Viable Option for TPS Replacement

[Tom]

Back to the TPS... @johnb can you tell what logic/criteria the KLR uses to trigger the WOT signal/maps?All the literature says it triggers at ~65%, which is fine, but the factory spec for the TPS pot is 2.7k to 4.7k ohms at full throttle. With such a wide range, ~65% of what? Assuming the KLR is using the pot as a voltage divider and doing an ADC on the resulting voltage, it seems like that could be a big swing -- with some cars triggering at 60% (or less) and others at 70% (or more) just depending on how high their particular pot goes. I need to test to make sure the pot is linear too, but either way it will be helpful to know the actual trigger logic to make sure any replacement TPS will work right. Any insights?

[johnb]

Back to the TPS... @johnb can you tell what logic/criteria the KLR uses to trigger the WOT signal/maps?All the literature says it triggers at ~65%, which is fine, but the factory spec for the TPS pot is 2.7k to 4.7k ohms at full throttle. With such a wide range, ~65% of what? Assuming the KLR is using the pot as a voltage divider and doing an ADC on the resulting voltage, it seems like that could be a big swing -- with some cars triggering at 60% (or less) and others at 70% (or more) just depending on how high their particular pot goes. I need to test to make sure the pot is linear too, but either way it will be helpful to know the actual trigger logic to make sure any replacement TPS will work right. Any insights?

It is indeed wired as a voltage divider, but that means that it always divides the supply voltage according to the ratio of the 2 resistances. So in theory the overall resistance of the TPS pot can vary without having any effect on the relationship between wiper position and output voltage, as long as the supply is constant.

Actually though, the supply isn't constant because the TPS resistance isn't that high, so there is a voltage drop on the supply that varies by throttle angle. The KLR reads both the supply voltage and the wiper position through separate ADC channels, and always calculates throttle position as a ratio of the two. So it's always 65 degrees regardless of the pot variation. That's actually around 72% of the available 90 degrees, so sources that say 65% are mistaken.

I wrote about this recently: https://jhnbyrn.github.io/951-KLR-PAGES ... ssing.html

...but it might have too much detailed info for practical purposes

[danmartinic]

If the goal is to replicate the TPS... due in part to it being out of production and extreme cost etc... could we not just buy the more readily available units for BMW and others and transfer the inside parts into our original housings?

I am assuming of course that those printed boards are the same inside and the incompatibility lies with mounting or number of output pins etc

[Tom]

If the goal is to replicate the TPS... due in part to it being out of production and extreme cost etc... could we not just buy the more readily available units for BMW and others and transfer the inside parts into our original housings?

I am assuming of course that those printed boards are the same inside and the incompatibility lies with mounting or number of output pins etc

It's possible, but finding a BMW part with the same internals would be the trick, if such a thing exists. If you know of one, I'm all ears. The pot traces on the 951 TPS look designed in place for that case, and the idle contact switch is either NLA or not off the shelf. Bosch may have used the same parts in other units, I really don't know, but without opening a lot of units, I wouldn't know where to start. My plan was to find off-the-shelf parts that are electronically compatible with the originals, and design a new case around them... That seems like the path of least resistance unless someone knows of another TPS with internals that would transfer over like that...

[danmartinic]

The trace board has a part number on it.. and there are only three connections to it? Perhaps one of the 3-pin connection TPSs have the same board (if any work as potentiometer like the 951s)

img_2362_b4775e683b667bf476600d0455c22d6a0e127ca5.jpeg (85.44 KiB) Viewed 371 times

How about a call out to BMW and Bosch car owners from the 80s & 90s to open up their TPSs

And I suppose there is no way to re-plate those traces?

[pizatski]

Back to the TPS... @johnb can you tell what logic/criteria the KLR uses to trigger the WOT signal/maps?All the literature says it triggers at ~65%, which is fine, but the factory spec for the TPS pot is 2.7k to 4.7k ohms at full throttle. With such a wide range, ~65% of what? Assuming the KLR is using the pot as a voltage divider and doing an ADC on the resulting voltage, it seems like that could be a big swing -- with some cars triggering at 60% (or less) and others at 70% (or more) just depending on how high their particular pot goes. I need to test to make sure the pot is linear too, but either way it will be helpful to know the actual trigger logic to make sure any replacement TPS will work right. Any insights?

FWIW, I have two of the 0 280 120 431 units sitting on my desk right now. I bolted one to the Turbo throttle body a couple of weeks ago to test it out and it both fit and worked. When I scoped the WOT signal from the KLR to the DME, it would indeed trigger with the pedal to the floor and the motor running. The idle switch worked as well. What I didn't get to test, though, is at what specific angle the WOT signal would trigger, but also with such a wide range of resistances does it matter as long as it triggers when the throttle is, indeed, wide open?

