Dyno Pull, or why I'm dumb as a bag of hammers

[jm1234]

Originally Posted by wolftrouble

I'm happy to be wrong; I know it doesn't just modify the sensitivity - you can have the throttle rock-steady in a certain position, hit 'power' and it lurches forward. Now I understand that may be (and it sounds like is? Please be true) that it's just because it shifts the throttle mapping.

If you want to check for yourself put the pedal to the floor and then push the Power button. In every throttle mapping (normal, sport and sport plus) that pedal position should be wide open throttle (WOT) and the Power button shouldn't change the throttle response there. Let the board know if it does.

[matt beard]

Originally Posted by PencilGeek

First of all, I'm 90% sure I know where you went, and I can tell you what went wrong. I've written about this on many occasions already. But hopefully a new crop of people will learn from these experiences.

First mistake: you let the dyno "guess" your gear ratio. In the absense of actual gear ratio data, the Dynapack can "guess" your gear ratio by holding the motor at a specified RPM and using the hub speed to calculate the gear ratio. This appears to be a 6MT, and 4th gear was used (not 5th). You would have hit the speed limiter in 5th gear. The dyno guessed 4.5:1, when the actual ratio is 4.586:1. This place only has one dyno operator, and he was lazy. If he wanted to, he could have looked up my dyno chart, and used the proper gear ratio. I dyno'd in 3rd, 4th, and 5th gear at this shop.

Second mistake: This particular shop does not have enough air flow for your M3 motor. The M3 motor reads air temperature and air flow and knows when it's sitting still. When there isn't enough air flow, the ECU will richen the fuel mixture, and you'll get a much lower horsepower output. The same thing happened to me at this shop. After I switched shops, also using a Dynapack, I picked up quite a bit of power -- all because of air flow.

Third mistake: You used 4th gear; you should have used 3rd gear. Using a 1:1 gear is a total myth. Every gear ratio (even 1:1) has a primary and secondary sprocket, an output shaft which drives a primary and secondary on the final drive. You're ALWAYS going through four gears sprockets before you measure the output -- even when the gear ratio is 1:1. Different gears have greater and lower mechanical efficiency -- and they go in an inverse ascending order. 6th gear is the least efficient, followed by 5th, 4th, etc. You should dyno in the gear that has the greatest mechanical efficiency -- and won't overload the torque rating of the dyno. On the M3 and Dynapack, that's 3rd gear. (6.084:1)

Using your power or "M" button will have no effect. I dynoed this exact test three separate times, and proved it has no effect. As you figured, turn off DSC. Also turn off all electronics (fans, lights, AC, etc.).

HP = (TQ * RPM) / 5252. It's not just typed in...it's a formula.

Lastly, call Aggressive Performance in San Jose; ask for Su. He'll do it right. He's dyno'd mine 6 or 8 times already. http://www.agpmotorsport.com/home.html

My choice for moderator.

[!Xoible]

dont get mad, get glad!

u learned something man. all good. next lesson?

[SenorFunkyPants]

Originally Posted by SenorFunkyPants

Anyone successfully run a DCT m3 on a Dyno?

Is that a "no" then?

[NGC]

I have had 4 sessions on a dynapack (hub based dyno) and there are way too many smarts in the car to just test out of the box as the car knows its not moving and will refuse to allow manual changing beyond 2nd gear to get to your desired gear.

This is what we did..
1. Turn of DSC fully at the button as well as change your Msetting in the idrive to DSC off just in case
2. Trick the car to trip ABS, we did this by pulling a few runs in D5/6/7 and it just went off (alternatively pull one of the ABS sensors of the wheel)
3. At this point we were able to pull 3 runs and then the car convinces itself that you have been using launch control and will retard the engine. We saw 70hp losses on subsequent runs.
4. To avoid above just run the car on the dyno for 5 miles and it resets itself.

Here was my last run following AA software and dinan throttle bodies, not as good as we expected but transpires BMW didnt calibrate the TB's correctly and we had AFR problems.

