| renhoek | 07-20-2006 08:14 PM |
Best gearshift points for track
�
�
Last session on the track was good, but I couldn't help thinking that revving my '02 WRX out to 6500+ rpm may not have been giving me the best straight line acceleration performance possible. I didn't want to spend too much time trying shifts at 6200, 6000 etc as my results via the butt dyno may have been too inaccurate.
Has anyone done any work on this? Anyone got a view on the best shift rpm for each gear to maximise acceleration ?
My old V8 did best shifting at 5500 in 1st and 5000rpm every other gear but don't really have a good feel for this car yet.
Has anyone done any work on this? Anyone got a view on the best shift rpm for each gear to maximise acceleration ?
My old V8 did best shifting at 5500 in 1st and 5000rpm every other gear but don't really have a good feel for this car yet.
| Mykl | 07-20-2006 08:49 PM |
Different car, but for reference I try to shift at ~6000 RPM's in the STi unless it's beneficial to hold it in that gear because I'm at the end of a straight away or the car's loaded down driving out of a turn.
| rkkwan | 07-20-2006 09:24 PM |
Theoretically, you want the RPM to be near the peak torque after you shift. You can calculate the shift point using the respective gear ratio, and peak torque for a stock WRX is 4,000rpm.
For example, for a 3-4 shift, shift point is 4,000 x 1.366 / 0.972 = 5,600.
At least that's the theory.
For example, for a 3-4 shift, shift point is 4,000 x 1.366 / 0.972 = 5,600.
At least that's the theory.
| ghschirtz | 07-20-2006 09:49 PM |
My 02 WRX power falls off after 6500, even a little before. I experimented a little yesterday and next track day I figure to shift just after 6000, not more than 6500, unless there is a reason to hold the gear and thereby avoid a shift.
There was another thread about this here or in the motor/gearbox forum, lots of people shifting at 7000 every time. Maybe the STi motor has the curve for that, but I don't think that makes sense, per my butt dyno.
rkkwan's computation makes that pretty clear. I guess I would add that if the motor is still pulling well between 5600 and, say, 6200, then maybe a 3-4 shift at 6200 would add further acceleration in a lower gear without losing much on the upshift. 4th gear at 4300 feels stronger than at 4000. Thrashing it to 7000 is clearly suboptimal.
George
There was another thread about this here or in the motor/gearbox forum, lots of people shifting at 7000 every time. Maybe the STi motor has the curve for that, but I don't think that makes sense, per my butt dyno.
rkkwan's computation makes that pretty clear. I guess I would add that if the motor is still pulling well between 5600 and, say, 6200, then maybe a 3-4 shift at 6200 would add further acceleration in a lower gear without losing much on the upshift. 4th gear at 4300 feels stronger than at 4000. Thrashing it to 7000 is clearly suboptimal.
George
| renhoek | 07-20-2006 10:38 PM |
[QUOTE=rkkwan]Theoretically, you want the RPM to be near the peak torque after you shift. You can calculate the shift point using the respective gear ratio, and peak torque for a stock WRX is 4,000rpm.
For example, for a 3-4 shift, shift point is 4,000 x 1.366 / 0.972 = 5,600.
At least that's the theory.[/QUOTE]
Sounds good. Actually, you raise an interesting point. What this means is that the gearchange rpm will be different for each gear. This does make some sense to me on your logic, but I'll probably just settle on one value - like 6000rpm.
One thing's for sure, there seems to be no point in pushing it up to 7000 unless you just want to avoid a change up due to a corner close by.
For example, for a 3-4 shift, shift point is 4,000 x 1.366 / 0.972 = 5,600.
At least that's the theory.[/QUOTE]
Sounds good. Actually, you raise an interesting point. What this means is that the gearchange rpm will be different for each gear. This does make some sense to me on your logic, but I'll probably just settle on one value - like 6000rpm.
