TPS/LC1 issues

[benc502]

I am running a 5.0 throttle body on my setup and am using a relatively new tps that I enlarged the holes on. I am having two issues. When I first turn the key on the tps gauge in t/s drops 5-10% while the wideband is warming up. Once the wideband warms up and shows a "real" reading the tps value goes back to somewhere around 0. The 2nd issue is that the tps value shown in t/s is very erratic (while running), it fluctuates from -5 to 5 somedays and somedays its 3-10, just kind of all over the place. Ive recently moved the ground for the tps from the ms sig return cirucuit to it being bolted to the block. this did not help, it functions the same.

One other thing you should know is that ive used logworks and t/s to get the afr reading in t/s to match what the actual wideband gauge reads. I have an LC-1 and a practically brand new sensor (only 500 miles). I had this same issue before I did any configuring in logworks/tunerstudio.

Grounds: MS is grounded to the factory ecu grounding location and a 10ga wire runs straight to the battery. Also coming off this spot is another 10ga wire going to the engine block. My battery is in the back and it is grounded to the frame in the rear with a 2/4ga wire. The engine block is also grounded to the frame in the engine compartment.

I built the ms myself and it has a 3.0 board, running ms2 extra. The tps operates correctly on the stim (in tunerstudio).

What am I doing wrong?

[Stinger]

The TPS being erratic is pretty common for a bad or loose TPS. Of course a wiring issue can cause the same issues. What does the TPS show for voltage when using an analog multimeter under these same conditions?

[wesk]

Are sensors grounded to the car somewhere, or MS? Years ago, the diagrams showed them grounded to the engine block. The best way is to pick a pin on the ground plane of the MS at the db37 connector, and terminate all sensor grounds to that pointl.

[benc502]

I wasn't able to drive the car tonight but I checked the tps voltage (green wire) when I first turned the key on and it was .80-.85. after the wideband warmed up it went to 1.05-1.10.
I tried a brand new tps for a 4cylinder (the ones with the already slotted holes) and had almost the exact same results.

All of the ms grounds go to the same spot. They are all wrapped/soldered to one 10ga wire that runs from a bolt by the ms (where the factory ecus are grounded) to the engine block. This spot also has a 10ga wire that goes to battery negative. The engine block is also grounded to the frame with a 4ga wire. None of the other sensors change values while the wideband is warming up. Ive tried moving the tps ground to a bolt on the intake manifold but I get the same results.

FYI: I have a pro-start battery from pepboys which I have already had one issue with (bad after 6 months and had to return for new). But all these tests were done with my battery charger hooked up on 2amp charge.

If I don't drive the car for a day or two I have to hook up the charger. If I do a battery drain test (meter between ground terminal of battery and ground wire) I get like 2-3 milliamps with all the doors closed and key off. Correct me if im wrong but I don't see any issue there.

Will drive car tomorrow and post results.

[benc502]

by the way the Wideband is grounded at the factory ecu grounding location as well.

[Stinger]

What MS system is this?

[benc502]

3.0 board/ms2 extra. Built by me

[Stinger]

Can you post a log of this happening and a tune that goes with the log?

[benc502]

here is zipped datalog file of me cycling the key

[benc502]

here is zipped tune

ill drive it today and get a driving datalog

[benc502]

here is zipped datalog from a 20 minute drive.
I cycled the key a couple times at the end and you can watch tps value move from 0 to 12 once the wideband warms up. while driving its not to bad but still fluctuating from 0 to 4 without foot on the pedal.
This is with a brand new tps, but the old one acted exactly the same, so im leaning towards wiring. thoughts?

[Stinger]

Looks like the LC1 is dumping high current ground where it shouldn't.

Where is the LC1 grounded relative to where the sensors and MS are grounded?

[benc502]

Problem Solved heres how:

I did redo the grounds-everything is grounded to the block and all in separate locations. Battery - to one place, wideband to other, ms to another. This seemed to help a little and I think necessary but I still had problems. So I went back and looked at the datalog again and realized I was missing something. If you look again you will see the battery voltage reading spikes intermittently throughout the drive. when it does it jumps to 18-21v or to 7v. This obviously throws off the sensors and was a sign that I had noise.

After reading several posts on msextra.com, I lifted and ran the middle legs of q9 and q12 to a 12v source outside of the ms and then put a 47uf capacitor in between the middle legs of q3 and q9. As soon as plugged the ms back in BOOM no more funky tps, no more spikes in my datalogs, no more funny numbers in my ve table because the megasquirt thinks the engine is at a certain kpa when its really at another.
car is running much better, and I can actually tune now that im getting true readings from the engine,etc.

Thanks for help

[Stinger]

Good to hear you got it worked out.