Final thing is that the DCT seems to have a torque controller which we are seeing as a roller coaster curve, rather than the flatter curves that the 6mt guys seems to be getting.

will be dynoing again in 2 weeks now that the software and TBs have been recalibrated,
Attachment 249667

[psuwcc112]

Originally Posted by wolftrouble

Thank you, thank you, thank you. Just talked to Su, heading over there this weekend. He says hi.

please let us know the result. I have very close set up as you.

[s78]

Originally Posted by PencilGeek

First of all, I'm 90% sure I know where you went, and I can tell you what went wrong. I've written about this on many occasions already. But hopefully a new crop of people will learn from these experiences.

First mistake: you let the dyno "guess" your gear ratio. In the absense of actual gear ratio data, the Dynapack can "guess" your gear ratio by holding the motor at a specified RPM and using the hub speed to calculate the gear ratio. This appears to be a 6MT, and 4th gear was used (not 5th). You would have hit the speed limiter in 5th gear. The dyno guessed 4.5:1, when the actual ratio is 4.586:1. This place only has one dyno operator, and he was lazy. If he wanted to, he could have looked up my dyno chart, and used the proper gear ratio. I dyno'd in 3rd, 4th, and 5th gear at this shop.

Second mistake: This particular shop does not have enough air flow for your M3 motor. The M3 motor reads air temperature and air flow and knows when it's sitting still. When there isn't enough air flow, the ECU will richen the fuel mixture, and you'll get a much lower horsepower output. The same thing happened to me at this shop. After I switched shops, also using a Dynapack, I picked up quite a bit of power -- all because of air flow.

Third mistake: You used 4th gear; you should have used 3rd gear. Using a 1:1 gear is a total myth. Every gear ratio (even 1:1) has a primary and secondary sprocket, an output shaft which drives a primary and secondary on the final drive. You're ALWAYS going through four gears sprockets before you measure the output -- even when the gear ratio is 1:1. Different gears have greater and lower mechanical efficiency -- and they go in an inverse ascending order. 6th gear is the least efficient, followed by 5th, 4th, etc. You should dyno in the gear that has the greatest mechanical efficiency -- and won't overload the torque rating of the dyno. On the M3 and Dynapack, that's 3rd gear. (6.084:1)

Using your power or "M" button will have no effect. I dynoed this exact test three separate times, and proved it has no effect. As you figured, turn off DSC. Also turn off all electronics (fans, lights, AC, etc.).

HP = (TQ * RPM) / 5252. It's not just typed in...it's a formula.

Lastly, call Aggressive Performance in San Jose; ask for Su. He'll do it right. He's dyno'd mine 6 or 8 times already. http://www.agpmotorsport.com/home.html

There's some good stuff in there pencilgeek, thanks.

However I have to wonder about a couple items:

Originally Posted by PencilGeek

Different gears have greater and lower mechanical efficiency -- and they go in an inverse ascending order.

Different gears necessarily have different gear ratios, not necessarily different efficiencies (unless you have data from BMW on efficiencies?). Third gear may be fine for dyno runs, but I question the logic for deciding its the best. Fourth gear might be fine also.

At the end of the day, if you want to compare runs, the important thing is you must stay on the same dyno and run in the same gear every run.

Originally Posted by PencilGeek

This particular shop does not have enough air flow for your M3 motor. The M3 motor reads air temperature and air flow and knows when it's sitting still.

Are you talking about airflow into the engine intake? Or airflow through the radiator? If there is not enough airflow through the radiator then the engine could run too hot, and then it is possible that fuel and / or ignition could be changed to cool and protect the engine. I'm curious how you came to your conclusions on this one, please expand.

Originally Posted by PencilGeek

Every gear ratio (even 1:1) has a primary and secondary sprocket

BTW sprockets are for chains (like on your bicycle) and I don't think the transmission is a box o' chains. It has gears.