One thing's for sure, there seems to be no point in pushing it up to 7000 unless you just want to avoid a change up due to a corner close by.
| oldmansan | 07-20-2006 10:56 PM |
It really depends on the car. I wouldn't shift an STI anything like my 02 WRX. The STI (stock) just doesn't pull up high as well. It makes power more quickly at lower rpms. My car is not stock, so this isn't a stock vs stock comparison. It also depends on how willing your car pulls to redline, and if after a certain rpm, it isn't pulling as quickly. Lastly, and most importantly imho, is where your car will be rpm-wise when you go for the next gear. This puts a lot of emphasis on gearing and where your turbo is at good/full boost.
San
San
| leecea | 07-21-2006 08:36 AM |
[QUOTE=renhoek]Sounds good. Actually, you raise an interesting point. What this means is that the gearchange rpm will be different for each gear. This does make some sense to me on your logic, but I'll probably just settle on one value - like 6000rpm.[/QUOTE]
I think some high-end shift lights allow different shift points per gear.
I think some high-end shift lights allow different shift points per gear.
| kwak | 07-21-2006 11:31 AM |
You need two things when determining shift points. You need to calculate the rpm drop for each gear. This gives you the size of the rpm window you place on your torque curve. And you need the torque curve, or even better get the dyno chart (the actual numbers at each rpm not the curve). Then take your rpm window and try to make it contain as many of the high torque and hp numbers as you can.
If hp starts falling off at high rpm, especially if they fall below the torque numbers if you were to shift, then you are waiting too long to shift. This is especially important if your car has lots of torque. Too many drivers are stuck thinking rpm = power.
If hp starts falling off at high rpm, especially if they fall below the torque numbers if you were to shift, then you are waiting too long to shift. This is especially important if your car has lots of torque. Too many drivers are stuck thinking rpm = power.
| sachilles | 07-21-2006 11:47 AM |
Make sure you look at the other end of that. Are you in the upper gear too early? If you are lugging in a higher gear, that won't help you out.
Just one more thing to consider.
Just one more thing to consider.
| drees | 07-21-2006 01:19 PM |
If anything I try to shift well below redline (around 6k) for a number of reasons:
* Power drops off quickly above 6k RPM stock
* 7k RPM is a lot harder on the engine than 6k RPM
* It's easy to overrev in the heat of the moment, so lower shift point gives more buffer room
As others mentioned, to determine optimal shift points you need a dyno chart and gear ratios to figure out what shift points give you the most area under the HP curve (don't worry about torque, it's HP which determines acceleration rate). Generally this means shifting at the point after HP starts falling where HP before the shift matches HP after the shift.
* Power drops off quickly above 6k RPM stock
* 7k RPM is a lot harder on the engine than 6k RPM
* It's easy to overrev in the heat of the moment, so lower shift point gives more buffer room
As others mentioned, to determine optimal shift points you need a dyno chart and gear ratios to figure out what shift points give you the most area under the HP curve (don't worry about torque, it's HP which determines acceleration rate). Generally this means shifting at the point after HP starts falling where HP before the shift matches HP after the shift.
| Patrick Olsen | 07-21-2006 04:43 PM |
[QUOTE=drees]As others mentioned, to determine optimal shift points you need a dyno chart and gear ratios to figure out what shift points give you the most area under the HP curve (don't worry about torque, it's HP which determines acceleration rate). Generally this means shifting at the point after HP starts falling where HP before the shift matches HP after the shift.[/QUOTE]
Exactly. When you shift, you want the HP in the next gear to be [i]no lower[/i] than the HP in the gear you're shifting out of. I've gone through this for my N/A 2.5L, using Excel and my dyno curve. In my case, it's always best to run it up to redline.
