M3 vs Competition - factoring in Torque, Weight and Gearing

[mixja]

Originally Posted by footie

Very possible, as swamp and bruce can confirm I am not as technical as they are and base everything I know on expereince. AWD may be play it's part in this but I still feel the torque curve is more important than anything else.

Yes you are exactly right that the torque curve is most important, remember the graphs above are all torque curves, nothing to do with power. At the end of the day in the real world, the drop off in torque to wheels may be much higher than what is shown in the graphs, if that's the case then you might be right - but if they are as shown in the graph, then it's going to be better to shift at redline.

From my own experience, I just traded my Audi RS4, it definitely was quicker shifting at redline. But my old Subaru STI, was quicker shifting around 7000RPM (vs 8000RPM) for most gears (except 1st gear) as it had phenomenal mid-range torque and power really dipped past 7000RPM.

[bruce.augenstein@comcast.]

Originally Posted by footie

Maybe Audis are different because every single one I have owned accelerate as quickly if not quicker when shift occurs at peak power, revving the extra 500rpm has not benefit what so ever.

You want to be shifting into that thick torque band and until your torque is also far up the rev range then there is no benefit holding on to the gear.

I can only gauge from experience and that is what I have found on every occasion.

footie, I think we can agree on the basics. Then we'll go from there.

The basic idea that I think we can agree on in this context is that, at any given vehicle speed, more power means more acceleration. No ifs, ands or buts involved. Gearing doesn't matter. Torque doesn't matter. It's just power and weight. As a follow on to that, any car will accelerate more quickly from one speed to another speed if it's making more average power between those two speeds than if it's making less average power.

You can work out torque at the drive wheels if you like, but you'll find that, 100% of the time, more power equals more torque at the drive wheels, which then equates to more acceleration at any given speed.

If we can't agree on this, then just stop reading here, and we'll agree to disagree. In fact, even if we agree, you can probably just stop reading here, since, if you think about it, you can see that revving past the power peak to some degree will give you more average power at the wheels. This is because you'll be making more power after the shift than if you shifted earlier.

However, from a driver's perspective, the problem is that, in any given gear, more torque means more acceleration, so any car will accelerate the hardest in that gear while at the torque peak, and will accelerate with less urgency at the power peak than at the torque peak while in that gear.

Furthermore, there is a physiological component to this as well. Human G receptors are lousy at differentiating actual acceleration forces, but are very sensitive to rates of change. Therefore, when accelerating in a given gear, torque production is falling off precipitously past the power peak (it's dropping even faster than rpm is climbing), and as a driver, you can feel that drop off in acceleration. Then, when you grab the next gear, the car gets that jump from the engine's rotating inertia (it slows down to normalize to the next gear, expending that energy to momentarily accelerate the car), and the car is near the torque peak in that next gear, pulling about as well as it can in that gear. In other words, it feels healthy in that gear, as opposed to the dying swan feeling in the previous gear.

So footie, when you say that there's no benefit to revving past the power peak, you're relying on what torque on your backside tells you, and not a stopwatch or time slip at the drags.

"Torque is what you feel, but horsepower rules."

For the reasons mentioned, an M3 would be even quicker than it is now if it could rev another 1000 rpm or so past the power peak before the shift.

Bruce

Edit: PS - I am ignoring dynamic rotational inertia factors which vary gear by gear, since I don't feel like arguing the point anymore with lucid and Swamp. Suffice it to say that, this factor included, you still need to maximize average power at the wheels to get best acceleration.

[footie]

Bruce,

I was not totally doubt what you had written as I know you know more about this stuff than I do, as swamp keeps reminding me maths isn't my strong suit.

What I was explaining to mixja is that from experience in previous cars and in my most recent car (Audi S5) was that shifting at peak power proved to be as quick if not a little quicker, and this observation was based on trap speeds as you call it, finding that I was recording a quicker speed by shifting before redline.

I know all cars are different and what worked for the S5 may not work for something else.

[bruce.augenstein@comcast.]

Originally Posted by footie

Bruce,

I was not totally doubt what you had written as I know you know more about this stuff than I do, as swamp keeps reminding me maths isn't my strong suit.

What I was explaining to mixja is that from experience in previous cars and in my most recent car (Audi S5) was that shifting at peak power proved to be as quick if not a little quicker, and this observation was based on trap speeds as you call it, finding that I was recording a quicker speed by shifting before redline.

I know all cars are different and what worked for the S5 may not work for something else.

