how do I improve torque?

[swamp2]

Originally Posted by pbonsalb

You must be one of the fast and furious crowd that lives on a race track. Why don't you datalog your driving since you like computers so much? You will find that about 99% of the time your car is running, it is below 5000 rpm.

Yes and then no. When I want to have fun, have a contest or boogie I'll simply use the right gear and the right rpm and stay above 5000 rpm exclusively.

If you want a performance mod that aids a car not being driven hard I don't know what to say other than (again) switch from an M3 to a 335. It is something like - "let's find a mod that offers the most improvement for my 30-80 time in 6th gear"... Ugh.

Originally Posted by pbonsalb

The vast majority of the 30-50 mph, 50-70 mph and 60-130 mph runs in your simulation show the 4.10 gears are best.

Again you are distorting the facts ("vast majority"). The BEST time is the only time from my perspective. If we take your reporting method magazines would report 0-60 times starting in 2nd gear. That's ludicrous. To be clear for this 6MT E92 M3 case looking at modded 4.10 gears. The stock car is:

-0.1 slower in 30-50
-about 0.4 faster in 50-70
-0.4 s slower in 60-130
-A wash in 1/4 time

If you want to call that "better" go ahead. I call it a wash. All for the paltry price of $3,000. Again knock yourself out.

Also I'm still waiting for any sort of real numbers. The burden of proof lies with you on this not me.

[swamp2]

Originally Posted by talontid

I've seen enough. I mean there's a reason dynos (specifically mustang) replicate real world data w/ load bearing rollers; there's a reason why street tuning is still great; there's a reason b/c countless variables not observed from a sim program. For 1/4 mile I will still bet, two cars, stock for stock, one w/ 4: 10 or lower gear (shorter gear (higher #) ) will be faster in the 1/4 mile. Shit, the OP was looking for best bang for the buck to increase torque?? And you still would spend 3x amount for a pulley?? OK, good advice.

Too bad dynos can not simulate an actual 1/4 mi run through the gears...

Have all of the hunches you want about which car will win. You can have your hunch, it is worth about as much as your opinion. I'll stick with physics over your opinion anyday. Again, just like above, the burden of proof lies with you. And since you don't get the whole statistics and variability thing yet prood requires multiple runs, under identical conditions with the SAME car. The results are simply too close to call without this type of methodology.

And I guess you meant quicker, not faster. Either way it would be a wash.

As the late Carl Sagan used to say, "Extraordinary claims require extraordinar evidence". You sir at this point have absolutely JACK.

Originally Posted by talontid

One last question though chief; you use a sex simulator before you do the real thing?

Great humor, great sarcasm.... uhhh not. If you are really interested in my sex life you really have bigger problems.

[swamp2]

Originally Posted by talontid

Shit, the OP was looking for best bang for the buck to increase torque?? And you still would spend 3x amount for a pulley?? OK, good advice.

Oh yeah this one too...

A FD increases peak wheel torque. It does not increase time averaged torque. The request was not specific

My point of the pulley recommendation is because most folks want more performance not more torque. Again, if all you care about is felt or peak in gear torque a FD ratio mod is "perfectly" acceptable. We covered that track of the discussion prior. Again the problem is that most folks will tell you it makes their car so much faster and in most cases it does not.

Last but not least pulleys will show an increase in wheel torque as well as power. Less parasitic losses yield more output.

[m3an]

Originally Posted by swamp2

Oh yeah this one too...

A FD increases peak wheel torque. It does not increase time averaged torque. The request was not specific

My point of the pulley recommendation is because most folks want more performance not more torque. Again, if all you care about is felt or peak in gear torque a FD ratio mod is "perfectly" acceptable. We covered that track of the discussion prior. Again the problem is that most folks will tell you it makes their car so much faster and in most cases it does not.

Last but not least pulleys will show an increase in wheel torque as well as power. Less parasitic losses yield more output.

not the OP of this thread so thanks for the suggestion

[pbonsalb]

Originally Posted by swamp2

Yes and then no. When I want to have fun, have a contest or boogie I'll simply use the right gear and the right rpm and stay above 5000 rpm exclusively.