As an example, borrowing this graph from PDX Tuning's website:
[img]http://www.pdxtuning.com/images/dyno_plots/Stage1-wrx.gif[/img]
(and I'm considering the blue, stock curves). Let's say we're shifting from 3rd (1.366:1) to 4th (0.976:1). RPMs will drop to 71% of whatever our shift point is. Their graph only goes to about 6800rpm, and if we shift at 6800rpm we'll drop to ~4825rpm in 4th. At 6800rpm, we're making about 165hp or so (just eyeballing it on the graph), and at 4825rpm we'll be making a bit less than that, looks like around 155-160hp. That's good, our average power is staying pretty high if we shift at 6800rpm.
If we shifted at 6000rpm (175hp), we'd drop all the way down to ~4250rpm (~135hp). Average power is significantly reduced. It might [i]feel[/i] better, because you're closer to the torque peak, but I guarantee you if the same driver went down the same straightaway with those two techniques, he's going to be going faster when he gets to the braking zone if he winds it out.
If you extrapolated the end of the blue line out a bit more to the 7000rpm line, I think we'd find that shifting at 7000rpm would be worse than shifting at 6800rpm. Again, that's just eyeballing it on the graph.
Pat Olsen
'97 Legacy 2.5GT sedan
Exactly. When you shift, you want the HP in the next gear to be [i]no lower[/i] than the HP in the gear you're shifting out of. I've gone through this for my N/A 2.5L, using Excel and my dyno curve. In my case, it's always best to run it up to redline.
As an example, borrowing this graph from PDX Tuning's website:
[img]http://www.pdxtuning.com/images/dyno_plots/Stage1-wrx.gif[/img]
(and I'm considering the blue, stock curves). Let's say we're shifting from 3rd (1.366:1) to 4th (0.976:1). RPMs will drop to 71% of whatever our shift point is. Their graph only goes to about 6800rpm, and if we shift at 6800rpm we'll drop to ~4825rpm in 4th. At 6800rpm, we're making about 165hp or so (just eyeballing it on the graph), and at 4825rpm we'll be making a bit less than that, looks like around 155-160hp. That's good, our average power is staying pretty high if we shift at 6800rpm.
If we shifted at 6000rpm (175hp), we'd drop all the way down to ~4250rpm (~135hp). Average power is significantly reduced. It might [i]feel[/i] better, because you're closer to the torque peak, but I guarantee you if the same driver went down the same straightaway with those two techniques, he's going to be going faster when he gets to the braking zone if he winds it out.
If you extrapolated the end of the blue line out a bit more to the 7000rpm line, I think we'd find that shifting at 7000rpm would be worse than shifting at 6800rpm. Again, that's just eyeballing it on the graph.
Pat Olsen
'97 Legacy 2.5GT sedan
| kwak | 07-21-2006 08:26 PM |
[QUOTE=drees]As others mentioned, to determine optimal shift points you need a dyno chart and gear ratios to figure out what shift points give you the most area under the HP curve (don't worry about torque, it's HP which determines acceleration rate).[/QUOTE]Remember, Torque = HP * 5250 / RPM. What does this mean? Torque and HP are just two ways of looking at the same thing. Below 5250 rpm torque is the bigger number and above 5250 rpm hp is the bigger number.
So if we look at Pat's graph and take the 200 line for TQ/HP we see a nice large rpm range with huge power available in the range 3600 - 6800 rpm. Also notice that torque is a little higher than hp. The [I]power[/I] peak is at 4600 rpm. If you have a wide enough power band to not have to use high rpm, or bounce off the rev limiter, you are still using all the power the engine has but you aren't putting the wear on your engine that high rpm causes.
[QUOTE=Patrick Olsen]Exactly. When you shift, you want the HP in the next gear to be [i]no lower[/i] than the HP in the gear you're shifting out of.[/QUOTE]This is incorrect. When your shift, you want the [B]TQ[/B] in the next geat to be [i]no lower[/i] than the HP in the gear you're shifting out of.
The cars the race team I'm with has always had cars with big low end torque and power which slowly tapers off at high rpm, just like Pat's chart. Once we got our drivers to understand this they started out accelerating other drivers in the same type car because they were using more of the available [I]power[/I].
