1987 GL-10 Wagon Throttle Position Sensor Issues

[OCDan]

I have a 1987 GL-10 which I haven’t been using and decided to gift it to a friend in need of an inexpensive vehicle. I’m helping him resolve some issues with it.

 

It’s getting a 31 code. The TPS is the 4-wire type. Without going into detail, the original TPS was irreparably damaged during removal.

Based on information posted here https://www.ultimatesubaru.org/forum/topic/153231-1987-subaru-gl-throttle-position-sensor-help/?do=findComment&comment=1283981

I located and purchased this:

https://www.partsgeek.com/dpq5225-isuzu-i-mark-throttle-position-sensor.html

with part description “Throttle Position Sensor - 1989 Isuzu i Mark - Walker 1272-06053178”

This is what it looks like:

And internally:

Box label:

 

Prior to cutting any wires, I checked resistance as described in the EA82 FSM:

And measurements were within specs. So, I took a leap of faith and cut the wires on this new TPS and on the engine side connectors and soldered in a new connector on each side.

 

I installed the TPS and double-checked the resistance on the TPS pins and everything was still within specs.

So, I went ahead and connected the TPS to the engine side connector, reinstalled the components and belts which had been removed to gain access to the TPS screws. The engine started on the first attempt and was running smoothly. After it had been running for a few minutes, the RPMs suddenly jumped from around 1000 to around 2000. At some point, the CEL came on. I looked at the ECM light to see that it was blinking 31 again. I pulled the connector to check the resistance on the TPS. There was no longer any continuity at all on any of the pins, regardless of the throttle position.

At this point, I don’t know if the part just failed due to some defect, or if it was somehow damaged by some condition when the engine was running. Regardless, it can’t be returned to PartsGeek, due to my having cut off the original connector.

I’m fairly certain that I matched wires correctly between the 2 sides. The engine side harness wires have the following functions:

I measured the voltage on the red and it is 5 volts.

Red goes to red, black to black, green on the engine side to white on the TPS side and blue/green to blue, for the idle switch.

I’m hoping someone might have some ideas on how I can troubleshoot this.

Can you think of any scenario which could have caused the new TPS to fail?

Without knowing, I’m reluctant to plunk down another $210. If I do go that route, I’ll temporarily use its existing connector with alligator clips on its tabs. That way, if it doesn’t work, I could return it.

I would appreciate any advice!

[azdave]

I don't see what you could have done wrong if all your readings were as you say. It a simple potentiometer setup with an on/off switch on the side to let the ECU positively know when the throttle is in the full idle position. If you currently have no continuity between pins B&D, unlike when you tested it new, that would indicate that the carbon trace on the board had burned in half or was separated somehow. If you now connect the Ohm meter between pins B &C and then turn the shaft from stop to stop, do you ever measure anything but an open connection as the wiper sweeps around the carbon trace? 

The 5 volt signal should not be able to supply enough current to damage the sensor. Even if that supply wire measured 12 volts, it would barely generate 0.003 amps (0.04 watts) which should easily be handled by the carbon wiper.

I don't know what kind of time schedule you are under but I can offer to bench test the TPS and see if I can determine what failed. Not much to lose except for the time to mail it back and forth and the postage cost. I myself would not risk another new TPS until I had a better idea what failed. 

 

 

 

[OCDan]

5 hours ago, azdave said:

If you now connect the Ohm meter between pins B &C and then turn the shaft from stop to stop, do you ever measure anything but an open connection as the wiper sweeps around the carbon trace? 

Thanks for your reply azdave!

No, the connection stays open through the whole sweep. Thanks for the offer to bench test it. What would that involve? Just wondering if, with guidance, it's something I could do. The FSM Fuel Pump Circuit diagram contains this section on the TPS. Presumably, the new TPS circuit is the same...

I do have a good multimeter.

Thanks again!

Edited Wednesday at 06:48 PM by OCDan
Additional comment

[azdave]

3 hours ago, OCDan said:

Thanks for the offer to bench test it. What would that involve?

I offered to help, not knowing your electronics skills but maybe you can figure it out with careful inspection. I would basically take it apart and find out if it failed due to a mechanical issue (like a hidden crack across the carbon path) or if it failed due to somehow suffering from an over-current condition. Your test results seem to indicate it failed in more than one way. I would be looking very closely for any hint of heat damage or circuit board burns. Also look for cold solder joints and micro cracks in any circuit traces. I will sometimes heat and cool a circuit to look for failures caused by expansion and contraction.

You say you get no continuity on any of the pins after the new TPS failed. Using the diagrams immediately above as reference, it does not make sense that you read nothing across R-B, nor across R-G or G-B. One or two of those paths should still be reading some type of resistance whether the part overheated or cracked. Very unusual to read an open circuit on any combination across the R,G, or B wires. That would have to mean the carbon path failed as well as the wiper on the G-wire. The connection between the B-LG wires we don't care about right now because it is a simple on/off switch and you can physically see if that contact is working at idle.

I had TPS issues when I first got my EA82 SPFI 87 wagon 3 years ago, but I was able to clean it up and make it work. I'm really temped to buy one of the cheap SERA484-10 TPS modules on Amazon, just to learn if it can be configured to work in our application. I see one example online where a person said it worked on their EA82 but yet another where they said it did not. I think I will order one and find out for myself. It would be good to know the answer and have a spare for myself if it is possible. I'm not concerned with connectors being plug-n-play of course.

 

Update: I went ahead and ordered one of the SERA484-10 TPS modules off eBay and should have it by the weekend to check out.

Edited Wednesday at 09:53 PM by azdave

[OCDan]

My electronics skills are pretty basic. So I really appreciate the suggestions.

I just checked for continuity across all possible pairings of the R, G and B tabs, on the inside of the module. It's open on all of them. As expected, when the contact on the idle switch is closed, LG to B shows continuity.

I'll go ahead and open it up and let you know what I find.

Crossing my fingers that the SERA484-10 can be adapted for our application!

[azdave]

18 hours ago, OCDan said:

Crossing my fingers that the SERA484-10 can be adapted for our application!

The ebay vendor just cancelled my order for the SERA484-10 TPS because it is out of stock. Looks like I'll have to order one shipped from New Jersey instead of California and wait an extra few days for it to get here.

[OCDan]

Bummer!

I tried very carefully to dismantle the failed TPS from PartsGeek.

Unfortunately, while drilling out the rivets holding the circuit board to the base, the circuit board cracked in 3 places. It seems surprisingly fragile. Other than the cracks, I don't see any places where the carbon trace is broken. Nor any sign of heat damage.

Edited 13 hours ago by OCDan