The physics work regardless. If your car worked better when short-shifted, it wasn't producing power according to Audi's published curve.

The MazdaSpeed 3 is a good example of this phenomenon. It's very quick, but forget about redline or anything close. The published specs don't meet the actual torque and power curves in that car.

Bruce

[footie]

The S5's peak power is at 7000rpm but the cut-off is 7450rpm. I found that stopping at the peak worked best, can't say if mine was different than any other.

[Hammer]

Originally Posted by footie

Bruce,

I was not totally doubt what you had written as I know you know more about this stuff than I do, as swamp keeps reminding me maths isn't my strong suit.

What I was explaining to mixja is that from experience in previous cars and in my most recent car (Audi S5) was that shifting at peak power proved to be as quick if not a little quicker, and this observation was based on trap speeds as you call it, finding that I was recording a quicker speed by shifting before redline.

I know all cars are different and what worked for the S5 may not work for something else.

If you argue with experience and bruce with maths, both of you might remember:

experience is subjectiv, math´s objective and (felt) results may vary between a subjectiv and a objectiv oberservation.

There maybe both of you are right

As you can imagine i agree with bruce: to accelerate quickest you ve to get the most avaible torque on the wheels all the time, it´s simply a combination of 5/6 choises (6 /7 gears) of the shifttiming which have to be optimised for the highest possible output. As far as i read the numbers it´s pretty simple: redlining means going the fastest ( in the stated cases).

You may to a empiric research by testing several shiftchoises and measuring the results. the problem might be to keep the conditions identical cause a human factor is part of the "experiment".

If the result doesnt match to the "maths" the maths might ve forgotten some aspects.

[lucid]

Originally Posted by bruce.augenstein@comcast.

Edit: PS - I am ignoring dynamic rotational inertia factors which vary gear by gear, since I don't feel like arguing the point anymore with lucid and Swamp. Suffice it to say that, this factor included, you still need to maximize average power at the wheels to get best acceleration.

Bruce, just caught your side note here. Nobody said that rotational inertia is not an issue or that its torque effects are not higher at lower gears where rotational accelerations are naturally higher. You know very well that Swamp and I have been arguing those effects scale linearly with rotational accelerations and are not as big as you claimed they are for the two vehicles we were considering for the drag race on that thread, which are operating in not so different rpm ranges as they accelerate. Never mind the highly likely possibility that the M3 has less rotational inertia to begin with.

I just wanted to set the positions straight as your note makes it sound like we were denying rotational inertia was relevant. I actually acknowledged that in that thread before you brought it up with you posts.

[bruce.augenstein@comcast.]

Originally Posted by lucid

Bruce, just caught your side note here. Nobody said that rotational inertia is not an issue or that its torque effects are not higher at lower gears where rotational speeds are naturally higher. You know very well that Swamp and I have been arguing those effects scale linearly with rotational speeds and are not as big as you claimed they are for the two vehicles we were considering for the drag race on that thread, which are operating in not so different rpm ranges as they accelerate. Never mind the highly likely possibility that the M3 has less rotational inertia to begin with.

I just wanted to set the positions straight as your note makes it sound like we were denying rotational inertia was irrelevant. I actually acknowledged that in that thread before you brought it up with you posts.

You guys think it's minor, but having Gillespie's book at my disposal, I know better.

Bruce

[lucid]

Originally Posted by bruce.augenstein@comcast.

You guys think it's minor, but having Gillespie's book at my disposal, I know better.

I am yet to pic up that book and locate the formula you are referencing. As I mentioned before my guess is that the formula is not simply relating torque to angular velocity, but other variables are involved. I will go to the library and request the book. Can you tell me the page number? Thanks.

P.S. I edited my previous post here as I just noticed I mistyped. Rotational "speed" should have been "acceleration", meaning, torque "losses" scale up linearly with rototional acceleration.

[bruce.augenstein@comcast.]

Originally Posted by lucid

...Can you tell me the page number? Thanks...

Just read chapter 2.

[lucid]

Originally Posted by bruce.augenstein@comcast.

Just read chapter 2.

I remembered you saying you were staring at a specific equation; that's why I asked. Anyway, I'll check it out.

[J08M3]

Is it just me or are all these charts just a bunch of wasted paper and time? In my opinion the only way to compare the cars is to get in and drive them. All these paper comparisons really don't mean anything. Spend all this time with these fancy formulas, talk about gearing, weight, torque, traction and whatever else, when all you really need to do is take the car for a ride. I mean when was the last time you flipped on the tv to watch a race and you see two guys sitting down comparing dyno charts? You don't, you see cars being driven. The only thing that counts is who crosses the finish line first.
-just my opinion on dyno charts and formulas.