So if we look at Pat's graph and take the 200 line for TQ/HP we see a nice large rpm range with huge power available in the range 3600 - 6800 rpm. Also notice that torque is a little higher than hp. The [I]power[/I] peak is at 4600 rpm. If you have a wide enough power band to not have to use high rpm, or bounce off the rev limiter, you are still using all the power the engine has but you aren't putting the wear on your engine that high rpm causes.
[QUOTE=Patrick Olsen]Exactly. When you shift, you want the HP in the next gear to be [i]no lower[/i] than the HP in the gear you're shifting out of.[/QUOTE]This is incorrect. When your shift, you want the [B]TQ[/B] in the next geat to be [i]no lower[/i] than the HP in the gear you're shifting out of.
The cars the race team I'm with has always had cars with big low end torque and power which slowly tapers off at high rpm, just like Pat's chart. Once we got our drivers to understand this they started out accelerating other drivers in the same type car because they were using more of the available [I]power[/I].
| Patrick Olsen | 07-21-2006 10:09 PM |
[QUOTE=kwak]Remember, Torque = HP * 5250 / RPM. What does this mean? Torque and HP are just two ways of looking at the same thing.[/quote]
Mathematically, yes, you are correct. However, keep in mind that the relationship between HP and torque changes throughout the drivetrain because the RPMs change. The RPMs at the crank are not the same as the RPMs at the wheel. Ignoring drivetrain losses, in an ideal world the HP at the output of the crank is the same as the HP at the output of the transmission is the same as the output at the wheels. The torque will not be the same, though, so we can't compare torque in 3rd gear to torque in 4th gear - the change in gear ratios (and RPMs) means they can't be compared directly.
[quote=kwak]So if we look at Pat's graph and take the 200 line for TQ/HP we see a nice large rpm range with huge power available in the range 3600 - 6800 rpm. Also notice that torque is a little higher than hp. The [I]power[/I] peak is at 4600 rpm.[/quote]
Since you yourself said that torque is the bigger number at lower RPMs and HP is the bigger number at higher RPMs, how do you figure that the power peak is at 4600rpm? The HP curve is the blue curve that peaks farther to the right. Here's another look at it - the actual brochure graph:
[img]http://www.submariner.org/thepno95/Pictures/Subaru/Dyno%20curves/2002%20WRX.jpg[/img]
The power peak is at 6000rpm. The advertised numbers for the 2002-2005 WRXs is 227hp [b]@ 6000rpm[/b] and 217ft-lb [b]@ 4000rpm[/b].
[quote=kwak]This is incorrect. When your shift, you want the [B]TQ[/B] in the next geat to be [i]no lower[/i] than the HP in the gear you're shifting out of.[/quote]
:huh: That doesn't even make sense. HP and torque are two different things. Yes, they are mathematically related, but you can't say you want torque to be higher than HP in the next gear - you're comparing different units.
Tell you what, ignoring drivetrain losses, tell me what the torque to the wheels is at WOT, 4000rpm in 3rd gear and in 4th gear. We already know it's 217ft-lb at the crank, so what will it be at the wheels in those 2 gears? Then (again ignoring drivetrain losses) tell me what the HP to the wheels is at WOT, 4000rpm in 3rd and 4th gears. If you don't understand why I'm asking for this information then you won't understand why your statement above is gibberish.
[quote=kwak]The cars the race team I'm with has always had cars with big low end torque and power which slowly tapers off at high rpm, just like Pat's chart. Once we got our drivers to understand this they started out accelerating other drivers in the same type car because they were using more of the available [I]power[/I].[/QUOTE]
I'm not sure what that's supposed to mean. I just know you need to [i]maximize your average horsepower to maximize your acceleration.[/i]
Let's say we wanted to maximize torque in order to maximize acceleration. That would mean we'd want to shift at about 5000rpm (making about 210ft-lb at the crank on the brochure graph), and that would drop us to ~3500rpm (again making about 210ft-lb). Does anyone think that's going to give us the best acceleration? I sure hope not.
