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 STM Tuned

1G 90 No voltage to Idle Position Switch (IPS)

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.

Scrubos

Probationary Member
14
2
Feb 11, 2018
Marshfield, Missouri
Key on, engine off, I have about a tenth of a volt at the IPS, with it unplugged. Confirmed no short between IPS and ECU. Also, key off, the IPS wire has continuity to ground, but only so long as the TPS is plugged in, throttle closed. It loses this continuity if you open the throttle about halfway or more.
Is there any condition that would cause these symptoms, or is my ECU just shorted internally?

It intermittently throws code 14. I'd say at random, but it only pops the CEL under load/boost. TPS itself checks out, we put it on the scope and it has no bad spots, returns to the same place every time. :thumb:

Car runs out great, but it's painfully obvious the ECU has no idea when the throttle is closed. :notgood:
 
I'm not sure I'm following you.

With the IPS unplugged you see a tenth of a volt at the IPS connector? So there is nothing connected to it's terminal but one meter lead and the other lead is to ground someplace and you're measuring voltage? That doesn't bother me. It's not uncommon for there to be small voltage differences between different points. Dissimilar metals in contact can generate small currents. Lookup electrochemical potential difference if you interested. Resistances in ground paths can also cause interesting readings.

If the green IPS wire is connected to the IPS it's supposed to be grounded when the throttle is closed. What it's not supposed to do is remain grounded until you open the throttle about halfway or more. It's supposed to loose continuity to ground as soon as the throttle moves off it's stop.

Just to be sure, the throttle body ground strap is attached and the Green IPS wire have either 5v or 12v on it with the key is on and the connector is disconnected from the IPS?
 
Let me try to clarify by answering you in reverse order.

It does now have multiple ground straps from the throttle body, as the factory strap was missing when acquired. Didn't help my hesitation off idle symptoms.

So next I go to check for voltage coming to the IPS. Unplugged the IPS wire from the switch, and test for voltage on that wire, other probe straight to battery negative with key on and I have only 1/10th of a volt, not the aforementioned 5v or 12v.

With the switch still unplugged, I measure continuity through the switch, and it behaves perfectly, losing continuity the instant I crack the throttle, and regaining continuity upon closing the throttle. However, with the IPS wire reattached, it has continuity to ground until the TPS detects the throttle approximately halfway open or greater. If I disconnect the TPS/IPS harness and measure continuity on the IPS pin at the harness connector, it behaves normally. But still only a tenth of a volt on that IPS pin.

It would seem to me the ECU has something awry with the TPS/IPS circuit causing it to ground the IPS internally through the TPS circuit.

I wish schematics we're available, my dad troubleshoots boards for John Deere (John Deere 1025R Problems). He might be able to fix it even without, but I just wanted to double-check this problem was new to the forums
 
Last edited by a moderator:
Thanks,
Just to be sure, you have a 1990 ECU and not a later one? This IPS wire is one of the two that changed position on the ECU between 90 and 91. The signal it's swapped with is the MAF reset line from the ECU.

Do you know how to identify a 90 ECU PCB? Many cases have been swapped over time so looking at the circuit board is the only 100% way to be sure.
 
First the ECU location (just to be clear).

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


Second a 1990 AWD Manual will have either a MD145903 or MD145902 ECU depending on if it's a Federal or California Car. The cover will look like this:

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


In the lower right corner it will have either an "A" or "E". The "E" indicated the case is from an EPROM based ECU which is handy for DSMLink or Chip Tuning.

The PCB inside should look like this non-EPROM board. You'll note that the number on the white connector matches the last 4 digits of the second line on the case. (The third line is a date code) Also mote that in the center of the board are three vertical ceramic boards (Hybrid IC's). Two on the left side of the capacitors and one to the right. The later 91-94 ECUs have one on each side.

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


The California version is exactly the same except for the software inside the processor and the part number. Here is a California EPROM ECU which is again like the Federal EPROM except for the Software version and part number. It's EPROM would say E742.

You must be logged in to view this image or video.
 
I have five running mid-90's Mitsubishis. About half came with either a wrong ECM or a wrong TCM. It's a pretty common problem.
 
Okay, sorry for the delay, but I have updates!

Case is 90, board is 90. Definitely had original caps leak out, but it looks properly remanned, cleaned up, traces peeled back, etc. Dated 2003 on reman. Replacement caps still test good.

On a whim I threw it back in there, but left it dangling in the floor board to see what it'd do, and son of me biscuit, the car runs perfect. Got almost the full 12v coming to the IPS. Ground the ECU case, and the problem comes back.

So, confirmed internal short to case ground. As for where, will require more diagnosis from my father, or more likely someone with schematics. Pin 6 does not have a hard short to the case, so somewhere in the circuitry.

For now, I plan to run it with the ECU dangling. I've read where 1g guys have done this with no problems. Someone correct me if I'm about to get myself stranded by leaving it loose. And/or do you think it's more likely to blow the ungrounded ECU, because of the internal short?

Thanks for all your help
 
Last edited:
This is probably pretty easy to find and fix. Find where the case is connecting to internal circuity. That will either be one of the four screws that mount the board to the case or some component touching the case. Fix that problem.

The case doesn't need to be grounded for normal operation -- if it did you couldn't run with it dangling as we've all done. So whatever's connecting to the case isn't the ground bus on the board: It's something else that doesn't need to be and you can probably find and disconnect it.

The only issue I could think of is that the case does shield the internal electronics from radio waves. The situation you've got is the case is gonna act like an antenna. So it is possible that going too near the wrong radio or TV station, microwave tower, industrial facility or ??? could leave you stranded. This is unpredictable though ... you might pass WXYZ-FM heading north and do fine but stall out in front of the driveway going south.

This is a weird problem -- one I would never have imagined.
 
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