[bruce.augenstein@comcast.]

Originally Posted by J08M3

Is it just me or are all these charts just a bunch of wasted paper and time?...

At the moment, it's just you.

Bruce

[swamp2]

Originally Posted by J08M3

Is it just me or are all these charts just a bunch of wasted paper and time? In my opinion the only way to compare the cars is to get in and drive them. All these paper comparisons really don't mean anything. Spend all this time with these fancy formulas, talk about gearing, weight, torque, traction and whatever else, when all you really need to do is take the car for a ride. I mean when was the last time you flipped on the tv to watch a race and you see two guys sitting down comparing dyno charts? You don't, you see cars being driven. The only thing that counts is who crosses the finish line first.
-just my opinion on dyno charts and formulas.

This is the a classic but massively uniformed point of view that represent to a great extent to death of true performance. If you care to look at the details you will find the that the butt-o-meter really pales in comparison with the accelerometer. In that same theme perhaps you can "inform" all of the F1 teams the telemetry/data acquisition is "over rated". They would laugh you all they way out of the pits. Similarly tell all the thousands of engineers from every specialty you can not even imagine, that care in ultimate minutia about the % of esoteric compounds in your tires, the bearing tribology in your your crankshaft, the flame front in your combustion chamber, the high speed data buses in your entire vehicle, the crush zones that protect your life... I could go on and on. Just in case you weren't aware, all of these folks care infinitely more about formulae and physics than what a car "feels like" when driving it. It seems fairly clear that you have no clue about the amazing correlation between you seat of the pants experience and the collective effort and knowledge captured by spreadsheet upon spreadsheet.

My 2 cents spoken....

Carry on....

[J08M3]

Of course the charts and formulas have a use. My point was directed at people who argue about one car being faster or better than another simply by comparing stats on paper.

[bruce.augenstein@comcast.]

Originally Posted by J08M3

Of course the charts and formulas have a use. My point was directed at people who argue about one car being faster or better than another simply by comparing stats on paper.

I don't know about "better", but, as long as the data is accurate (and as long as all the pertinent data is there), such charts can tell most of us a bunch more than just having reams of data to look at - and can give a very good indication about which car is potentially quicker.

In addition, you can't reliably tell which car is faster by a stint behind the wheel of each. As a for instance, I'm pretty sure the Merc will feel faster than the bimmer to the driver, but with the auto, the bimmer will be very, very close.

In addition, charts can tell you things you will never be able to ascertain with test drives. As an example, take a look at the first chart. The first (and perhaps most obvious) thing you can ascertain is that the M3, if it had, say, a 9500 rpm rev limit, would be even faster than it is now. You can tell this because there is a major loss in acceleration when you shift to second gear. Just look at that vertical gap! If you could rev higher in first, you'd be filling that gap in. Even though thrust in first would be falling off a cliff, you'd still be quicker than if you were in second. There would also be advantages (although lesser ones) in the next couple of gears.

The second thing is that, if you were willing to abuse the machinery*, the six-speed car would be quicker to 60 than the auto, and might in fact still be with (or even ahead of) the auto at the quarter mile point. The idea here is that an early lead is very hard to overcome because, in a quarter mile drag race, you have very little time to first reverse the trend, and then overtake. I know BMW says the auto is quicker, and I'm quite sure it is under their test procedures, but what that first chart tells me is that, everything done perfectly, the six-speed car has the potential early advantage that might in fact even hold up through a quarter mile. Of course as mentioned, you'd have to do everything perfectly, which is clearly not possible on any consistent basis, especially if you're me.

Another item is that, from, say, 70 and up, it's all MDCT, pretty close to all the time.

Bruce

*Abusing the machinery means, in this case, launching from whatever rpm the track will hold, going instantly to the floor, and holding rpm constant at the clutch point until vehicle speed catches up with engine speed, at which point the clutch is finally all the way out. Then, it's just dab-pull, dab-push, dab-pull, attempting to keep deceleration times during each shift to a minimum by using the tiniest travel of the clutch pedal consistent with the actual ability to complete the shift, all of this with right foot on the floor. None of this hitting the floor with the clutch pedal stuff. If you can keep deceleration times at or slightly below two tenths during shifts, my contention would be that you very well may be able to postpone the inevitable, and hold off the wonderbox long enough to take the win light.