Pat
Mathematically, yes, you are correct. However, keep in mind that the relationship between HP and torque changes throughout the drivetrain because the RPMs change. The RPMs at the crank are not the same as the RPMs at the wheel. Ignoring drivetrain losses, in an ideal world the HP at the output of the crank is the same as the HP at the output of the transmission is the same as the output at the wheels. The torque will not be the same, though, so we can't compare torque in 3rd gear to torque in 4th gear - the change in gear ratios (and RPMs) means they can't be compared directly.
[quote=kwak]So if we look at Pat's graph and take the 200 line for TQ/HP we see a nice large rpm range with huge power available in the range 3600 - 6800 rpm. Also notice that torque is a little higher than hp. The [I]power[/I] peak is at 4600 rpm.[/quote]
Since you yourself said that torque is the bigger number at lower RPMs and HP is the bigger number at higher RPMs, how do you figure that the power peak is at 4600rpm? The HP curve is the blue curve that peaks farther to the right. Here's another look at it - the actual brochure graph:
[img]http://www.submariner.org/thepno95/Pictures/Subaru/Dyno%20curves/2002%20WRX.jpg[/img]
The power peak is at 6000rpm. The advertised numbers for the 2002-2005 WRXs is 227hp [b]@ 6000rpm[/b] and 217ft-lb [b]@ 4000rpm[/b].
[quote=kwak]This is incorrect. When your shift, you want the [B]TQ[/B] in the next geat to be [i]no lower[/i] than the HP in the gear you're shifting out of.[/quote]
:huh: That doesn't even make sense. HP and torque are two different things. Yes, they are mathematically related, but you can't say you want torque to be higher than HP in the next gear - you're comparing different units.
Tell you what, ignoring drivetrain losses, tell me what the torque to the wheels is at WOT, 4000rpm in 3rd gear and in 4th gear. We already know it's 217ft-lb at the crank, so what will it be at the wheels in those 2 gears? Then (again ignoring drivetrain losses) tell me what the HP to the wheels is at WOT, 4000rpm in 3rd and 4th gears. If you don't understand why I'm asking for this information then you won't understand why your statement above is gibberish.
[quote=kwak]The cars the race team I'm with has always had cars with big low end torque and power which slowly tapers off at high rpm, just like Pat's chart. Once we got our drivers to understand this they started out accelerating other drivers in the same type car because they were using more of the available [I]power[/I].[/QUOTE]
I'm not sure what that's supposed to mean. I just know you need to [i]maximize your average horsepower to maximize your acceleration.[/i]
Let's say we wanted to maximize torque in order to maximize acceleration. That would mean we'd want to shift at about 5000rpm (making about 210ft-lb at the crank on the brochure graph), and that would drop us to ~3500rpm (again making about 210ft-lb). Does anyone think that's going to give us the best acceleration? I sure hope not.
Pat
| drees | 07-22-2006 02:57 AM |
Thanks Pat, now I don't have to try to correct kwak. :)
| silentbob343 | 07-22-2006 06:45 AM |
I'm guessing he meant TQ instead of HP
| ghschirtz | 07-22-2006 12:56 PM |
Looks like shifting between 6-6500 is about right for best acceleration, to keep inside the meat of the torque curve. I think the goal should be to keep the motor between 4000 and 6500. 3500 might be okay in 2nd, but it definitely is logy in 3rd, needs to be higher.
That all said, sometimes the best acceleration and WOT are not what you need to make the best laptimes. Unbalancing the car at the wrong time can scrub speed off. The true "feel", as Donohue said, is the laptime. Bruce McLaren also tended to take it easy on some sections of a course so he could set up for the best launch onto the straighter, faster sections.
