MSefi.com read only forum (Jan.15,2018)

So I've gone through something like five TPS units since I started this project. Keep in mind that with the issues (non MS related) this car has had that keep it off the road, this means that the longest life I got from one was 1000 miles - the rest died after only a few days of testing in the driveway.

I thought that maybe it was the wiring that was preventing the signal from getting through but I've tested every single unit that stopped working with a VOM per the megamanual and get ZERO response. They're actually dead. And every time I hook up a new one, it works beautifully, no noise in the logs or anything.

The TPS units have all been shielded from water, and the engine bay doesn't get THAT hot - I've touched the sensor after the engine was warmed up and it's not even warm to the touch. I don't know what else to look for. I'm using GM lever style sensors like this one: The ground for the circuit is grounded to pin 19 per the megamnual along with the CLT and IAT sensor grounds. The MS and LC-1 circuit is grounded directly to the battery.

What could be killing them? I wasn't surprised when I lost a junkyard unit, but the rest have all been brand new from different sources. I've had them die both using the regular code and after I switched to Extra some time ago.

In the meantime, I have a related issue. I'm using TunerStudio at the moment and can't seem to figure out how to disable TPS input either by hardware or software. Without a working TPS installed, the TPS reading is wildly erratic and unpredictable. Using Speed Density, the erratic TPS readings from having no incoming signal shouldn't affect the fueling equation (accel enrichment is set to 100% MAP control) but it does make the car VERY hard to start when the ECU thinks the throttle is at 125% and goes into flood clear mode every other cycle. Is there a way to temporarily bypass the TPS wiring so that this isn't an issue - i.e. it reads a steady signal so I can set that as 0% throttle and stop trying to clear the flood condition? I'm thinking a jumper or something across the connector, but I haven't been able to get that to work.

AS,

About the only things I can think of that might kill a TPS are if it is overtightened, or is overextended (i.e. the throttle wants to move it further than it should go).

To disable a TPS signal, ground the signal wire (there is a 1.0K Ohm internal resistor on the input). That will tell the CPU that the throttle is always closed, so you won't have flood clear or TPSdot accel enrichment.

Lance.

Fair enough. I thought that would work, but the last time I tried grounding the pins (couldn't remember which one so I just tried all three one at a time) all three of them triggered a click from the injectors and the MS unit reset.

Incidentally, I also should revise my earlier assertion - the most recent unit is NOT dead after all! I was using a different VOM last time I tested it - I just tried it on another one and it does show varying resistance as I adjust the lever, so it's actually still good. Doesn't explain the other four that died but at least I'm not out the money for a new one again.

I know I didn't overextend any of the units, since the throttle travel is less than what they can handle, and I'm pretty sure I didn't overtighten them either.

So now I guess this means I have to test the length of the TPS circuit and make sure there isn't a break somewhere...

AS,

Yeah, if you ground Vref (the 5 Volt supply), that will definitely cause issues with the controller. The signal pin you want to ground is DB-37 pin #22. Leave the Vref from pin #26 unconnected and insulated (so it doesn't short out accidentally).

Lance.

Okay, I tested out the entire TPS harness. It all tests out fine, nothing weird or intermittent.

However, when I hook up the sensor, it still reads wild and crazy voltages. I've noticed that if I spend a few minutes stretching the wires in various directions, it behaves normally for about ten seconds and then freaks out again.

I'm going to try making jumpers from the TPS pins to the harness connector, and see if that makes it go away. It seems like the harness connector isn't seating deep enough to make consistent contact. Anyone ever heard of this happening?

Lance, I was able to successfully ground the signal pin and remove the sensor from the equation, but now that I know my sensor is still working I'd like to be done with working on it and have a working system.

The only way you can burn up a TPS pot is by wiring up the 5volts wrong. Use a ohm meter and measure between the pins, the two pins that read about 5K ohms and don't change when the pot arm is moved, is where the 5 volts and ground should connect. If the 5 volts is connected to the wiper on the TPS pot this will cause the TPS pot to burn up.