Inconsistent Virtual Dyno Runs, Same Tune
 

Been tuning AFR and such. I just finished fabbing up my 3" exhaust system so I was curious about the gainz. Heres where stuff starts getting weird. Keep in mind, I'm on Brain's basemap as far as the spark map goes, haven't touched it. I've also been running my Chinese ebay "interheater" so IAT temps kinda skyrocket.

Did a pull on wastegate pressure, near-empty tank of 91 in the morning, ambient was like 70F. Car made about 170whp. Did a pull at night around the same air temps with a near full tank of 91. Car made about 150whp. Only thing I did in between pulls was get gas, and fix a boost leak. One thing I do notice is that the run with less power, made more boost, and the run with more made less boost so the run with less boost got more timing. Is that the reason? Another thing that changed after the 170 run is I recalibrated the afr table to try and match my gauge and TS, but I don't think that has anything to do with it.
AFR with 10psi i saw on the gauge was like 11.8-12 while 11psi I saw 11.5-11.8 not sure if the logs lined up with that.

The boost leak in question is weird as well. The threaded holes for the wastegate actuator bracket on the compressor housing aren't being used as the actuator is mounted on the turbine side of the turbo. So the holes for the compressor actuator bracket are left empty, weird thing is, one of those holes is an actual hole into the compressor housing which I guess makes it a boost leak? intake leak? not sure if that has anything to do with the weird readings. I just put a bolt in the hole to "fix" the boost leak post-168whp's run.
https://cimg9.ibsrv.net/gimg/www.mia...286daf02a4.jpg

Logs and tune is attached, thanks yall.

EDIT: Forgot to list specs. Kraken manifold/ 3in DP, GT 2554r, full 3in catless exhaust with Dynomax race bullet. 2" hotside 2.5" coldside piping. Last compression test on the motor was all 150psi even

One thing with virtual dyno is you need pretty much the same stretch of road to compare pulls. Were the pulls on the same road?

Same stretched of road except in opposite directions from each other. Google Earth put it at Grade 0% with a 10ft elevation change over like 400ft so it's pretty flat.

Ah that's good

10ft over 400ft is like a 2% grade. Definitely not flat.

Just got back from another couple runs. I think you're right the roads I'm on ain't flat.

I did 2 identical runs in 2nd gear starting from 2k rpm. Going "up" the road put me at like 140whp 10psi, going "down" put me at 160whp at 10psi. Guess the roads I'm using arent the best... :facepalm: The road going "down" also goes under a bridge so air temps get a few degrees cooler than going "up". Guess I'll see if I can find a flat road and do some tests there.

I mean if you consistently get one number up and one number down, its safe enough to kinda average them, since its a virtual dyno anyway. Probably making around 150 I'd guess.

:vash:Damn that's kinda weak. But you're probably right... I'll work on finding a straight and flat section of road....

Don't worry about the absolute numbers. IME, it's really hard to get virtual dyno to generate numbers that are consistent with physical dynos. Lower gears generate less power due to flywheel effect, higher gears generate a lot of drag due to speeds which is difficult to calculate out because drag coefficient varies a lot car to car and there's no real way to measure it.
​​​​​
If you're careful to keep everything consistent between runs and can find the right road (minimal traffic, no "official" attention), then it's possible to generate numbers that are sufficiently consistent with each other to tune the car. If you want to compare your car's power to anyone else's car, though, you need to find a real dyno.

--Ian

Hah thanks for making me feel better :rofl:

Yeah I guess I'll just use VD just to calculate differences in the tune between runs and such. Time to go find a good road.

You should always do 2 way pulls and then average. Accounts for slope and wind. The flatter the better.

Yeah I saw people recommending to do that. Good ideas, thank you.