"Virtual Engine Dyno" Thread - from Stock to Bolt-ons

[mixja]

This is a new mechanism I thought of for measuring engine power. The BMW ECU has a parameter called "Current Torque", which outputs the engine's current torque in Newton Meters. Using a tool like the Bavarian Technic Diagnostic Tool (http://www.bavariantechnic.com/) you can log this value in conjunction with engine speed from the ECU to give you a reasonably cheap and accurate crank horsepower calculator.

Exactly how the Current Torque value is calculated I'm not sure - my guess is that it is calculated based upon a number of other sensor readings in the engine. I'm very confident however this figure will be very accurate, given it is a parameter logged by the BMW ECU.

The main point of this thread is to establish a baseline of my car in stock trim, and compare the effect of a number of bolt-on modifications which will include in likely order of testing:

- Envy Charge Pipe
- AFE/BMC Air Filter
- AFE Air Scoops
- AA Crank Pulley
- Powerchip Stage III Software (claimed +40WHP)
- Turner Primary Cat Delete

All results will be normalized using SAE J1349 to provide comparative readings as each power measurement is taken. Three runs are taken each time and the best run posted - note that the results from run to run are very similar so far...

Note I've posted an historical dyno plot for results captured a few months back when my car was not stock - I didn't have any measurements for SAE correction so I've presented the results uncorrected...

Historical Dyno: Powerchip Software, AA Crank Pulley, AFE Filter and AFE Scoops



Dyno #1: Stock Results


Interesting results - looks pretty much as advertised until the engine hits 7000RPM - past here a fairly uniform decline in torque causes power to peak well short of the claimed 414HP @ 8300RPM figures. My car had the very latest software applied one week ago.

[mixja]

Placeholder

[Singletrack]

I used a similar value, and VAGCOM, on my old Audi 2.7T to generate a "virtual" dyno and it was pretty accurate as well. Good idea.

[Singletrack]

Originally Posted by mixja

Interesting results - looks pretty much as advertised until the engine hits 7000RPM - past here a fairly uniform decline in torque causes power to peak well short of the claimed 414HP @ 8300RPM figures. My car had the very latest software applied one week ago.

What gear were you doing the virtual dyno in? I would assume that could affect the results. I'm sure it will be pretty hard to dyno in 4th gear on the road ; )

[mixja]

Originally Posted by PencilGeek

Brilliant idea. I've looked in the list of BT "actual values" and completely missed the torque value. Next time I get my hands on the stroker, I'll run through some gears and log the results. Better yet, next time I get the stroker I'll be on the dyno and will log the values and compare against the dyno results. It's possible that in my case, the two won't correlate well because the ECU has been completely retuned -- and I'll bet those torque values are dependant on programming in the ECU. But we will see.

I like this idea so much, I'll add a new Dyno Type to the DynoDB.

Also curious how you did the SAE correction. I've spent a LOT of time with those formulas, bought the spec from SAE, and cross-checked my formulae against known industry standards. When I ran your numbers through my formulas, I get the following: =SAECF(65, 29.1, 55) = 1.053574619.

If you go to any of my dyno threads, you can download the spreadsheet that has all of the SAE 1349 formulas in visual basic script.

Thanks!

On the SAE Correction - I just used an online calculator - http://www.csgnetwork.com/relhumhpcalc.html - of course I have no idea how accurate it really is...

[mixja]

Originally Posted by PencilGeek

Right, we need to establish a standard for this. It should be 3rd gear for the 6MT and 4th gear for the DCT. 2500-8400 RPMs. Traction control off. Flat surface (or as flat as it can be).

Yeah I agree, although a 4th gear run on the street even in DCT is a little risky. I personally am going to have to stick to 3rd gear runs, but I'll chuck in a few comparison 4th gear runs to see if there are any major differences.

Note I didn't have TC off for the Dyno #1 run above and the surface wasn't exactly flat (kind of dipped a bit then climbed a bit)...

For DCT it should probably be in S6 as well (Dyno #1 above was in S5, M Dynamic Mode)..

[swamp2]

Wow, very cool. This might be in the car in part for working with DCT. Looking forward to seeing some logged data. Curious as to how BMW does this. Could be some strain based sensor or could be extracted from some combustion/cylinder monitoring.

[Jonmartin]

Its kinda weird how it doesn't have a normal graph shape it just gets flat up top dunno if I like this idea to much. Good luck anyway.

[michaeldorian]

Originally Posted by PencilGeek

Right, we need to establish a standard for this. It should be 3rd gear for the 6MT and 4th gear for the DCT. 2500-8400 RPMs. Traction control off. Flat surface (or as flat as it can be).

Exactly. If we can agree on a uniform standard on the forum, we have a great way to do some independent testing. We should also standardize on a correction formula to use. Regardless of if the formula is 100 percent accurate, if everyone is using the same formula, we can all map the relative gains of each mod correctly.

I think this idea is genius. Definitely going to pick up the tool for a xmas present to myself.

[michaeldorian]

Originally Posted by PencilGeek

I will provide a spreadsheet that includes the SAE correction factor formulas. The intention is for people to cut/paste their results from the BT tool into the spreadsheet and post them to the forum, and submit them to the DynoDB. The users will need to collect temperature, barometric pressure, and relative humidity in addition to the results from the BT tool.

This is great. Thanks! Now time to start hunting down a long flat road somewhere.

[mixja]

Originally Posted by PencilGeek

Getting accurate weather data for your dyno runs can be difficult if you rely on the national weather service. Here's a quick and easy way to get weather data for the past few days. Use this method when submitting your Virtual Dyno results.

Collecting weather data immediately before or after your virtual dyno run:

• Go to www.weather.com
• Enter your zip code.
• On the main screen, obtain the temperature, pressure, and humidity.

Collecting weather data within seven days of your virtual dyno run:

• Go to www.weather.com
• Enter your zip code.
• Click the "Yesterday" tab and wait for the screen to refresh.
• In the pulldown tab (with "Yesterday" in the window), select the date of your virtual dyno run and wait for the screen to refresh.
• Click "hourly reports" and wait for the screen to refresh.
• Obtain the temperature, pressure, and humidity of the time of day of your virtual dyno run.

The BMW ECU also logs ambient temperature and ambient pressure, so the only real weather measurement you need is humidity. A low cost, fast response hygrometer would be most ideal for obtaining relative humidity...

[mixja]

One issue I have at the moment is the BT scan tool is not displaying the correct values for many variables - for example torque is multiplied by 10 (easily fixed), but the pressure/temperature readings are not so easily corrected (not sure as yet how to convert them correctly, I have support calls in with BT about this)...

[Jaypod]

This sounds pretty interesting - I dont have a BT Tool, but I have a PLX Kiwi Wifi - will have to see if I can collect the same data somehow.