• The site migration is complete! Hopefully everything transferred properly from the multiple decades old software we were using before. If you notice any issues please let me know, thanks! Also, I'm still working on things like chatbox, etc so hopefully those will be working in the next week or two.

Virtual Dyno

well on smoothing 0 you can definitely see the spike at the end of the run ... that makes the end of that run useless. Other than that were the roads straight and without potholes and flat that you were testing on? Was it the same stretch of road you are comparing here?
 


Heres 3 I put did back to back on the same stretch of road this time.

0 smoothing

0smoothing.jpg


1 smoothing

1smoothing.jpg


2 smoothing

2smoothing.jpg
 
That blue graph definitely has a spike in it which when smoothed gets smoothed to a hump in the other graphs. That is characteristic of a bad log.
 


matt5112 said:
Also which tire diameter is this thing supposed to use? Actual or theoretical? Doesn't specify and automatically selects theoretical.
Click to expand...

This.

By actual, I mean effective. Since we have significant sidewalls, 225/60R16, the deformation makes the effective radius of the tire around 25.71-25.84 (GT lists first size, GTP lists second)

Also curious as to why there is no compensation for torque converter slip.

In that red curved I removed the last data point and it jumped back to the next high of ~330 HP. Still nowhere near accurate because of spikes.

The light green pull is the most believable curve however, does it consider air drag? Third gear covers ~84-125 mph. Significant increase in rolling resistance up there. If anything though, it seems to think there is more drag than what was seen by the vehicle.

Based on how sensitive this program is to bad data, I'm not sure it is worth using with our logs as they all seem to be error ridden.
 
Last edited:
There is not mathematical way that Im aware of for compensating for a torque converter. It would be complete guessing. Its the same as a slipping clutch how do you guess how much power is being lost there.

Wind resistance is taken into account and added as drag hp. You can check the frontal area and drag coefficient for your vehicle by looking in File -> Car Editor.

Virtual Dyno has been proven to be very accurate on 14 other forums.

Also you arent deforming your tire that much. Not enough to make a difference. These differences are from data in the logs. The issue with tires that you mention is a valid point in dragsters but not in your car.
 
Do WOT logs in a gear where the TC is already locked and turn off the A/C to keep the TC locked.

Bad data in, bad results.


Tapadatass
 
The part about bad data in bad data out is VERY TRUE!! In the software engineering world we call it Garbage In Garbage Out.

I would like to have a dime for every time that a TC wasnt locked and people were complaining about how VD wasnt accurate because look at this hump ... I am through explaining that to people :) They get blank stares from now on because they first need to learn how a car works. (Wow ... that was grumpy of me. I just traded my Evo for a GT500 and drove 400 miles ... Im tired and my gf is 2.5 hours away so Im having PMS.)
 


I think this would be accurate but you have to do probably 5 runs at least to get an average and then compare it to 5 more runs after modifications to see increase. When youre running E85 like me, thats like a whole tank so no bueno. Also matt, why not just go measure your tire diameter? Thats what I did and got around 25.5 or a little more.
 
Without a converter being locked you cant get an accurate number because its just flat drive train slippage ... same as a clutch slipping or wheels spinning. As long as you can recognize the parts of the graph that are showing this you can trim it out with the RPM trim bar at the bottom of the run.

This feature should have been used on the blue and green runs above due to the spikes in the tails of the graph.
 
You must log in or register to reply here.
Back
Top