Now if I could just make all this lovely theory work all the time....
George
That all said, sometimes the best acceleration and WOT are not what you need to make the best laptimes. Unbalancing the car at the wrong time can scrub speed off. The true "feel", as Donohue said, is the laptime. Bruce McLaren also tended to take it easy on some sections of a course so he could set up for the best launch onto the straighter, faster sections.
Now if I could just make all this lovely theory work all the time....
George
| hotrod | 07-22-2006 06:36 PM |
This is one of the reasons drag racing is useful to you track guys, this has all been sorted out on the drag strip and there are several threads on it.
On the early 5MT transmission 2.0 L WRX's take low all the way to red line, shift the other gears at about 6700 - 6800 rpm depending on the gear.
A quick look at the gear splits explains why -- low second gear split is very wide so you make more torque at the axle at red line in low than you will at any rpm in second gear, in the other gears the gear splits are more reasonable, and by working with a dyno plot and your calculator you can quickly work out the shift points that give you the most area under the curve.
Larry
On the early 5MT transmission 2.0 L WRX's take low all the way to red line, shift the other gears at about 6700 - 6800 rpm depending on the gear.
A quick look at the gear splits explains why -- low second gear split is very wide so you make more torque at the axle at red line in low than you will at any rpm in second gear, in the other gears the gear splits are more reasonable, and by working with a dyno plot and your calculator you can quickly work out the shift points that give you the most area under the curve.
Larry
| kwak | 07-23-2006 02:40 AM |
[QUOTE=Patrick Olsen]Mathematically, yes, you are correct. However, keep in mind that the relationship between HP and torque changes throughout the drivetrain because the RPMs change.[/QUOTE]That's a confusing statement! The relationship between hp and torque changes at the location it is measured, like at the crank, based on rpm.
[QUOTE=Patrick Olsen]The HP curve is the blue curve that peaks farther to the right.[/QUOTE]I mistakingly assume we were talking about the red curve, not stock power numbers on the blue curve. In that context your 5000 rpm shift point example obviously makes no sense.
[QUOTE=Patrick Olsen]Tell you what, ignoring drivetrain losses, tell me what the torque to the wheels is at WOT, 4000rpm in 3rd gear and in 4th gear. We already know it's 217ft-lb at the crank, so what will it be at the wheels in those 2 gears? Then (again ignoring drivetrain losses) tell me what the HP to the wheels is at WOT, 4000rpm in 3rd and 4th gears. If you don't understand why I'm asking for this information then you won't understand why your statement above is gibberish.[/QUOTE]Gibber, gibber, gibber. :) What you are asking is commonly referred to as transmission output power which can be measured either as hp or torque.
[QUOTE=Patrick Olsen]I'm not sure what that's supposed to mean. I just know you need to [i]maximize your average horsepower to maximize your acceleration.[/i][/QUOTE]Since you brought up transmission output power you should know that you want to shift when and if the transmission output power curve for the current gear crosses the curve for the next gear. Right?
Here is an example calculated from an actual dyno sheet. The examples show the crossover points between the two gears.
Transmission output hp:
4th gear at 6000 engine rpm = 180
5th gear at 4800 engine rpm = 180
Transmission output torque:
4th gear at 5300 engine rpm = 200
5th gear at 3900 engine rpm = 200
Should we shift when the hp curves cross? When the torque curves cross? Some combination?
It is possible to look at torque and hp, and the relationship between their maximum values and the shape of the curves, to get a good idea of whether you should run to redline or not without going through all the math. The simple rule of thumb is if hp >= tq then just run to redline. It's when tq > hp that the issue needs to be looked at more carefully.
When might you find yourself "on track" where tq is enough more than hp and you should look more closely at shift points? In the road racing world this can happen when you run a stock turbo car with lots of boost, or when you have a large n/a engine and have to run a restrictor. You will never have to think about it if the peak tq number is above 5252 rpm, which by the math gaurantees tq < hp.
