeds, maneuvering into the garage, stuff like that, and a much more stable idle speed. It’s a lot nicer to drive and I’m going to save some wear and tear on my clutch!

Here are before and after shots of that area in the table. You can see I changed 12 cells.

Reduced timing:
You must be logged in to view this image or video.


Original timing:
You must be logged in to view this image or video.
 
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Innovation Products Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com RixRacing RockAuto RTM Racing STM Tuned

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Back
Top