|
|
08-23-2008, 08:49 PM | #23 | |
Major General
374
Rep 8,033
Posts |
Quote:
The more significant issue is which part of what has been measured during the shift and displayed in the first graph is the signal, and which part is the noise? Those two spikes can't both be entirely signal. What is the frequency of the signal vs the noise during that interval? Anyway, that'll hopefully clear up once the sensor is mounted properly.
__________________
|
|
Appreciate
0
|
08-23-2008, 09:54 PM | #24 | |
Lieutenant General
609
Rep 10,407
Posts |
@Art and lucid: There is much interest in a greater resoultion of the accleration profile during the shift and hence you need many sample points across the event. 300 Hz is marginal given Nyquist to capture a great deal of the actual dynamics during the shift. I think longer term, if shifts get much quicker than 30 ms 500 Hz would be a good sampling to stick with. I am not very interested in the frequency content myself, enlighten me.
Quote:
D modes are: -Mathematically more jerk (surge) -Psychologically smooth and very little surge -Very short in duration S modes (S4->S6, i.e. those that have surge) are: -Mathematically less jerk (surge) -Psychologically rough with great surge -Significantly longer in duration It is slightly contradictory, S4->S6 FEEL faster but aren't. The exact same phenomena happens with the shape of cars torque curve. Lumpy/peaky can feel fast but not be as fast. |
|
Appreciate
1
DrFerry6724.00 |
08-24-2008, 12:21 AM | #25 | |
Colonel
108
Rep 2,279
Posts |
Quote:
In any event, great job getting this up and running, swamp. Your initial efforts have already yielded some very interesting data and, as usual with these types of things, raised almost as many questions as they've answered. I'm curious to see the next stage -- hopefully, Ken or someone can fab up a mount that is successful in dialing out much of the noise. Great job, swamp. |
|
Appreciate
0
|
08-24-2008, 12:32 AM | #26 |
Major General
1072
Rep 8,008
Posts
Drives: i4M50
Join Date: Oct 2007
Location: No where fast
|
swamp,
I totally understand what you are saying and for once I agree with you, the length of the shift isn't important if the acceleration doesn't decrease during it, as is the case with the surge that is felt (S4~6). But I know from a purely personal interest I would love/need to know what the improvement was because of what I was told all those months ago, that the benefit doesn't out weight the harshness of the surge. P.S. Please swamp I am not wanting to bring up old arguments as we have come to an agreement on the surge and I see it's position after actual sampling but I too know why they (Audi) felt it wasn't needed. |
Appreciate
0
|
08-24-2008, 02:17 AM | #27 | |
Lieutenant General
609
Rep 10,407
Posts |
Quote:
By the way saying the benefit, which can be quantified, "does not" outweigh the the harshness seems like a very difficult thing the either quantify or make objective. It would very much depend on the person. |
|
Appreciate
0
|
08-24-2008, 02:37 AM | #28 | |
Brigadier General
233
Rep 3,302
Posts |
Quote:
__________________
"...it's not about the money and not about the brand of the car, it's about handling,performance and passion......And that, no other car has all together like an M3........when you talk about the most complete car the M is invincible." --Tony Kanaan. |
|
Appreciate
0
|
08-24-2008, 02:45 AM | #29 | |
Lieutenant General
609
Rep 10,407
Posts |
Quote:
Short shifts are better than long shifts in general. But this point of view comes from the fact that short DC shifts are better than long MT shift mostly because their is significant time with no power and hence with deceleration when shifting with a MT. DC systems change this game quite a bit. When shifts can maintain or even increase acceleration during the shift (using engine and flywheel momentum) then the length of the shift becomes almost irrelevant. I have not yet tested S5 nor S6 but because S4 shows no deceleration and actually shows acceleration (above and beyond the steady state in gear acceleration) during the shift, it is hence faster in overall acceleration than D3. S5 and S6 offer even more feeling of surge and hence they are almost for sure going to be even better accelerating than S4 or any D modes. Hope this helps. |
|
Appreciate
0
|
08-24-2008, 03:06 AM | #30 | ||
Major General
1072
Rep 8,008
Posts
Drives: i4M50
Join Date: Oct 2007
Location: No where fast
|
Quote:
Quote:
Yes each person will come up with their own opinions on this and what I might feel as little or no real benefit someone else might disagree based on what they are looking for. The very fact the even D5 doesn't offer surge leads me to believe the benefit is very small indeed, but again until some raw data is available this can only be an opinion. P.S. I am glad with are finally fight of the same side and what difference we have seem to small enough not to cause a 20 page back and forth debate on what is only fine detail. P.S.S. It's still engineered. j/k. |
||
Appreciate
0
|
08-24-2008, 03:18 AM | #31 |
Major General
1072
Rep 8,008
Posts
Drives: i4M50
Join Date: Oct 2007
Location: No where fast
|
On a different note,
Everyone, including yourself swamp thought that the lower modes D1~2 and S1~3 slurred the shifts to not only smooth out the change but disperse the energy stored in the flywheel and the higher modes that shift was far quicker and the surge were delivering it in a great big lump, the higher the mode the greater the amount. But after this test the opposite seems to be true. Everyone including myself is head scratching on this one. |
Appreciate
0
|
08-24-2008, 08:53 AM | #32 |
Banned
11
Rep 471
Posts |
I think the faster D times surprised Swamp too...why?