[Tom]

Back to the TPS... @johnb can you tell what logic/criteria the KLR uses to trigger the WOT signal/maps?All the literature says it triggers at ~65%, which is fine, but the factory spec for the TPS pot is 2.7k to 4.7k ohms at full throttle. With such a wide range, ~65% of what? Assuming the KLR is using the pot as a voltage divider and doing an ADC on the resulting voltage, it seems like that could be a big swing -- with some cars triggering at 60% (or less) and others at 70% (or more) just depending on how high their particular pot goes. I need to test to make sure the pot is linear too, but either way it will be helpful to know the actual trigger logic to make sure any replacement TPS will work right. Any insights?

FWIW, I have two of the 0 280 120 431 units sitting on my desk right now. I bolted one to the Turbo throttle body a couple of weeks ago to test it out and it both fit and worked. When I scoped the WOT signal from the KLR to the DME, it would indeed trigger with the pedal to the floor and the motor running. The idle switch worked as well. What I didn't get to test, though, is at what specific angle the WOT signal would trigger, but also with such a wide range of resistances does it matter as long as it triggers when the throttle is, indeed, wide open?

Great info!! What is the resistance range -- how many ohms when closed vs wide open. Based on Johnb's explanation above, I'm not sure it's critical but if it's in the same range, it would remove all doubt. Does the factory 951 connector fit onto it?

[pizatski]

Back to the TPS... @johnb can you tell what logic/criteria the KLR uses to trigger the WOT signal/maps?All the literature says it triggers at ~65%, which is fine, but the factory spec for the TPS pot is 2.7k to 4.7k ohms at full throttle. With such a wide range, ~65% of what? Assuming the KLR is using the pot as a voltage divider and doing an ADC on the resulting voltage, it seems like that could be a big swing -- with some cars triggering at 60% (or less) and others at 70% (or more) just depending on how high their particular pot goes. I need to test to make sure the pot is linear too, but either way it will be helpful to know the actual trigger logic to make sure any replacement TPS will work right. Any insights?

FWIW, I have two of the 0 280 120 431 units sitting on my desk right now. I bolted one to the Turbo throttle body a couple of weeks ago to test it out and it both fit and worked. When I scoped the WOT signal from the KLR to the DME, it would indeed trigger with the pedal to the floor and the motor running. The idle switch worked as well. What I didn't get to test, though, is at what specific angle the WOT signal would trigger, but also with such a wide range of resistances does it matter as long as it triggers when the throttle is, indeed, wide open?

Great info!! What is the resistance range -- how many ohms when closed vs wide open. Based on Johnb's explanation above, I'm not sure it's critical but if it's in the same range, it would remove all doubt. Does the factory 951 connector fit onto it?

The form factor is identical to the 944 Turbo factory TPS - connector fits, same direction of rotation, fits the TPS correctly, the pins are numbered the same, and do the same things according to the tests I've run and the various pieces of literature on these units:
https://forums.quattroworld.com/s4s6/msgs/20405.phtml
http://www.sjmautotechnik.com/trouble_s ... throt.html
http://www.s2central.com/ds_F60_G69.html

The difference is indeed the resistance on what is the Bosch G69 potentiometer (wonky internal designation, I guess).

At idle on the Audi Unit on my desk right now, it is 1k ohm at idle, 2.8k ohm wide open tested across pins 2 and 3. The wide open reading is the low end of the 944 Turbo range and would translate to about 95% throttle.

When I had this TPS temporarily installed on the car it started, ran, idled, and the scope showed the WOT signal to the DME was as expected when I stabbed the throttle pedal to the floor and let off. I tested this with the F9Tech DME case open and scope leads attached across pin 3 and ground checking voltage.

[pizatski]

I'll add one more thing for the original poster...

This person offers a TPS repair service, transaction through ebay - https://www.ebay.com/itm/197603630834

I don't know what precisely is or is not repairable on the TPS, but the whole reason I ordered one of the Audi TPS sensors is because my old one was at their shop getting tested and I'm impatient. My TPS ended up not needing a repair (I performed the test procedure incorrectly) and was sent back with a full refund, so the person running that shop is at least honest and may be worth a shot to get one that is really broken repaired.

[chrischrischris]

I think that repair person is George Beuselinck. Long time 944 legend. He gives advice to people a lot on Pelican forum and also provides feedback to 944 questions for the PCA.

There's another active thread on the forum about an exciting plug-play "classic ecu" offered by a vendor in Germany. What's relevant to this discussion is it looks like he offers the option for an upgraded TPS that integrates with the new ecu.