Part Throttle Run On

[SPN]

Just tuned a 2003 K2500 8.1 with great results. However, upon releasing the throttle after a 60-70% TPS at 40-55MPH pull, the vehicle continues to accelerate (for 1-2 seconds) as though throttle is still being applied. I did isolation test the TPS and it did show some minor irregularity so I may replace it. Just wondered if anyone had similar experience.

[JamesLinder]

Sounds like too much throttle follower air; to begin with, zero out all throttle follower delay (in gear) above 2 MPH. Leave OEM
delay for 0 and 2 mph speed cells, but zero out all the rest. Make certain that you have not lowered the OEM throttle follower
decay values, and make certain that the throttle follower airflow and airflow multiplier entries are not greater than the OEM
values. With an automatic transmission with stall converter you can increase the decays for 3rd and 4th values quite a bit; but
I would not try this until I zero out the delays described above first. If you increased you IAC Park values too much this can
cause similar problems too, but it is typically at speeds of 40 mph and less where the car wants to continue driving itself
instead of slowing down. If you have increased the IAC steps in the lower part of the curve this can also cause problems like
this. If you have raised the low load timing in the High Octane table along the first 2 or 3 rows (lower gram/cylinder rows)
this can also cause this type of problem. Probably best you post your tune and a log, or at least your tune so we can see if
anything obvious stands out.

[JamesLinder]

Yes, my TPS at times will hold at 2% after flashing a new tune and I have to pull the PCM fuses to get it to
return to zero like it should. Yesterday I ordered a new GM TPS to replace the parts store replacement I
installed a couple of years ago; we are finding that sensors made by after-market companies often are
not reliable. This is true for MAP, ECT, TPS, CAM, and RPM sensors. I recommend using genuine GM
parts when replacing sensors because they are the most important components you have working for you
to obtain a "good" tune - otherwise, you are not getting accurate input measurements for the PCM to use
for making feed-forward corrections. I guess I would not compromise on oxygen sensors either if you are
running open loop, which I only recommend if you find the car performs just as well open loop as closed
loop after you calibrate VEs and MAF with wide band oxygen sensor. When you calibrate using lambda
instead of AFR, then it does not matter what the precise ethanol content your gasoline is every time you
fill up; your chosen lambda (that works best for your vehicle) will always prevail and provide the exact
correct power enrichment regardless of the precise AFR optimal for the ethanol content of the fuel you have
in the tank. I would assume that the variable ethanol content might be the reason the OEM narrow band
oxygen sensors sometimes do not seem to target the optimal LTFT values. Either that or the long-tube
headers place them too far back to work correctly with the PCM - too much delay so that the oxygen
content is not truly representative of what is burning in the cylinder at the time it is "seen" by the PCM.