Veloster Turbo Forum banner

Virtual Dyno and Data-logging

68K views 435 replies 28 participants last post by  swordfish 
#1 · (Edited)
Virtual Dyno and Data-logging

Want to see how much power your car is making, but can't afford or find the time for a dyno? Need to log AFR, Boost, Timing, and more? This is the solution. Proven time and time again to be within 3-5% accurate when compared to Dynojet dynamometers, Virtual Dyno can be a great tool for comparing with others or revealing power issues.

All of the tools needed are completely free, except for the OBD2 adapter. You can find those on ebay/amazon for around $10-25, which is peanuts compared to a single dyno test!

This should be a fully comprehensive guide to setting up, taking logs, and graphing them to be saved for future use.

First thing you need is a smartphone of the Android or iPhone variety. Sorry, windows phone does not have an app for logging.
Download an app called Torque for Android. If you have an iPhone and know how to take logs please post it so that I can incorporate it into the guide.

Once you have the Torque app installed, there is two ways to set it up. One for Lite and one for Pro.

Setting up Torque Lite
Torque Lite will give you the minimum parameters you need to log in order to make a Virtual Dynograph. You are not able to log Boost or AFR in Torque Lite. It has an option for Boost, but the values it gives are incorrect. Also, one of the limitations of using Torque Lite is that it does not save your PIDs for logging. So each time you start the Torque Lite app you have to re-add all the PIDs again.

Text Font Games Screenshot
Text Font Line Screenshot
Text Font Line Screenshot Document
Text Font Screenshot Technology Electronic device
Text Font Line Screenshot Number
Text Font Technology Screenshot Electronic device

Black White Darkness Text Red
Black Text Font Screenshot Sky
Text Font Line Technology Screenshot
Text Font Line Photo caption Screenshot
 
See less See more
10
#2 · (Edited)
Now you can actually start logging.

Technology Electronic device Screenshot Digital clock Electronic instrument
Technology Electronic device Games Electronic instrument Gadget


When you're done, press the same button again to stop the log.

Technology Games Electronic device Electronic instrument Screenshot


Torque Lite does not separate the log files between each new log, so you'll have to either manually edit the spreadsheet yourself or clear logs after each log.

You can transfer the logs to Google Drive or your own e-mail. Always choose CSV, it's the only format supported by Virtual Dyno.

Text Screenshot Games Adventure game Font


Setting up Torque Pro
Paying the $4.99 asking price gets you a lot of extra goodies and saves you some hassle with a better dash, extra settings/PIDs, USB cable support, and more! Much of the setup is the same, however, there are slight differences that need explaining.

Font Speedometer Technology Screenshot Electronic device
Text Font Screenshot Technology Electronic device
Text Font Screenshot Technology Electronic device
Text Font Screenshot Technology Electronic device
Text Font Screenshot Sky Technology
Text Font Screenshot Technology Electronic device
 
#3 · (Edited)
Text Font Screenshot Technology
Text Font Screenshot Technology Electronic device
Text Font Screenshot Technology Games


Now when you want to start the log, just press the cog wheel and 'Start Logging'.
Text Font Screenshot Technology Adventure game
Text Font Screenshot Technology Games


You can transfer the logs using the 'Email Logs' button, make sure to select CSV.

Setting Up Virtual Dyno
Download virtual dyno from www.virtualdyno.net and install it. You must have Windows XP or above. Sorry Mac users.

First you have to make a vehicle profile so that Virtual Dyno can properly calculate torque and power.
Text Blue Font Screenshot Technology


If you have a manual, make sure to use these specifications. If you happened to weigh your car at a truck scale or using corner scales, then use that.
Text Font Line Screenshot Number


Auto's, you can use these.
Text Font Line Screenshot Number


Make sure you click 'Save as Custom' when you finish. Don't change the directory it defaults to, but name it whatever you want.

Now you need to change some of the preferences, make sure yours look like these images.
Text Font Line Technology Screenshot
 
  • Like
Reactions: gsrdude
#4 · (Edited)
You can set the Correction Factor to whatever you want, just make sure that you always use it when comparing to other runs. This is the only way you can compare mods. I recommend Virtual Dyno, it seems to be the closest to reality. I'm sure once more people make graphs we will find out what the perfect Correction Factor will be.
Text Font Line Number

*I'm now using a custom Correction Factor of CF1.01.

