No TPS, but TPS "creep"?

[opethmike]

Don't have a log handy at the moment to demonstrate the problem, but I have a funny one -

My setup is ECU -> Relay board -> sensors

I recently removed the throttle position sensor from my car, including the wiring from the relay board. But, for some reason, I'll see the TPS % in Tuner Studio slowly creeping up. Uh..... shouldn't it just plain do nothing?

[Spittybug]

Don't have a log handy at the moment to demonstrate the problem, but I have a funny one -

My setup is ECU -> Relay board -> sensors

I recently removed the throttle position sensor from my car, including the wiring from the relay board. But, for some reason, I'll see the TPS % in Tuner Studio slowly creeping up. Uh..... shouldn't it just plain do nothing?

Weird. I vaguely recall seeing something in the MS boards about grounding it even if unused..... just a thought.
Why did you remove it? Have you got AE working well with just MAPdot?

[opethmike]

I actually don't use AE at all. I've had numerous issues with TPS noise, regardless of what I did. Several TPS units used, wiring routes tried, etc, but I still received funky signals.

I only used the TPS signal to activate/de-activate the ignition vacuum advance solenoid. But the sporadic TPS signals would cause it to turn on/off at times it shouldn't, which obviously caused all sorts of running issues.

So after I realized (duh) that I could just turn the vacuum solenoid on/off based on RPM, I got rid of the TPS. I have no interest in going back to having one.

The grounding idea makes sense. I'll give it a shot and see what happens.

[Spittybug]

Don't give up on AE, it makes a difference!

[dmcjohn]

Are you sure you're using the right cover sheet for the TPS?

If you could go ahead and try to remember to do that, that would be great... Mmmk?

[dmc6960]

yes, it creeps up if nothing is connected to it.

[opethmike]

yes, it creeps up if nothing is connected to it.

Is it possible to stop it from doing that? Or some way to stop it from causing an erroneous flood clear condition?

[dmc6960]

Is it possible to stop it from doing that? Or some way to stop it from causing an erroneous flood clear condition?

Grounding the TPS input should accomplish that. Pin 22 of the DB-37.

[opethmike]

Grounding the TPS input should accomplish that. Pin 22 of the DB-37.

Thank you.

[Spittybug]

Have any of you guys tried/looked at my timing calculations in the forum? Would be interested in getting your feedback.

[opethmike]

Jim,

That solved my issue. Thank you very much.

Owen,

I've looked at it, but my understanding of timing is VERY rudimentary, so I'm afraid I can't offer much useful feedback :(

[jangell]

In case you're wondering why you had to ground it, and why you got weird values when it wasn't, I'll throw in my novice electronics knowledge here: Basically, the MegaSquirt expects the input to be between 0v and 12v. When it's not connected to anything, the value is "floating", meaning that its value is undefined (I think it actually acts as a radio antenna and generates slight voltages from random RF, but I could be wrong). This shows up as seemingly random values in the signal. Floating inputs are generally considered bad for this reason. Tying it to ground (i.e. "pulling it down"; "pulling up" would be tying it to 12v) ensures it always has a constant value of 0v.

I'm sure someone actually knowledgable like bitsyncmaster would be able to provide a more accurate answer than this.


Incidentally, I'm hoping to start my EFI conversion some time in October; I got rails from Owen and fittings from JEGS, but I need to save up a bit more before I can order the MS, sensors, fuel regulator and whatever else I need. I was thinking of going with a TPS, but I'm not really sure.

-- Joe

[Bitsyncmaster]

The input impedance on most microprocessor analog input pins is quite high (> 50 Kohms). So the humidity in the air can change what value of charge an open input pin will see. Of course the car has a lot of wiring harness and high impedance will pick up and charge high impedance with different voltages.

Normally the designer would design a jumper to ground the pin if not used. Or he can option via a user function to disable that part of the software to alway load zero. Another option the designer can take is program the input pin as an output if disabled.

It's still better to not leave input pins floating. Voltages greater than the VCC or less than ground can feed through the micro and affect other hardware inputs.

[dmc6960]

Basically, the MegaSquirt expects the input to be between 0v and 12v.

Almost. The TPS POT runs off of a 5V reference voltage supplied by the ECU. Other external sensors (air and engine temp) are dual wire, and run from a reference ground in the ECU, to a calibrated pullup circuit to give and approximate input range of 0-5v for them.

[dmc6960]

Have any of you guys tried/looked at my timing calculations in the forum? Would be interested in getting your feedback.

I'll get there eventually...