he chose the slower sampling rate 300hz for them, as opposed to the 500hz for the S mode, thinking they would be 'slower'... this is good stuff... I wonder if the ECU bus could be tapped to 'steal' the onboard acceleromters signals? what format is the raw data in? I'ld be interested in getting the file to play with... re: sampling rate let's assume we want to capture at least 5 ms info/events...~200 hz Nyquist would require >400hz...so 500 should suffice... |
Appreciate
0
|
08-24-2008, 11:40 AM | #33 | |
Major General
374
Rep 8,033
Posts |
Quote:
This again brings up the issue of what we think the "event" we would like to measure is exactly, and if the broader "shift event" can be broken down to shorter sub-events that should be characterized. For instance, the first graph suggests some deceleration which probably corresponds to the the disengagement of the first clutch, and then acceleration, which probably corresponds to the engagement of the second clutch. If that is the case, clutch engegament/disenagement constitute events. Then we can think about the frequency of that "event" and use the Nyquist–Shannon sampling theorem to come up with a required sampling rate. Another example is defining clutch slippage during engagement as an event although that is over the top and not relevant to the question being posed with this test. Does that occur as linearly as the first graph suggests? I'd be interested to know, but even with a higher sampling rate, noise would get in the way.
__________________
Last edited by lucid; 08-24-2008 at 12:04 PM.. |
|
Appreciate
0
|
08-24-2008, 11:47 AM | #34 | |
Brigadier General
233
Rep 3,302
Posts |
Quote:
__________________
"...it's not about the money and not about the brand of the car, it's about handling,performance and passion......And that, no other car has all together like an M3........when you talk about the most complete car the M is invincible." --Tony Kanaan. |
|
Appreciate
0
|
08-24-2008, 11:48 AM | #35 | |
Lieutenant Colonel
48
Rep 1,664
Posts |
Quote:
Most likely, 335i guys with the Procede and others, tapped into anything they wanted for data and more control. Maybe, Sharkedit and other tools will help with that on the M3 if ever released.
__________________
2008 E92 M3 Jerez Black,DCT,Fox Red ext,Prem,Tech,19", ipod/usb, CF roof and trim
2010 E91 328i Space Gray,Black int, M sport, most options 2007 Montego Blue 335i (retired) |
|
Appreciate
0
|
08-24-2008, 11:59 AM | #36 | |
Brigadier General
529
Rep 4,021
Posts
Drives: 2008 335xi Coupe
Join Date: Feb 2007
Location: The land where we kill baby seals
|
Quote:
This will be difficult, I think there are 3-4 CAN busses on the M3. You would need to find the bus that the MDCT sits on and then understand the data. I would call Swamp2's current work phase 1.....need easy repeatable data....nothing beats intercepting wires.
__________________
"Aerodynamics are for people who cannot build engines"......Enzo Ferrari
|
|
Appreciate
0
|
08-24-2008, 12:37 PM | #37 |
Major General
1072
Rep 8,008
Posts
Drives: i4M50
Join Date: Oct 2007
Location: No where fast
|
Based on the evidence that swamp has uncovered the whole lag after flicking the paddle is purely interest and has no real bearing on the performance of the car. Acceleration seems to be unbroken with DCT as all the manufactures have claimed and with surge there is a slight increase during that 200ms transition from one gear to the next, the only benefit with discovering to amount of lag at different levels is to learn why it's there at all.
We all know that SMGIII might quote a shift in 80ms but I doubt anyone believes that to be the case, I have read here that some reckon it's closer to 180ms which even using the none surge modes mean the DCT improves on this by up to 5~6 times. One can only imagine what difference it will bring to the likes of the M5/6, RS6 and other truly bonkers cars. I will be interested to see how quick the shift can actually be after additional testing, is it truly possible that DSG and DCT actual shift in the often quotes 8ms. That's mind blowing. |
Appreciate
0
|
08-24-2008, 01:17 PM | #38 |
Lieutenant Colonel
48
Rep 1,664
Posts |
Swamp, others who pitched in ideas, money, great work and info so far. Learning the mechanics and details of the transmission only aids in further enjoyment of the car!