Text Font Line Number Screenshot


You no longer have to mess with the CSV files! However, if you have multiple runs in the same CSV file it will load all of them up at the same time. I prefer to keep my runs separated, but that's just me.
 
  • Like
Reactions: ECKO
#5 · (Edited)
Opening & Graphing Logs
After opening a log there are a few ways you can tweak the correction factor, smoothing, and more.

Text Font Line Screenshot Number


Text Blue Font Line Parallel


It's important that you double check the selected Gear, if the graph looks wrong this is a good place to check.
Text Font Line Technology Parallel


Add the total weight of yourself and anything else inside of the car. You can include fuel if you want. Just make sure you let others know when posting the graphs.
Text Font Line Technology Number


SAE correction allows you to correct for different temperature and barometric pressure. You can find this on wunderground.com.
Text Font Line Number Parallel



It's very important that you take as many runs as you can and compare them. If you get a lot of torque then you are likely on a road that is on a decline. Keep trying different roads until you find the one that allows you to take repeatable results.
Text Line Plot Parallel Design


You should load up multiple runs at the same time so you can compare them and figure out which is the most accurate.
Text Font Line Technology Parallel


Since all three runs have similar curves after 4000 RPM, we can assume that the road itself is the problem and not the correction factors. Another way to know if the road is the problem is if adding more smoothing results in a difference greater than 15WHP/TQ. That is a sign of bumps in the road or a road that isn't flat.
Text Line Plot Parallel Pattern


That's it! It will take some trial and error, but eventually we will get to a point where we can find out if a mod adds power and where.
 
#6 · (Edited)
Ah-HA! I figured out why the torque is so high in the beginning!

This program is looking for Cobb Accessport time stamps, which counts time cumulatively. Torque counts time as in the current time on the device. So all I did was convert all the time stamps with the equivalent times the program expects and now it is reading much better!

Comparison to show you
Text Line Plot Diagram Slope



Convert This
Text Font Line Pattern Number


To This
Text Font Line Number


I think what happens is that the program calculates torque based on the time it takes to reach an RPM. So it needs to see time in this way, otherwise it isn't able to properly do the math.
 
#426 ·
Ah-HA! I figured out why the torque is so high in the beginning!

This program is looking for Cobb Accessport time stamps, which counts time cumulatively. Torque counts time as in the current time on the device. So all I did was convert all the time stamps with the equivalent times the program expects and now it is reading much better!

Comparison to show you
View attachment 33478


Convert This
View attachment 33479

To This
View attachment 33480

I think what happens is that the program calculates torque based on the time it takes to reach an RPM. So it needs to see time in this way, otherwise it isn't able to properly do the math.
I don't understand the math you're using to convert the time stamps here. Can anyone help?

Sent from my SM-G920V using Tapatalk
 
#8 ·
Yeah as we discover more about this program we can tweak it. The high torque was the biggest issue, but now it seems like it is pretty close to an actual dyno.
 
#10 · (Edited)
We use VD in EVO land as well. Dead accurate once set up properly. I have been using it for quite a few years now

As you said you'll want to find the flattest road possible and make multiple runs to overlay

Another great post. Nice to see good info getting to the community here instead of the usual "This Oil Is The Best Thread"
 
#11 ·
Snail, snail, snail.. bro, exceeeeelllllleeeeennntttttt write up! Pretty sure this could be titled as Virtual Dynoing for Dummies! HAHA! I'm def going to have to try this sometime tonight or this weekend as I'm very curious to see the actual curve difference with installing the Tork pipe this weekend too! I will be trying to do logs with stock intake on and with my DIY intake too. Just gotta get the time in my schedule, not to mention we're currently in rain right now and are expecting torrential downpours tomorrow. Great job and thanks for doing it for us!
 
  • Like
Reactions: SnailPower
#13 ·
Can't wait to see your results with the Tork pipe because I just installed mine last weekend and I can't wait to go home and try this as well! Thanks Snail!
 
#14 ·
I'm just curious if there is actual any loss like claimed on the low end and what gains are to be had on the high end. Fortunately for me, I have a mile long access ramp that runs right next to a road where there's about 3 miles of burn out tracks where it's flat surface of nothing but runs happening lol! This was years ago, but now there's a trailer park there and no one ever goes down it. So, my results should be pretty consistent! We shall see!

