E92 M3 6MT vs. M DCT - Simulation

[T Bone]

With the M DCT press release, I plugged in data into the CarTest 2000 software. I just used the data in the press release.

For the 6MT, I used a 0.35 second shifttime. For the M-DCT, I used a 0.05 second shifttime. Slightly more weight for the M-DCT. Else everything else is the same.

One thing that I noted when I entered the data.....the M-DCT has a very tall rear-end. In fact, it looks like the 7th gear is used exclusively used as a fuel economy gear. It is never even used in obtaining the acceleration to top speed.

Also, I wonder if we will be able to see any of the numbers that one of the rags put up. This software has proved to be pretty accurate.

Please take a look and discuss.

[masmole]

nice work!

[gasman]

I hope real life numbers look better than that!

[lucid]

Swamp ran the same set of simulations, and posted here:

http://www.m3post.com/forums/showthread.php?t=107582

Did you see those? He's got faster outcomes. I don't know how you guys are setting up the simulation and what's different.

One thing Swamp caught earlier with regards to DCT based on a tangential comment of mine was the time step. What is your time step, and is it significantly lower than the DCT shift time?

[T Bone]

Originally Posted by lucid

Swamp ran the same set of simulations, and posted here:

http://www.m3post.com/forums/showthread.php?t=107582

Did you see those? He's got faster outcomes. I don't know how you guys are setting up the simulation and what's different.

One thing Swamp caught earlier with regards to DCT based on a tangential comment of mine was the time step. What is your time step, and is it significantly lower than the DCT shift time?

I didn't see that but here are the assumptions and the data chart from the sim that I ran.

I used a DCT shift time of 0.05 seconds and a drag coefficient of 0.29 (this was just a guess since BMW only gives Cd X A number)

Hey Swamp - what other variables did you change?

[T Bone]

Here are the speed and distance charts. I changed the default parameters to use higher speeds and also the standing 1/4 mile, standing kilometer and standing mile.

I am really puzzled by BMW's choice in gearing with the M-DCT. The sim program doesn't even use 7th gear to acceleration.....presumably the wheels get more torque at all speeds in 6th gear.

Did BMW waste the 7th gear for fuel economy?? What a waste!!

Looking at the data, the 6MT closes the intial gap after the quarter mile.....is the 6MT better geared than the M DCT 7 speed for higher speed runs??

Confused.....

[PhoenixM3]

Some recent magazine tests indicating (for example) the 0-100MPH times. I distinctly recall reading 9.4 seconds in one of the monthly rags.... I currently own an '02 M3 and I'm debating on a new M3.... Again, well done. I'm hoping that you'll be able to refine this even more as time passes.

Ciao,

Jeff

[InJapan]

I was wondering if this data was based with / without full launch control. The 4000+ rpm dump vs 1800 rpm that we have now with the SMGII.

My only concern with the DCT is that BMW will disable the full launch control on the US M3's just like they did with the E46 M3 SMGII.

[MrHarris]

car and driver for 4.4 to 60 and they mentioned they had traction issues and weighed the car in at 3571lbs (showed the car on a scale in one of the pictures), also r&t got 4.1 to 60 and 9.8 i believe to 100. pretty shocking numbers in my opinion.

[T Bone]

Originally Posted by MrHarris

car and driver for 4.4 to 60 and they mentioned they had traction issues and weighed the car in at 3571lbs (showed the car on a scale in one of the pictures), also r&t got 4.1 to 60 and 9.8 i believe to 100. pretty shocking numbers in my opinion.

Originally Posted by InJapan

I was wondering if this data was based with / without full launch control. The 4000+ rpm dump vs 1800 rpm that we have now with the SMGII.

My only concern with the DCT is that BMW will disable the full launch control on the US M3's just like they did with the E46 M3 SMGII.

Originally Posted by PhoenixM3

Some recent magazine tests indicating (for example) the 0-100MPH times. I distinctly recall reading 9.4 seconds in one of the monthly rags.... I currently own an '02 M3 and I'm debating on a new M3.... Again, well done. I'm hoping that you'll be able to refine this even more as time passes.

Ciao,

Jeff

I realize these numbers are slower than some rags but in my experience this simulator is pretty good. The question is can someone replicate the numbers in the rags with their own car.

The software also has a mode that test different rpm clutch dumps, which is similar to launch control. The software uses different rpm clutch dumps and then plugs in the rpm that gives the best times.

[JEllis]

I feel the greatest advantage of the DCT over the 6MT for us everyday drivers is that it gives you the ability to hit say 4.5 every single time. Thats great for those of us willing to challenge our buddies from a stop light...

That being said, something bothers me about your simulation...thr 60ft times

Optomized they should be 2sec or less....no question

Anyone that has been to a drag strip knows that most cars should hit the 60ft mark in 2sec or less. Its the sign a of a good launch. I have gotten under 2 sec in my E36 M3. I have not doubts this car is capable of the same.

Jason

[sayemthree]

car mags have already got 4.4 to 4.1 seconds in 0-60 mph. with the 6MT.

[swamp2]

Yup, repost T-bone! I just have to drop the:


I don't know if we will be getting BER but I am using 0% accessories loss which should simulate BER fairly well. I also used BMWs quoted shift times roughly, using 1ms (software does not take 0ms) for the shift time and 5ms for the clutch engage time. I am also using 420 hp instead of 414. The first and last points are equivalent to simply a little bit better total drivetrain loss than defaults. You absolutely need to use a time step, called "time increment" in CarTest of 20ms or so, I use 10ms. The default is 50ms and it simply can not capture the ms time scales of DCT workings. I still can not get the M-DCT to break a 2 second 60 ft time so maybe the launch/wheelspin algorithms are not perfect. All of the other times I have predicted for the 6MT fall withing the observed bounds on our performance "database" thread.

[T Bone]

Originally Posted by swamp2

Yup, repost T-bone! I just have to drop the:


I don't know if we will be getting BER but I am using 0% accessories loss which should simulate BER fairly well. I also used BMWs quoted shift times roughly, using 1ms (software does not take 0ms) for the shift time and 5ms for the clutch engage time. I am also using 420 hp instead of 414. The first and last points are equivalent to simply a little bit better total drivetrain loss than defaults. You absolutely need to use a time step, called "time increment" in CarTest of 20ms or so, I use 10ms. The default is 50ms and it simply can not capture the ms time scales of DCT workings. I still can not get the M-DCT to break a 2 second 60 ft time so maybe the launch/wheelspin algorithms are not perfect. All of the other times I have predicted for the 6MT fall withing the observed bounds on our performance "database" thread.

Didn't mean to generate data and repost

I will try these settings.... BTW, I always enter KW instead of HP / PS for power.....the Metric system is the only way not to get confused.

[lucid]

Originally Posted by T Bone

I always enter KW instead of HP / PS for power.....the Metric system is the only way not to get confused.

So, true, but there is something romantic about horses pulling on stuff...

[sward]

0-100km/h M3 Coupe manual 4.8s
0-100km/h M3 Coupe M-DCT 4.6s

http://shrani.si/f/r/10E/3xFS3lZm/mdctgbfv.pdf