Get in the real world. Life is not a computer simulation. It is not when you want to have a contest. It is when the contest happens. But in your simulation world, you are always running the perfect quarter mile and always driving your car around at 5000 rpm in the right gear waiting when the contest occurs.

I like decent acceleration most of the time, not just when I am on a dragstrip running to redline and shifting as fast as I can through the gears. Your simulations show that the 4.10 diff improves acceleration more that an underdrive pulley in most gears when doing the very common 30-50 mph and 50-70 mph acceleration as well as the less common 60-130 mph acceleration. That is daily driving.

I was the one saying long ago that the 335i is a great car and probably the faster car in daily driving, particularly when modded. It is always ready to go. No need to first be in the right gear and at the right rpm. I have driven stock and modded examples and have one of the N54 engines in one of my other cars (and had a JB4 on it for a while). I like them. Torque makes daily driving lots of fun.

5% better multiplication of it would be welcome. Its just doing a tiny bit more of what BMW thought was necessary to make these motors perform in these cars -- use aggressive gearing to multiply the low torque of the high revving engine.

If you just want to go fast on rare occasions, when you decide in advance and are making sure you are in the right gear and the right rpm, you should just get a 328i and run nitrous. That is essentially what you are doing now when you drive fast. You are planning ahead and "arming the car." Nitrous would be perfect for someone like you who either has no interest in performance or is going all out in the quarter mile.

[swamp2]

Originally Posted by pbonsalb

Get in the real world. Life is not a computer simulation. It is not when you want to have a contest. It is when the contest happens. But in your simulation world, you are always running the perfect quarter mile and always driving your car around at 5000 rpm in the right gear waiting when the contest occurs.

That is what M-DCT is for. Instant proper gear at any time.

I have no idea what "life is not a simulation" really means here. Are you saying the Matrix does not exist?

Originally Posted by pbonsalb

I like decent acceleration most of the time, not just when I am on a dragstrip running to redline and shifting as fast as I can through the gears. Your simulations show that the 4.10 diff improves acceleration more that an underdrive pulley in most gears when doing the very common 30-50 mph and 50-70 mph acceleration as well as the less common 60-130 mph acceleration. That is daily driving.

You really keep missing the forest for the trees or can't read the data. Comparing pulley to 4.10 gears:

• A pulley is basically a wash for 30-50 (0.03 s worse).
• The pulley is substantially better for 50-70 (0.5 s)
• The pulley is basically a wash for 60-130 (0.05 s better)

The pulley just happens to slightly win the 1/4 mi as well, time and trap...

Originally Posted by pbonsalb

I was the one saying long ago that the 335i is a great car and probably the faster car in daily driving, particularly when modded. It is always ready to go. No need to first be in the right gear and at the right rpm. I have driven stock and modded examples and have one of the N54 engines in one of my other cars (and had a JB4 on it for a while). I like them. Torque makes daily driving lots of fun.

If you just want to go fast on rare occasions, when you decide in advance and are making sure you are in the right gear and the right rpm, you should just get a 328i and run nitrous. That is essentially what you are doing now when you drive fast. You are planning ahead and "arming the car." Nitrous would be perfect for someone like you who either has no interest in performance or is going all out in the quarter mile.

Again, see above, M-DCT is the answer. As well, what you think is fun is purely an opinion, I absolutely love driving above 7500 rpm on a daily basis. Nitrous is by no means a good solution for me. Stick to your high torque, diff modded turbo lagging "brilliant" every day car and I'll stick to mine.

[swamp2]

Originally Posted by m3alabama

1-Swamp can you elaborate on the DCT adding equivelent to 20-50 hp? Is that purely due to shift time, or delivery of power or what?

Plenty of threads here on the forum. There is my simulation work that contributed to this finding as well as some empirical evidence from the guys running superchargers. We had some disagreements on the actual value but totally agreed on principal. The benefit is ENTIRELY due to shift time reductions. It really depends on how fast of MT shift time and which metric you want to compare 1/4 mi, 0-100, 60-130, etc. (more or less how many shifts are involved). If you can't find the threads let me know. I'm also assuming you use google to search the forum, it is the only way to fly. Use booleans and append your search with "site:m3post.com".