__________________
2008 E92 M3 Jerez Black,DCT,Fox Red ext,Prem,Tech,19", ipod/usb, CF roof and trim
2010 E91 328i Space Gray,Black int, M sport, most options 2007 Montego Blue 335i (retired) |
Appreciate
0
|
08-24-2008, 01:59 PM | #39 |
Lieutenant General
609
Rep 10,407
Posts |
@foot, Art:
Of course the longer shifts in S4 are a big surprise. One of many classic cases of how the butt-o-meter can be very wrong. I am amazed that the D3 shifts, despite their very large mathematical jerk can be perceived so smoothly. I can't figure how there is enough time in a 30 ms shift to do anything with the extra engine and flywheel momentum, let along smoothly compensate for the extra torque at lower rpm. I may have choosen 500 Hz for S modes with some thought they would be faster, however I also did it just to beging a sensitivity analysis of test parameters. @lucid, Art: In terms of reporting results. Based both on the difficulty and subjectiveness of saying where the actual shifts begin and end, as well as the noted noise problems, providing a +/- is not only good basic scientific practice but REQUIRED. In particular look at the S4 shift. It almost looks like there could be a secondary peak making that shift maybe as much as another 50% longer than I quoted. As well look at the size of the oscillations at the beginning of the D3 data and just after that shift. A couple of points show data with negative acceleration. That surely is not real, no way. I have already tried some simple moving average filters and they do clean up the data a bit but not tremendously. You can't get anywhere near a 10 ms window for this since the D shifts will be mostly lost. There is much work to do on mounting, test procedures and post processing to get some truly quality data. I think the range of 300-500 Hz is just about right to capture not only the shift but some of the dynamics of the shift. |
Appreciate
0
|
08-24-2008, 02:02 PM | #40 | |
Banned
11
Rep 471
Posts |
Quote:
and the fastest numbers we've heard for the shift event is 8 ms (which seems like wishful thinking) even if we divide that into 2 sub-events of 4ms ~ 250hz, we need to sample at 500hz to capture ALL we need to replicate/model the event... sometimes more is not better, you may get noise that has nothing to do with the signal... either way, 500hz seems good, and some filtering would be nice... but if we assume an event of 20 ms, in 2 sub-events...~100hz, 200hz would do it...500hz would be a great margin of safety... 0.002 sec sampling in an event that takes 0.020, a factor of 10 or 10% resolution...that's pretty precise |
|
Appreciate
0
|
08-24-2008, 02:11 PM | #41 | |
Banned
11
Rep 471
Posts |
Quote:
I agree, a margin of error or tolerance is always required... what are the specs/accuracy on the accelerometer, A/D convertor, etc. (rhetorical question ) what are the bits of the signal? this will help us select the sampling rate... why couldn't you have -a? wouldn't the car 'slow down' (relatively) during the shift... 40 mph shift 39.9 mph the accelerate again? just curious |
|
Appreciate
0
|
08-24-2008, 02:44 PM | #42 | |
Major General
1072
Rep 8,008
Posts
Drives: i4M50
Join Date: Oct 2007
Location: No where fast
|
Quote:
Maybe the figure have gotten less over time, we all know how things change with each telling of the story but it would be a heck of a jump from 30ms down to 8ms, maybe something in between is nearer the true minimum time for a shift. |
|
Appreciate
0
|
08-24-2008, 02:59 PM | #44 | |
Lieutenant General
609
Rep 10,407
Posts |
Quote:
You can get the entire spec sheet in my post #10. The data sheet for the DAQ is attached. So many "accuracy" specs available it is tough to determine how much noise may be from the DAQ. Not my area of expertise. My gut tell me very, very little. In general a car can slow down during a shift (well D3 yes, S4, no, if we believe this data!). My point was that the car can not be decelerating before any shift initiation (as the car was nearly pegged!). This makes the mounting and noise problems very obvious. As well, as I mentioned previously, this accel was overloaded on input based on a typo on a supplied data sheet. I am going to get it repaired and recalibrated or replaced (on their dime!) soon, just to be sure. Lastly a truly regulated DC power supply like this one. Might improve noise as well. Any more donors?? |
|
Appreciate
0
|
Post Reply |
Bookmarks |
Thread Tools | Search this Thread |
|
|