Fuzzy TPS

[ErnieJones]

I just finished off my Microsquirt Module install (the hardware side) and so I began following all the steps in the Megasquirt II Setting Up guide.

In section 2.16 it tells you to check the sensors while cranking and to watch for gauge movement and that CTS and IAT gauges should not be jerking around or else there's a ground issue. I'm all good except for one thing....

With the ignition on I can see that the TPS gage moves a bit. Not much, just a little. It's fuzzy. I then did a cranking datalog and I can see that the TPS does jump around a little bit.

I'm also finding that I often have to re-calibrate the TPS as the low side (off idle) seems to increase a bit over time.

What causes this? The sensor is powered by VCC and Sensor ground same as with the IAT and the MAP as well as the external Baro. These are all good. It's just the TPS that's fuzzy. It's the only obvious thing wrong. What would cause that?

[DaveEFI]

Does it do the same on a stim? A TPS is normally just a pot, and can wear or be dirty and have a poor connection between wiper and track. They normally wear most at the beginning of the track. If you don't have a stim, you could buy a 5000 ohm (approx) linear pot and use that for testing. If that gives smooth readings, the chances are yours is faulty. If not worn, they can sometimes be cleaned.

[ErnieJones]

I don't have a stim but I did discover something that might help narrow it down but it also baffles me....

With the ignition on and the engine off, and the TPS connector removed - it still does it! So the bike is silent - nothing is running. No fan, no injector, no coil, no fuel pump, nothing. And yet, the TPS bounces around 0 to 5% on the gauge.

What could possibly cause that?

[DaveEFI]

Not sure what an open circuit input would do. Try shorting the input to ground, and see if that stops it happening.

[elutionsdesign]

It's not a big deal, the TPS is a RATE-OF-CHANGE measurement. Having one end or the other float a bit, yawn.

[dontz125]

It's not a big deal, the TPS is a RATE-OF-CHANGE measurement. Having one end or the other float a bit, yawn.

Yes, and if your 'rate of change' is random and has no bearing on what the engine is doing? There lies madness ...


Ernie: inspect your board for shorts, damage, and general crap in and amongst the TPS components.

[DaveEFI]

If you simply unplug the TPS from the engine loom, I'd not be surprised to get some form of random 'interference' picked up on the open circuit wiring. So either disconnect at the DB37, or short it to ground at the TPS connector as I said.

[ErnieJones]

It's not a big deal, the TPS is a RATE-OF-CHANGE measurement. Having one end or the other float a bit, yawn.

Yes, and if your 'rate of change' is random and has no bearing on what the engine is doing? There lies madness ...


Ernie: inspect your board for shorts, damage, and general crap in and amongst the TPS components.

I've been doing that and found a lot of vibration induced issues. I now know that vibration and electronics don't mix at all

I'll come back to the TPS in a second but first a quick detour as something else has cropped up that's more important. In Tuner Studio my Coolant Temp Gage is not working. In fact, it reads 0F.

All other gages powered by VCC work fine. So, I thought it was an issue with a connection to that specific sensor. I just did a ohms test right from the top of the Microsquirt module CTS pin all the way through to the CTS Sensor and it has continuity all the way through. In fact, I can read the OHMS that the sensor is putting out and it's accurate.

The sensor has been calibrated in Tuner Studio. What I don't understand is that, if their is continuity all the way from the sensor to the top of the Microsquirt Module, how is it that Tuner Studio shows zero on that gage? Is it possible that Tuner Studio gets corrupt somehow?

Edit: I can rule out Tuner Studio corruption (I think) because my desktop shows the same thing

[grom_e30]

if you get a volt meter on on the signal wires for the clt sensor you should be able to read some voltage.

[ErnieJones]

CLT fixed This turned out to be an open between the top of the Microsquirt connector and the CPU.
Fuzzy TPS fixed: This turned out to be the TPS unit itself. I replaced it with a used one with 2,300 kilometers on it and it calibrated perfectly, held it's calibration, and reads dead accurate all the way through regardless of ECU re-starts etc.

The last thing on the repair list is the external baro - I've got continuity all through on each of the 3 lines and even tracked the signal right up to the pin on ECU - still nothing. This would tell me that either my new baro just kaffed or there's an issue on the CPU side. I'm guessing something happened to the new baro.

I have another one here (now) but for the time being I can just go back to speed density instead of % baro for testing/tuning purposes.

I'm happy so far, 2 out of 3 ain't bad and I should be back to tuning tomorrow

[grom_e30]

you got that wide band sorted yet?

[ErnieJones]

you got that wide band sorted yet?

I'm still looking for a 1" stainless bung with reasonable price and shipping. I found a place with them and they wanted $110 USD for shipping. Found one on e-pay and they wanted $50 USD for shipping.

I mean, I do have a WB here, I do have a steel bung but I'd rather put a stainless one on if I'm going through the effort. But I just can't get ripped off to that extent on shipping. It's against my nature.

[grom_e30]

im pretty sure i got a 1inch stainless lambda bung free with my lc2 lol

[ErnieJones]

im pretty sure i got a 1inch stainless lambda bung free with my lc2 lol

And that helps me how???

[ErnieJones]

you got that wide band sorted yet?

I should have my 1" stainless bung any day now I managed to find one with somewhat reasonable shipping I already have the WB so...it should be just a matter of placement, drilling & welding