Originally Posted by m3alabama

2-People arguing for the gears are the same people who say the 335 is faster. Gears will give a harder "punch" for a shorter amount of time, just like the 335. 335 will kick you in the back because the torque at a lower rpm is higher and it comes on and off more quickly. WIth gears you FEEL more torque which is accurate but it does not last as long as without gears as you run through rpms quicker.

So looking at average or AUC numbers and looking at MOST point A to point B measures, they both will be pretty much identical given the oem car will have a bit less wheel torque but will be able to stay in a lower gear for a bit longer in each gear providing a comparable average advantage. The average is the same but the gears will "feel" like your pulling harder as you are-but not lasting as long.

All about what you feel-I really think performance wise they pretty much make little difference but if you like the feeling great-if you have 5k (with install) to blow and you like a small bump in the feeling of being pushed back in the seat-than its a fine mod.

Agree, that is restating what I already said.

Originally Posted by m3alabama

Clearly catless xpipe/HFC and race fuel/vishnu meth is good for 40 hp which destroys the pulley or the gears. Most people fret about these other mods but don't realize the oem tune has a timing target that cannot be reached without higher octane or meth-so focus on that for easy gains

I never said a pulley was THE best bang for the buck and it is probably not. It is pretty good bang for the buck though. I used a pulley because that other fellow mentioned using simulation to prove a pulley doesn't work or some such rubbish.

Isn't race fuel good for about 20 hp itself? It's quite hard to compute a hp/$ metric for race fuel though as the amount you will use is uncertain. It would be interesting to rank all of the common M3 performance mods on a hp/$ basis. Lucid did a very nice thread on a similar topic long ago with weight savings/$. Hopefully we would not have to revisit the topic of do RPI scoops actually add power for such an effort...

[pbonsalb]

When you plug your numbers into your simulator for the pulley, do they reflect that the UD pulley adds nothing at idle, maybe 5 of its 10 rwhp at 4200 rpm, and the full 10 rwhp not until 8400 rpm? When you do it for the 4.10 diff, does it reflect that the diff multiplies torque 5% better at every rpm? This may not be relevant when you are armed and ready as you drive around in the right gear at the right rpm waiting for your speed contest, but it is relevant in daily driving at mid rpm where you feel the 5% more torque multiplication -- the equivalent of 12 lbs -- more than you feel the 5 rwhp of the underdrive pulley. This reality is why the 4.10 diff blows away the underdrive pulley in your daily driving simulations for 30-50, 50-70 and 60-130 mph runs in any gear other than the absolutely perfect gear and perfect rpm (which we know you are always in, at least on your computer).

[pbonsalb]

I do agree that the DCT is the way to go, especially for this engine. It will shift faster than a human on any engine and improve acceleration through the gears for that reason alone. But on a high revving low torque motor where you have to be in the right gear at the right rpm to go really fast, the DCT will get the gear and rpm for you faster and more accurately than you can.

It also helps that the DCT is a closer ratio transmission -- 7 gears versus 6 gears for the manual. Part of the reason its speeds in some gears are better is that it has better torque multiplication in some gears. This has been discussed here before --

http://www.m3post.com/forums/showthread.php?t=157633

and part of the reason is that it shifts so fast that the normal issue of costing time because it is running out of gear while the longer geared car is still pulling in the same gear becomes a non issue.

While I would like a 6MT with a 4.10, I would like a DCT with a shorter diff even more. Let's get some DCT simulations going. Dinan, which according to one contributer to this thread has absolutely no idea what it is doing, offers two shorter diff ratio options for the DCT.

[pbonsalb]

Originally Posted by swamp2

• A pulley is basically a wash for 30-50 (0.03 s worse).
• The pulley is substantially better for 50-70 (0.5 s)
• The pulley is basically a wash for 60-130 (0.05 s better)

I think you are having trouble reading your own simulation chart in Post #73 for 30-50, 50-70 and 60-130 mph times. There are 15 speed contests listed in that chart. The pulley wins 3 of them. The 4.10 wins 12 of them. Which is faster in daily driving when you are not always in the perfect gear at the perfect rpm? The 4.10.

I live in the real world, not a computer simulation and freely admit that I do not start every speed contest in the perfect gear at the right rpm. Unfortunately, life just does not work that way.