[QUOTE=Patrick Olsen]The HP curve is the blue curve that peaks farther to the right.[/QUOTE]I mistakingly assume we were talking about the red curve, not stock power numbers on the blue curve. In that context your 5000 rpm shift point example obviously makes no sense.
[QUOTE=Patrick Olsen]Tell you what, ignoring drivetrain losses, tell me what the torque to the wheels is at WOT, 4000rpm in 3rd gear and in 4th gear. We already know it's 217ft-lb at the crank, so what will it be at the wheels in those 2 gears? Then (again ignoring drivetrain losses) tell me what the HP to the wheels is at WOT, 4000rpm in 3rd and 4th gears. If you don't understand why I'm asking for this information then you won't understand why your statement above is gibberish.[/QUOTE]Gibber, gibber, gibber. :) What you are asking is commonly referred to as transmission output power which can be measured either as hp or torque.
[QUOTE=Patrick Olsen]I'm not sure what that's supposed to mean. I just know you need to [i]maximize your average horsepower to maximize your acceleration.[/i][/QUOTE]Since you brought up transmission output power you should know that you want to shift when and if the transmission output power curve for the current gear crosses the curve for the next gear. Right?
Here is an example calculated from an actual dyno sheet. The examples show the crossover points between the two gears.
Transmission output hp:
4th gear at 6000 engine rpm = 180
5th gear at 4800 engine rpm = 180
Transmission output torque:
4th gear at 5300 engine rpm = 200
5th gear at 3900 engine rpm = 200
Should we shift when the hp curves cross? When the torque curves cross? Some combination?
It is possible to look at torque and hp, and the relationship between their maximum values and the shape of the curves, to get a good idea of whether you should run to redline or not without going through all the math. The simple rule of thumb is if hp >= tq then just run to redline. It's when tq > hp that the issue needs to be looked at more carefully.
When might you find yourself "on track" where tq is enough more than hp and you should look more closely at shift points? In the road racing world this can happen when you run a stock turbo car with lots of boost, or when you have a large n/a engine and have to run a restrictor. You will never have to think about it if the peak tq number is above 5252 rpm, which by the math gaurantees tq < hp.
| renhoek | 07-23-2006 05:54 AM |
[QUOTE=Patrick Olsen]snipped
Let's say we wanted to maximize torque in order to maximize acceleration. That would mean we'd want to shift at about 5000rpm (making about 210ft-lb at the crank on the brochure graph), and that would drop us to ~3500rpm (again making about 210ft-lb). Does anyone think that's going to give us the best acceleration? I sure hope not.
Pat[/QUOTE]
Geez Pat. Now I feel stupid for asking, but I'll put my hand up and say I thought Yes! :D
And on this topic, there is a suggestion by Kwak (at least that's how i read it) that for a basically stock WRX, shifting gear to keep TQ high (@5000rpm in your example), would provide faster acceleration going up thru the gears.
Anyway, one good thing is that it appears that going much beyond 6000rpm is a waste of time. That's cut 1000rpm off my shift point already! :D
Got to also say a big thanks to all the contributions so far. I reckon I know a little bit about this stuff and the contributions to date have added some good practical comments for me (and others) to work with.
Let's say we wanted to maximize torque in order to maximize acceleration. That would mean we'd want to shift at about 5000rpm (making about 210ft-lb at the crank on the brochure graph), and that would drop us to ~3500rpm (again making about 210ft-lb). Does anyone think that's going to give us the best acceleration? I sure hope not.
Pat[/QUOTE]
Geez Pat. Now I feel stupid for asking, but I'll put my hand up and say I thought Yes! :D
And on this topic, there is a suggestion by Kwak (at least that's how i read it) that for a basically stock WRX, shifting gear to keep TQ high (@5000rpm in your example), would provide faster acceleration going up thru the gears.