[SenorFunkyPants]

Originally Posted by NGC

I have had 4 sessions on a dynapack (hub based dyno) and there are way too many smarts in the car to just test out of the box as the car knows its not moving and will refuse to allow manual changing beyond 2nd gear to get to your desired gear.

Thanks for the info, it looks a bit more complicated than I had hoped...I have the impression that my car is a bit more powerfull than the norm and wanted to test it to find out but maybe I'll just settle for living with the illusion!

[NGC]

It sounds complicated but its not....and getting a baseline is really good for future mod assessment. Also hearing your car redline in a closed environment is intoxicating

[s78]

Damn pencilgeek after I saw the sprocket thing I have to admit I assumed you were rather out to lunch and was therefore extremely skeptical of the rest of the post. Looking up your other posts I see you know your shit, which I appreciate... teaches me to jump to conclusions.

Originally Posted by PencilGeek

There's far too much history with me on this subject for me to start explaining it from scratch. It's all here on the forum for anybody who wants to look it up.

I'm looking up the gear ratio stuff...

Originally Posted by PencilGeek

I have mixed feelings about this for a couple of reasons. (Let's assume we're always using Dynapack dyno's.)
1) SAE 1349 correction factors should take care of atmospheric condition differences on the dyno from day to day, and hour to hour. That is...if you trust the correction formulas.
2) When the dyno is accurate withing 1/10th of a ft/lb of torque over it's entire hub speed, then it shouldn't matter if you're on Dynapack-A or Dynapack-B.

I agree, SAE corrections will have to come in to play even if you use the same dyno all the time since you probably will not be controlling the temp / press / humidity where you are testing. I assume from item number (2) that Dynapack must claim 1/10 ft-lb accuracy on all its dynos? That's really good. So if that's true, and all shops keep those things calibrated and don't have any special fudge factors then yeah, you would probably be good to go between dyno's. But that's a lot of faith in things you don't have control over... I guess I don't have that kind of faith!

Originally Posted by PencilGeek

Short answer -- the place this guy used only had one very small fan, and in my case, they didn't even turn it on.

Wow. I would not trust anything having anything to do with that shop if they didn't even have a fan during the runs. Pretty sad.

Originally Posted by PencilGeek

Out of curiosity, I just looked up one of the big industrial gear manufacturers, and noticed they refer to the individual pieces of cog/helical gears as "sprockets" too -- so maybe I was right on the money after all.

...hmmmm.... I've never heard of that... are they british, or of some other unworldly origin?

[NGC]

PG

I will document and film where appropriate the next Dyno tests with the DCT transmission.

Once the ABS is triggered, you can change up and down through the gears. The last run I did we tried 6th gear (given there are 7 with the DCT box) and no issues other than a starting torque number of 273ft/lbs at 2900 RPM.

Did you find your AFR numbers constant between 4k and 7k rpm

[SenorFunkyPants]

Originally Posted by NGC

PG

I will document and film where appropriate the next Dyno tests with the DCT transmission.

Thanks you guys for an interesting read...I have subscribed to this thread so if you do film and document a DCT M3 dyno run could you note it here please...Thanks.
I was at a dyno session the other day where they ran an M5 with a modified exhaust and the sound was spine chilling.

[s78]

More gear efficiency stuff for PG or whoever is interested:

I looked around a bit for gear efficiency stuff and really only found one place that has a calculator for spur or helical gears, here:
http://www.roymech.co.uk/Useful_Tabl...fficiency.html

I used the spur gear equations in the "spur gears / helical gears" section. I'm sure helical would be different, but I'm really just looking for trends more than absolute numbers, so I think this is ok.

PG posted some educated guesses of gear tooth counts for the 6MT, that's what I used for the calcs. (If anyone tears down their gearbox please count some teeth) I estimated the friction coefficient to be 0.05 for all gears. It's just a guess, but again we are just looking for trends so it doesn't matter too much. Below see the "Power loss" for each gear as calculated. The lower the power loss, the higher the efficiency.

