Another TPS/ground issue MS3x and LC-2
 

I am having TPS/ground issues too

I basically narrowed my problem down to my LC-2 wideband. There is no separate sensor ground on the LC-2 and I think it causes a ground loop or voltage difference.

My TPS will read correctly everywhere EXCEPT in overrun. When in overrun my TPS % drops to 0 and then holds...UNTIL the overrun fuel cut kicks on then the wideband reads 22.2 and boom the TPS starts fluctuating from 0-6%. Get back on the gas and the TPS holds steady at the right %. At idle it holds steady at 0%. Only when the wideband reads 22.2 does the TPS start acting up.

It's actually better now, before when the wideband read 22.2 the TPS would jump to 0-15% causing the overrun fuel cut to turn off, inject fuel, then wideband reads normal fuel range and the TPS goes back to 0%, then overrun fuel cut cuts fuel, causing the wb to read 22.2, causing the TPS to jump to 15%, causing the computer to inject fuel etc etc, creating a nasty feedback loop.

Also when I first start up the TPS will read something like -6% and then as soon as the wideband warms up and starts working, the TPS jumps to 0% and holds steady.

Anyway I basically narrowed it down to the LC-2 because I disconnected the LC-2 from the car/ECU and my TPS held perfectly solid at 0% when in overrun. Combining sensor and chassis grounds also makes it worse. The kicker is I never had this problem with my MS1 using all the exact same hardware, but now I have the problem on my MS3x.

Where does everyone get 12v and ground from for their LC-2? Currently I pull 12v from the OEM O2 sensor plug and ground at the chassis ground point of my MS3x. I'm pulling the gauge headlight-dimmer source from somewhere under the dash, not sure if that matters at all. I also think it's a bad idea to ground the LC-2 only to the sensor ground location of my MS3x.

Here's my tune

I tried multiple TPS sensors with the same results, forgot to mention that. I will do a log asap hopefully tonight if the weather holds.

what is the TPS lag factor set to?

Currently set to 20.

Here is a log of a start-up, you should notice the TPS snaps to about 0% at the same time the wideband comes alive. Rain and hail and nastiness right now so I'm keeping it in the garage for now.

Here is a log of accelerating to 4th gear and then coasting to a red light in overrun. During overrun the car went into the feedback loop I described.

Try grounding the LC2 to the same point as the ECU - the cylinder head if you're running it through the stock harness.

That's where it was grounded. Just to be sure tonight I pulled all the LC-2 wiring that had been in there for about 5 years and ran new wires, cleaned up the ECU ground point and bolt and use conductive grease and the problem remains just the same. I recently uncovered some mistakes in the wiring harness from the PO of my MS3x and built a whole new wiring harness from the MS3 to the car's loom. I'm wondering if they messed up something inside. What in the MS3 should I be looking at for clues?

Nobody has ever had this issue? No things to try?

I am currently having a similar issue. I have a MS2 and a Innovate MTX-L wideband. Did you ever sort this out?

My TPS reading fluctuates a lot as well when I am having the issue, which makes my car idle like dog shit.
Today when it happened, I turned on my headlights, and the TPS reading went back to normal from doing that. Maybe the toyota itb TPS I run is very sensitive to heat?

I am suspicious of it being my TPS or grounding to my TPS

Sort of. I changed turbo set ups and in the process moved the LC-2 module to a new spot. I also had the excess wiring looped up, but now it's been straightened out. The problem remains, but is no longer as severe. My TPS now only fluctuates sometimes, and only 2-4% so about half. I think it may be related to heat on the LC-2 module or the length of the excess wire. I plan on replacing the O2 sensor soon and at the same time I want to move the LC-2 module inside the cabin and completely re-wire it. I will simply cut out any excess wire also so it is only as long as it needs to be.