All this from a chart...

PS - This is not meant to denigrate the bimmer box. It's clearly the wave of the future. And in fact, pulling off a run such as I described might in fact be cause for breaking out the champagne and calling it a day.

Edit: PPS - As an incredible example of what a simple chart can tell you, take a look at this. Unbelievable.

[T Bone]

Originally Posted by J08M3

Of course the charts and formulas have a use. My point was directed at people who argue about one car being faster or better than another simply by comparing stats on paper.

Only a fool would ignore valid and reliable data. (not calling you a fool)

Looking at the data below, it is obvious that the data set appears good and measuring the torque through the drivetrain shows the engines torque / power characteristics.

Also explains why the RS4's otherwise decent engine is wasted by attaching the horrible power sapping Quattro drivetrain.

No doubt Footie will jump in and try to use diversions and irrelevant arguments to say how great the RS4 is.... But all observable races the RS4 is slower than its competition and this dataset shows why....Quattro.

[J08M3]

Originally Posted by bruce.augenstein@comcast.

The second thing is that, if you were willing to abuse the machinery*, the six-speed car would be quicker to 60 than the auto, and might in fact still be with (or even ahead of) the auto at the quarter mile point. The idea here is that an early lead is very hard to overcome because, in a quarter mile drag race, you have very little time to first reverse the trend, and then overtake. I know BMW says the auto is quicker, and I'm quite sure it is under their test procedures, but what that first chart tells me is that, everything done perfectly, the six-speed car has the potential early advantage that might in fact even hold up through a quarter mile. Of course as mentioned, you'd have to do everything perfectly, which is clearly not possible on any consistent basis, especially if you're me.

As stated by you all the chart does is give you theoretical possibilities. So on paper a car might be faster or slower than another, but that doesn't mean it is when it's on the road being driven. I'm not saying there is no use for charts, but for comparing one car to another all you are doing is comparing theory, not necessarily "real world" facts.

[bruce.augenstein@comcast.]

Originally Posted by J08M3

As stated by you all the chart does is give you theoretical possibilities...

As actually stated by me, the chart gives you stuff you're unlikely to get in your real world of test drives. Not as a replacement, mind you. As a helpful addition.

Never mind, you're unlikely to get it no matter how much explanation is offered.

Bruce

[footie]

Originally Posted by T Bone

Also explains why the RS4's otherwise decent engine is wasted by attaching the horrible power sapping Quattro drivetrain.

No doubt Footie will jump in and try to use diversions and irrelevant arguments to say how great the RS4 is.... But all observable races the RS4 is slower than its competition and this dataset shows why....Quattro.

Sad to think that your insecurity against Audi products are there at every opportunity.

There is enough data out there is European tests to prove that Quattro cars, despite they additional weight and accelerate as well as anything powering only two wheels.

It's just like the old torque vs hp debate, each group have their opinions on which is better. (Not saying that quattro is better, only a different way of getting about very quickly).

[bruce.augenstein@comcast.]

Originally Posted by T Bone

...Also explains why the RS4's otherwise decent engine is wasted by attaching the horrible power sapping Quattro drivetrain...

T Bone, unless you're of the opinion that any and all systems that drive four wheels are horrible (based on the fact that they take more power to run, I suppose), then you are way off base on this. The fact is that awd systems offer significant advantages, and Audi's is a pretty damned good one. Especially the latest iterations. Perhaps not as effective in improving at-the-limit handling as, say, the GT-R's is, but pretty damned good nonetheless.

In fact, no matter the disparity in quoted numbers, do you think the GT-R would be as quick around the 'Ring with two wheel drive?

Bruce

[J08M3]

Originally Posted by bruce.augenstein@comcast.

As actually stated by me, the chart gives you stuff you're unlikely to get in your real world of test drives. Not as a replacement, mind you. As a helpful addition.

Never mind, you're unlikely to get it no matter how much explanation is offered.

Bruce

I get it, I just feel theoretical is irrelavant when you have the option to actually drive the cars and see how they perform against each other. I mean if you're unable to get those results in real world as you say then those results really are pointless. Don't get me wrong here, I'm all for dynoing a car and using whatever formulas necessary against itself to see how modifications are working or original design is working. But to take those numbers and use them against an entirely different car to say it is faster or slower is ridiculous. For example to say "my torque curve is bigger and flatter so my car is faster than yours" is ridiculous. Then it's just theoretical numbers when the real proof comes from getting in the cars and driving them against each other.