Is your simulation for the pulley accurate? Simulations depend on user inputs. Did you plug in 12 hp across the board for the pulley or does it vary depending on rpm as it does in reality? If you are always in the 5000 to 8400 rpm range, the gains might be 6 rwhp to 10 rwhp over that spread, rather than a constant 10 rwhp. The 5% torque increase will be constant so that is easier to input.

[m3an]

I have left this thread, but wanna make one point.

OP was looking for BEST way to increase torque. For city driving, IMO a rear gear is a huge bang for the buck. Who gives a fuck about 1-20mph and 30-50 mph and 50-75?? I mean c'mon. Most cars are measured by 1/4 mile ET. I will still bet anyhting at all, that the car with gears will be faster through the 1/4; that's it. DONE. And to those who say, yes, you have to shift more and yada yada; don't forget that even if I have to shift one more time, the fact that I can get out of hole quicker should allow the geared car to hold that gap..

I can't believe this is a discussion. Keep in mind, the Germans have roads where high speeds are king, and in the US its' a bit diff. Gears will no doubt make the car quicker stoplight to stoplight.

[swamp2]

Been out of the loop and out of town for some time...

Originally Posted by pbonsalb

When you plug your numbers into your simulator for the pulley, do they reflect that the UD pulley adds nothing at idle, maybe 5 of its 10 rwhp at 4200 rpm, and the full 10 rwhp not until 8400 rpm?

Yes this is reflected, although it is largely irrelevant. When doing any WOT runs (which nearly all simulations of this type are) you are almost always in the 6000-8400 rpm range. Thus the data for what the engine is doing below that range is not relevant. Now of course if you choose the wrong gear, drive lazy and nearly lug the engine trying a 60-130 in 6th gear this fact will not be quite so true.

Originally Posted by pbonsalb

When you do it for the 4.10 diff, does it reflect that the diff multiplies torque 5% better at every rpm?

The differential gearing is treated exactly, regardless of the engine rpm, torque curve or the ratio used. You might consider brushing up on the fundamentals of vehicle performance and acceleration with some actual equations. They are quite insightful as to how and why these simulations actually work.

Originally Posted by pbonsalb

the 4.10 diff blows away the underdrive pulley in your daily driving simulations for 30-50, 50-70 and 60-130 mph runs in any gear other than the absolutely perfect gear and perfect rpm (which we know you are always in, at least on your computer).

Well "blows away" is certainly subjective, it may offer a small advantage that increases as the driver adopts a lazier and lazier approach to trying to actually have some or the best performance. Again, if you think the lazy driving numbers and benefit of this mod is justified on a performance/dollar ratio that is up to you. Anyone rational will disagree.

Again M-DCT... Argument pretty much over. It adds the equivalent of 20-50 hp depending on the contest and makes rectification of being in the wrong gear trivial. It also costs less than new FD gearing.

[swamp2]

Originally Posted by pbonsalb

I do agree that the DCT is the way to go, especially for this engine. It will shift faster than a human on any engine and improve acceleration through the gears for that reason alone. But on a high revving low torque motor where you have to be in the right gear at the right rpm to go really fast, the DCT will get the gear and rpm for you faster and more accurately than you can.

It also helps that the DCT is a closer ratio transmission -- 7 gears versus 6 gears for the manual. Part of the reason its speeds in some gears are better is that it has better torque multiplication in some gears. This has been discussed here before --

http://www.m3post.com/forums/showthread.php?t=157633

and part of the reason is that it shifts so fast that the normal issue of costing time because it is running out of gear while the longer geared car is still pulling in the same gear becomes a non issue.

While I would like a 6MT with a 4.10, I would like a DCT with a shorter diff even more. Let's get some DCT simulations going. Dinan, which according to one contributer to this thread has absolutely no idea what it is doing, offers two shorter diff ratio options for the DCT.

The benefits of M-DCT are primarily due to the incredibly short shift times and this benefit in totally independent of any particular feature of the engine (power, torque or shapes of those curves).

A secondary benefit is as discussed just above. The ability to get in THE RIGHT (best) gear for any performance contest quickly and in an automatic fashion.

I've simulated the piss out of M-DCT as have others here on the forum. Search around a bit.

In some gears at some speeds the M-DCT will have an advantage of the 6MT however, those specifics pretty well disappear across multiple gears. Such benefits would be pronounced on a track with a sprint between tight corners when using perhaps only 2nd and 3rd gear.

