91 formula tpi map. Been having my tps voltage go all over the place with an intermittent high idle due to the iac moving from incorrect tps readings. Ground on the passenger side of the head is perfect. Checked the 5v ECM reference on the tps sensor and when the intermittent issue occured, and the reference voltage dropped all over the place and eventually read zero. When the car is running normal and the tps voltage is steady, the reference is exactly 5v. Where can I find the ground specifically for the ECM? what other testing can I do to find out why my ECM reference is dropping out and screwing with my TPS readings?

 

The 5V reference source can aslo be monitored at the MAP sensor. If the voltage is also varying there, it may indicate a connection problem at the ECM or in the harness/connectors outside of that.

Also, pins A4 and A5 at the ECM are the 5V reference source in case you decide to monitor the voltage at those points for comparison.

If the voltage varies at the ECM, the source is a regulator on the the logic gate U17. It is filtered by a 47µF electrolytic and protected by a 0.1µF ceramic bypass cap. We know that 30+ year old electrolytics can become questionable, so it wouldn't be a complete surprise either way.

 

If you're measuring at the TPS sensor, I'd move back to the ECM connector and monitor there as well.

Also, with engine off, key on, monitoring at the TPS sensor, jiggle the wires around and see if you can get a drop out or fluctuation on the 5V.

Basically trying to rule out the wiring harness.

 

Yes I'm measuring at the tps sensor an I gave the wires a tug while measuring and they where fine. the tps sensor and connector is new and the solder job was very solid. ill move to the ecm reference wire and measure off the ecm connector.any thing else I should look for while im back there?

 

sounds good. ill measure at the map too. ill get a measurement off the a4 and a5 pins. also that regulator, if I do find that is the case, how would i go about replacing that? I've never heard of that logic gate or whatever. I'm very new to the third gen electronics area.

 

In addition to the 5V reference and sensor voltage back to the ECM, I'd check all the grounds, not just the sensor grounds.

 

yes, I've also had trouble finding all the grounds. Do you have pictures, or a diagram? I know there's one on the passenger side cylinder head but I can't seem to get a straight answer about any other grounds. I also don't know how many there are.

 

https://tpiparts.net/1227730_pinout_diagram
all ecm grounds, map, tps and o2 all are at the back of the drivers side head.

 

all grounds are good, my mechanic told me it could be a bad ecm because of the 5v reference circuit going bad. Could that be the case? Recently it's been tripping a code 32 also.

 

Could be multiple issues if your getting a egr code.

 

yeah it only happens when I have that same reference 5v dissappear and it's got a new map sensor too. Plus if I had a vacuum leak wouldn't the high idle be all the time?

 

I would think a bad egr solenoid if anything but need to fix the 5v ref to tps before you further diagnose another issue.

 

Did you do whay Vader suggested and monitor the 5V at the other locations? It appears the 5v reference for the MAP sensor comes from the same source within the ECM. If they're all going down simultaneously, then most likely it's the ECM. Other possibilities is an intermittent short to ground on the 5V external to the ECM, thereby pulling down the 5V to 0V.

How are you measuring this? With a multimeter on the sensor? Or via datalogging the ECM?

 

Been having an issue on my 91 formula where the tps signal starts going all over the place when the 5v reference disappears. The 5v ref reads zero on the tps connector when the issue starts. Usually starts when the engine reaches closed loop. I read it off the ecm with my probe and it is a solid 5v the whole time while the tps side is at 0. Right and left grounds on the cylender head are good,, new tps new connector. Is there anything but wire in between? Is there a fuse box? Can I just run a wire straight to the tps from the ecm? Any answers are appreciated!

 

If you aren't getting the 5v reference from the ECU (and that 5v is good at the ECU connector), then you have a wiring issue and you don't have 5v going to the TPS. You need to trace the wiring and find out where you have an open or short to ground that is causing the 5v to disappear at the TPS connector.

GD

 

thanks, so no fuses in between? And is there a diagram for that harness or am I just gonna have to wing it?

 

You can buy a factory service manual on CD for $25 on eBay or you can go here and start tracing your wiring. http://www.austinthirdgen.org/index.php?pid=19

 

From the diagrams I have, it doesn't appear that the wiring goes through even the C100 connector, but directly through the "C" connector at the ECM to the TPS. From having modified some factory TPI harnesses, I do know that there are splices in some circuits right in the harness, but those are usually soldered and insulated.

It would be useful to closely inspect the ECM and TPS connectors, especially the crimps. Any corrosion there is going to be a problem.

 

Do you have 5v at the MAP sensor? Should be on the same 5v supply from the ECM. There are no fuses - circuit protection is provided by the ECM itself.

GD

 

There appears to be two threads going on this issue...

https://www.thirdgen.org/forums/dfi-...2-tpi-ecm.html

Should consolidate into one.

 

thanks, that's what I needed. I'll check the 5v at the map and post back here.

 

yeah I'll go ahead and do that

 

I've had problems with my 5V reference voltage being too low at the TPS. My coolant temp sensor also showed 30 degrees C when it was 6 degrees outside.

The problem was the ECM, the 5V reference circuit was bad. Replaced ECM = all good after that.

T/A GTA -88 with 1227165 ECM

 

yeah I thought I had the same issue, but I have a solid 5v at the ecm at all times. It only drops out at the sensor. Did you get a reading directly off your ecm connector? I'm pretty sure if mine was bad I wouldn't get a 5v at the ecm when the issue happens.

 

This was several years ago, but I think I measured backprobing at the TPS connector with TPS connected. Can not remember if i measured at the ECM unfortunately.