TPS works when not running, doesn't work when running

[beast02]

Ok, I seem to have a whacky TPS sensor. I can calibrate it and get a low value when closed, high when open (0 and 426 are the values I believe). The TPS registers fine in tunerstudio when I move the throttle. THEN, it all changes when the car is running. I am in the early start up tuning stage right now, running ITB's on Alpha-N, so I REALLY need this TPS value! As I try to tune off idle, it of course gets nice and lean as it gets more air but doesn't register throttle movement.

I have attached a data-log where you can see this, I believe the car is fairly warm at this point. Near the end of it, you can see where the engine is off, but now you can see actual throttle movement.....

Hope you can help!

Thanks,

Bryan

[Matt Cramer]

Where is the TPS getting power and ground?

[beast02]

It's wired up per the DIY autotune harness, grounded at the megasquirt.

-Bryan

[Matt Cramer]

What happens to the TPS voltage and VREF voltage with the engine running?

[beast02]

Where would I get these values, at the megasquirt pin within the board with case off?

Or just unplug the TPS from the pigtail and take the value at the end of the pigtail (obviously can't alter the TPS position if it's unplugged)

[Matt Cramer]

You could probe the MegaSquirt, but I was thinking more back probing the sensor.

[beast02]

Ok, tested, both values are 0 when the TPS is closed and not running but key on, when I open the TPS signal goes up in V, got up the .5v (didn't open all the way). Vref was 0 for this.

When I start the car, the V was at .1V, and then didn't change on the TPS signal as I opened the throttle (didn't get a chance to check the Vref but assumed it was still 0)

-Bryan

[Matt Cramer]

Vref was 0 for this.

You will need to investigate what is pulling VREF to zero.

[beast02]

I lied, there is 5V at the 5vref wire. I am an electrical tard (good project for me eh?) and was grounding to the chassis...

Once grounded to the sensor ground, I had 5V. I'll do a comprehensive test tonight when I can get the car running etc and figure out the differences.

Thinking this possibly is a Signal and 5Vref wire swap issue?

-Bryan

[DaveEFI]

I lied, there is 5V at the 5vref wire. I am an electrical tard (good project for me eh?) and was grounding to the chassis...

Once grounded to the sensor ground, I had 5V. I'll do a comprehensive test tonight when I can get the car running etc and figure out the differences.

Thinking this possibly is a Signal and 5Vref wire swap issue?

-Bryan

You should still have seen 5v (approx) using the chassis as a ground. The sensor ground and chassis grounds are connected together inside MS.

[beast02]

Results as follows (key on, engine off).

Source to ground: 5ish volts closed, goes down as i open the throttle to about 4.5v

Vref to ground: stays 5.5 volts as i open throttle

[Matt Cramer]

What do you mean by "source"?

[beast02]

What do you mean by "source"?

I meant signal, TPS signal wire

[DaveEFI]

Results as follows (key on, engine off).

Source to ground: 5ish volts closed, goes down as i open the throttle to about 4.5v

Vref to ground: stays 5.5 volts as i open throttle

That sounds like there is no ground connection to the TPS track.

[beast02]

The resistance between the VREF and ground pins will remain constant.
The resistance between the ground and signal pins will be low with the throttle closed and high with the throttle wide open.
The resistance between the VREF and signal pins will be high with the throttle closed and low with the throttle wide open.



Without the key on, i tested the resistance. All of it is correct except for the last statement, the Vref and signal start low and get higher. BUT, the statements above it are true. Herm, does this point to anything?


Im not thinking the ground is an issue because the other sensors which ground the same are working fine

[beast02]

I think I have figured it out. I believe the TPS is actually working when the engine is running, I think the issue is there seems to be an amount of throttle travel before any movement registers in the tunerstudio. I was looking in to grounding stuff when I noticed this after re-calibrating the TPS. WHy I think this is I can push the gas pedal down a bit and it still shows 0% (While engine is off), then it begins to go up in value. I'm guessing I never reached that threshold when the engine was on due to a) the engine reving up with the opening, b) the mixure obviously leaned out with more air and not any more gas.

SO, now, I either need to figure out if there is a TPS sensor issue itself, whether there is free play between the throttle shaft and the sensor itself, or if I can dial in the resolution of the TPS a bit. ADC count reads 0 and 467-ish

-Bryan

[beast02]

I think I figured the issue out. Since this is a completely custom set-up with a GM TPS sensor adapted to use on GSXR750 ITB's, it's easily changeable. Hopefully it's a future lesson for someone making their out set-up.

It looks like the issue was the TPS was positioned with the sensor at absolute 0, which as I have discovered takes some degree of movement of the shaft/sensor itself before it registers a Voltage change in the megasquirt.

So, my fix will be to re-grind the adapter insert from the shaft to the sensor itself so that the TPS closed position registers an ADC count above 0! WOOO HOO.

-Bryan