The issue of whether or not a FD gear ratio modification offers much benefit is also independent of whether or not the car has a DCT or MT.

[swamp2]

Originally Posted by pbonsalb

I think you are having trouble reading your own simulation chart in Post #73 for 30-50, 50-70 and 60-130 mph times. There are 15 speed contests listed in that chart. The pulley wins 3 of them. The 4.10 wins 12 of them. Which is faster in daily driving when you are not always in the perfect gear at the perfect rpm? The 4.10.

I live in the real world, not a computer simulation and freely admit that I do not start every speed contest in the perfect gear at the right rpm. Unfortunately, life just does not work that way.

As stated multiple times on the past, I feel it is only fair to compare the best possible times for any given modification. If you want to drive lazy why bother worrying about your actual times at all. It is the same argument about the 335i vs. the M3, one takes more shifting to enjoy and exploit.

Who in their right mind would quote a 0-60 or 1/4 mi starting in 2nd gear? Your argument is the same.

[swamp2]

Originally Posted by talontid

I have left this thread, but wanna make one point.
...
I will still bet anyhting at all, that the car with gears will be faster through the 1/4; that's it. DONE.

I'll take your bet, let's make it pinks (if you have a comparable vehicle). Ball is in your court to provide QUALITY data. Without that you have NOTHING but wild (and incorrect) speculation.

Physics just does not lie. There is no measurable benefit in the quarter mile for this modification to this particular vehicle. Again depending on your driving style and particular goals (if you drive "lazy") there may be some small benefits. Either way the gain provided for the cost is incredibly poor compared to many other modifications.

Now that is DONE..

[pbonsalb]

Your own data shows the 4.10 beats stock diff and UD pulley for 12 out of 15 speed contests in the real world ranges of 30-50, 50-70 and 60-130. Hard to argue against that. Only in the simulated world of fawless perfect shifts when you are starting fom the ideal rpm in the ideal gear does the UD pulley with stock diff prove faster than the 4.10 for these contests. The rest of the time, which is about 99 percent of the time, the 4.10 is faster.

I am honest and live in the real world. I do nt always drive at the perfect rpm and in the perfect gear for whatever I might need to do. This is one reason why simulations are of limited interest to me. At least your simulations did include the results for the less than perfectly conducted speed contests -- and the 4.10 won all or almost all of them.

Most of us do not drive our cars only in the 6000 to 8400 rpm range. If you want to make your car faster only in that range there is actually more you can do. How about some cams to trade off low to midrange power sine that is unimportant? The car would make more power in the very narrow and limited rpm range that is the only one of interest to you.

[m3an]

this swamp guy is ridiculous. I don't bet based on data. Come correct, two cars all stock, one with 4:10 rear and we can go 1/4 mile. I will def put an ass whoppin on ya!

[burakkose]

I'm a fan of "buy a 500 hp car if you want 500hp"
I can only suggest you to buy a car with torque close to your expectation

[swamp2]

Originally Posted by talontid

I have left this thread, but wanna make one point.

Originally Posted by talontid

this swamp guy is ridiculous. I don't bet based on data. Come correct, two cars all stock, one with 4:10 rear and we can go 1/4 mile. I will def put an ass whoppin on ya!

Well, you haven't left... You're funny. Just can't quit, eh?

If you call validated physics and engineering ridiculous then you better not even be driving a car. That is how YOUR CAR was conceived, designed, engineered and made possible. For that matter you better also go get a Ph.D. and prove that physics doesn't work...

Let's examine this a bit more closely. I have some evidence. You think physics based evidence is invalid. However, you have ZERO "real world" data whatsoever. You don't even have bad and insufficient real world data. You have unadulterated speculation and I am being ridiculous? Let's let basic logic decide who is being ridiculous here.

My bet was not based on simulation. You do not appear to have very good reading comprehension. It is based on running a real world test. However, you still do not understand what it takes to do accurate testing. You simply cannot use two cars for this kind of test. There can be more than enough variation in those cars due to natural variation in output as well as variation is output from their state of break in. There can also be enough variation from vehicle weight, tires and tire wear and a host of other relatively small factors. You obviously should not use 2 drivers as well. That effect would clearly overwhelm any conclusion one was trying to form. Now that being said, if you established the equality of two particular cars before the FD mod, and established this with some statistics and measured the certainty of their equality in performance then perhaps you could accomplish the test with two cars, but again probably not with 2 drivers. Almost for sure not with you and I driving.