Gear/Power loss %
1 /0.92
2 /0.52
3 /0.42
4 /0.49
5 /0.64
6 /0.62

I would love to post more info but I couldn't figure out how to get an image of my Excel worksheet in this post so I stripped it down to just the Gear and "Power loss" columns.

The data does not show a trend of efficiency vs. gear ratio. The data shows third gear is most efficient, and 1st gear is least efficient.

*Insert big caveat* Please note this is frictional power loss in the teeth only. "Windage", or the losses of churning up all the gear oil, are not included. There's a whole bunch of other stuff that goes into gear efficiency like loading, gear speed, etc. which makes me think the only way to ever know the efficiency is through testing. Here's an interesting NASA paper http://gltrs.grc.nasa.gov/reports/20...003-212222.pdf

[wolftrouble]

I promised I'd update the thread when I got the new dyno pulls. Thanks to PencilGeek's excellent advice I went to Aggressive Performance in SJ and Su took care of it professionally and quickly. Really nice guy and really great work.

Dyno chart's below, with what I'd consider to be much more accurate numbers. Also, for funsies I brought along a digitial recorder and some hi-end condenser mics and got a great audio recording of the last pull; I'll probably post it somewhere if anyone wants to hear it. Not the kind of sound you usually get to hear with it standing still.

Thanks again everyone, especially Pencilgeek for all the help. Su was kind enough to give me a copy of the Dynapack reader file so if anyone's curious about any of the stats besides hp/torque let me know and I can forward it along.

[DevJ]

post the clips!!

[dcstep]

Originally Posted by wolftrouble

I promised I'd update the thread when I got the new dyno pulls. Thanks to PencilGeek's excellent advice I went to Aggressive Performance in SJ and Su took care of it professionally and quickly. Really nice guy and really great work.

Really stout numbers for stock. Have you got any mods? Thanks for the followup.

Can't wait to hear your sound clips. I may take a condensor mic and DSD recorder to my next dyno session.

Dave

[wolftrouble]

Not stock. That's with the AA underdrive pulley, Remus Sport exhaust, and RPI scoops (which we'll just pretend are red decorations, for the purposes of a dyno pull. )

Assuming 15% drivetrain loss that gives it ~424bhp at the crank, yes? Listen to me pretending I know what I'm talking about.

[psuwcc112]

Originally Posted by wolftrouble

Not stock. That's with the AA underdrive pulley, Remus Sport exhaust, and RPI scoops (which we'll just pretend are red decorations, for the purposes of a dyno pull. )

Assuming 15% drivetrain loss that gives it ~424bhp at the crank, yes? Listen to me pretending I know what I'm talking about.

No filter mod?

[wolftrouble]

Oops, sorry, yes - that also includes an MS air filter. I forgot, the mod took me like 10m to install.

My audio recording, FWIW, was done with an Olympus LS-10 and a pair of Core Sound Binaural mics. Kind of screwed up the positioning a little, I put them about 15ft behind the rear of the car, offset to the side about 3 feet but not enough to avoid the wind from the exhaust making a slight whooshing sound (sounds a little like tape hiss) on the recording.

[wolftrouble]

Audio is here: http://web.me.com/wolftrouble/M3/M3.html

[dcstep]

Originally Posted by wolftrouble

Audio is here: http://web.me.com/wolftrouble/M3/M3.html

Nice! Sounds more like an American V8 than I would have expected. (That's not a diss. I like the sound).

It'll be interesting to contrast my Active Autowerke Signature. At WOT it also sounds very American-8ish from inside, so they may be similar. Under light throttle/high revs it sounds way different inside. Cruising up and down parking structures at 1500 to 3000 rpm in second it sounds like a huge, high reving, exotic motorcycle.

I just need a couple of hundred more miles on my 4.10 FD before I go back to the dyno.

Dave