Snail, is there the option to factor in other logging measures, such as IAT's into the graphing so that you may see how the car reacts under different intake temps?
 
#15 ·
Yea, use SAE if you want to see how temperature would effect the car. However, because our car is turbocharged, you can probably add a few extra HP/tq since we rely on cold temps more than naturally aspirated cars.

I feel like uncorrected may be simpler though. When it asks for barometric pressure, it wants the pressure that the ECU sees. If you log the Barometric Pressure PID, then you will know what it is.
 
  • Like
Reactions: Gerid
#16 ·
Ahhh, ok. Well, i'm going to have lots of questions this weekend.. so bare with me!
 
#17 ·
Most of the work is just setting up the apps. After that its fairly straightforward. We need someone who recently dyno'd to take some logs that way we will know what we need to use to make it more accurate.
 
#19 ·
Wonder if Nox or Zex have the app, being that there the most 2 recent tunes that I know of with dynos...
 
#21 · (Edited)
So I did a test and it shows some interesting results.

Stock Air Box connected to TS piping vs Blox filter (velocity stack) connected to TS piping. It's like the dip in power moved to another RPM band with the Blox Filter... maybe I'll do another test with the 10" Spectre. Also look at the Boost and you can see that the Blox filter holds more boost longer.
Text Line Plot Design Parallel


Well it proves how restrictive the airbox is. It also shows that my torque dip is somehow related to the intake...
 
#22 ·
So I did a test and it shows some interesting results.

Stock Air Box connected to TS piping vs Blox filter (velocity stack) connected to TS piping. It's like the dip in power moved to another RPM band the Blox Filter... maybe I'll do another test with the 10" Spectre. Also look at the Boost and you can see that the Blox filter holds more boost longer.
View attachment 33525
The velocity stack works eh?
 
#28 ·
This test was with a modified airbox, removed snorkel, removed magnet thing, smoothed all internal ridges, covered hot side port, etc.

Its just such a small filter and the inlet is only 2.5". Very restrictive.
 
  • Like
Reactions: Gerid
#30 ·
It didn't gain, it caused the turbo to spool up sooner due to less air. Look at the boost in the graph, the stock airbox made 2-3psi more boost below 3500 RPM. But it lost 2psi above 5500 RPM.

Look at how lean the AFR is with the stock airbox until 3500 RPM, lot more air.

The torque curve shifted left, that's generally what happens when you restrict the flow of air in or out.
 
  • Like
Reactions: Gerid
#32 ·
That's why I'm looking forward to the downpipe install and then the Big Gulp. I just don't understand why this car is so picky about restrictions. It shouldn't be causing a massive dip, just less power.
 
#34 ·
I get that it would move some due to less restriction, I was just having trouble grasping the extra trq.

Going to be playing with this tonight hopefully, I've got the virtual dyno downloaded and torque app setup -- now just for the runs. It's been raining and temps are wayyyyy down (may even hit an all time record low) and I bet we could pull some nice numbers over the high 90's that's expected this weekend. I'm assuming I can shift out of 3rd or do I need to stop the log right at redline?
 
#35 ·
It took me a few tries before I found a road that gave me consistent numbers. Just saying, take 2 runs on each stretch of road and whichever one was the most consistent always use that road.

You can use 3rd or 4th gear, just make sure that you put the pedal to the floor right before the car spools up (2000-2500 RPM).

Also, you don't have to edit the CSV files. The program is now smart enough to read Torque data. So that's even easier.
 
  • Like
Reactions: Gerid
#36 ·
Just bought an Elm 327 bluetooth adapter and torque pro.full ver to do this. Can't wait to see the log of my VT.
Thanks OP :)
 
#37 ·
Ok so I'm guessing that since my VT is an auto I should put it into the "manual" mode and once I get to the 3/4th gear at around 2 to 2.5k rpm I should go WOT? And do I just go to the redline and then change gears? Sorry about the noob questions. First time ever with a turbo car haha
 
#38 ·
Start the log as low as you can. If you start too close to where the turbo spools it will just show 300+ torque which is wrong. I'm still testing to see what RPM works best. But for now, I'd recommend 1500 RPM or so in 3rd.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top