Again, as I have been trying to drill into your head over and over: To test the effectiveness of a modification such as supercharging that makes a large power increase, such great care in testing is not required. However, when the modification will produce miniscule gains in a particular contest the only way to determine if it actually makes a difference is careful, methodical testing along with multiple tests to build up some basic statistics.

I know this is all likely completely lost on you but I know many others simply "get it".

Please, please do some more thinking on this before you knee jerk reply.

[m3an]

Originally Posted by swamp2

Well, you haven't left... You're funny. Just can't quit, eh?

If you call validated physics and engineering ridiculous then you better not even be driving a car. That is how YOUR CAR was conceived, designed, engineered and made possible. For that matter you better also go get a Ph.D. and prove that physics doesn't work...

Let's examine this a bit more closely. I have some evidence. You think physics based evidence is invalid. However, you have ZERO "real world" data whatsoever. You don't even have bad and insufficient real world data. You have unadulterated speculation and I am being ridiculous? Let's let basic logic decide who is being ridiculous here.

My bet was not based on simulation. You do not appear to have very good reading comprehension. It is based on running a real world test. However, you still do not understand what it takes to do accurate testing. You simply cannot use two cars for this kind of test. There can be more than enough variation in those cars due to natural variation in output as well as variation is output from their state of break in. There can also be enough variation from vehicle weight, tires and tire wear and a host of other relatively small factors. You obviously should not use 2 drivers as well. That effect would clearly overwhelm any conclusion one was trying to form. Now that being said, if you established the equality of two particular cars before the FD mod, and established this with some statistics and measured the certainty of their equality in performance then perhaps you could accomplish the test with two cars, but again probably not with 2 drivers. Almost for sure not with you and I driving.

Again, as I have been trying to drill into your head over and over: To test the effectiveness of a modification such as supercharging that makes a large power increase, such great care in testing is not required. However, when the modification will produce miniscule gains in a particular contest the only way to determine if it actually makes a difference is careful, methodical testing along with multiple tests to build up some basic statistics.

I know this is all likely completely lost on you but I know many others simply "get it".

Please, please do some more thinking on this before you knee jerk reply.

Dude you sound ridiculous, this is a car forum.. What do you say to all the drag racers in NHRA or IHRA competing against one another?? That they shouldn't b/c there's too many variables, that two diff people cannot drive diff cars? WOW!! I'm flawed at what I am reading.
That's a real scenario, Me and you, 2 m3s, one (mine) w/ 4:10 gears in a 1/4 mile. What's not to get?? That's called grudge racing, look it up and leave your physics sims to another forum, this is car talk; we don't need that crap here.

[m3an]

and just b/c it isn't an M3, who's to say I don't have real data (i.e. timeslips) from FD changes. You should go outside more and drive the car rather than read about it and make statements that would only apply in a perfect world/vacuum scenario

Later dude

[swamp2]

^ Your cluelessness continues and worsens. I have never made any statements about competitive drage racing. That's a great sport and one in which simulation is used! Last I recall this was a debate about the performance improvements from a FD mod to an M3, not some personal grudge match. Ugh...

For other cars, all bets are off. There are certainly cars where such modification will improve certain performance metrics. It is just for the M3 there is no benefit FOR THE 1/4 mi. Either way, simulation can very effectively answer those questions about the other cars and other ratios and which particular contests may be helped and which may be hurt.

If you were a longer term member here (or one who was more widely read and informed) you would find that the simulation work accomplished here on the forum (not just by me) is something appreciated by the community and has helped settle a variety of debates. Keep up the cluelessness buddy. Your foot just keeps going deeper and deeper down your throat.

All a drag race between you and I would accomplish is showing one or two things irrelevant to the actual debate. It could show who the better driver is (and trust me NO ONE cares a hoot about that). It could also show that M-DCT is superior to a 6MT (if that is what you drive).

All data (time slips) here are welcome. However, data showing very small effects will be relentlessly questioned as to their true meaning (statistical significance).