Anyway, one good thing is that it appears that going much beyond 6000rpm is a waste of time. That's cut 1000rpm off my shift point already! :D
Got to also say a big thanks to all the contributions so far. I reckon I know a little bit about this stuff and the contributions to date have added some good practical comments for me (and others) to work with.
| hotrod | 07-23-2006 05:59 AM |
You want to shift at the point that the output power "at the axle" in the new gear (after the shift is completed) is equal to the output power in the old gear when you shifted. This is not always possible!
To get the maximum power under the curve you need to go well past peak power in all gears so the average through out the rpm range is the highest.
I find it easier to simply work with torque and look at the total accelerative thrust at the tire as you go through the rpm range.
Take the low - second shift.
Low gear has a ratio of 3.454:1 and the second 1.947:1 in the stock 5MT.
that gives a ratio split of 0.56367 times the rpm of the engine in low gear.
Lets say in low gear you shift at 7000 rpm iyour engine rpm will drop to 3945.9 rpm as you drop into second gear. Now it takes about 1/2 second or so to make the shift so the cars road speed will drop a bit while coasting so your true rpm drop will be a bit larger, but lets disregard that here.
If you extend the plot above, the engine will be making about 160 ft/lb of torque, but the torque mulitplication of the transmission increases that to about 552.64 ft lb at the tail shaft. Now looking at second gear its torque output is about 217 ft-lb at 4000 rpm, so multiply by 1.947 you get 422.499 ft-lb of torque at the tail shaft in second gear. This means the actual accelerating force at the tire contact patch will also drop by the same ratio. With a final drive ratio or 3.90, and using a 1 ft lever arm at the tire your actual accelerative thrust at the tire patch, in low gear at 7000 rpm will be 2155 lbs of thrust. When you complete the shift into second, the thrust at the tire patch will instantly drop to 1647.7 lbs. That means shifting to second gear from low gear drops your potential acceleration to only 76% of the acceleration you had in low gear at 7000 rpm.
If you compute the power at the axle you get about the same numbers.
Low gear
7000/3.454=2026.6 rpm
2026.6/3.9=519.6 rpm
2026.6/3.9=
519.6 rpm
hp = torque*rpm/5252
(519.6*2155)/5252 = 213.2 hp
Second gear
4000/1.947=2054rpm at the tail shaft
2054/3.9 = 526.8rpm at 1647.7ft-lbs
hp = (torque x rpm)/5252=165.3hp as you complete the shift to second gear.
That is why you need to hold low gear to the highest possible rpm without hitting fuel cut. The gear split on 1-2 shift is too large for the stock WRX rpm range, and you fall out of the torque curve as you go into second and deliver only 3/4 the power to the wheels at 4000 rpm in second gear, as you do at 7000 rpm in low.
This is also why, a laggy turbo that does not build boost quickly at low rpm hurts you so badly. On the stock WRX if you really push up the boost at the mid range rpm's [b]and[/b] have enough octane to avoid detonation, it significantly improves your drag strip time. By getting lots of torque at 4000 rpm you can fill up that torque/hp hole at the bottom of second gear right after the shift.
The same sort of thing happens in the other gears but I'll leave the computations to the reader.
Larry
[edit correction to second gear power output thanks to Patrick Olsen's proof reading]
To get the maximum power under the curve you need to go well past peak power in all gears so the average through out the rpm range is the highest.
I find it easier to simply work with torque and look at the total accelerative thrust at the tire as you go through the rpm range.
Take the low - second shift.
Low gear has a ratio of 3.454:1 and the second 1.947:1 in the stock 5MT.
that gives a ratio split of 0.56367 times the rpm of the engine in low gear.
Lets say in low gear you shift at 7000 rpm iyour engine rpm will drop to 3945.9 rpm as you drop into second gear. Now it takes about 1/2 second or so to make the shift so the cars road speed will drop a bit while coasting so your true rpm drop will be a bit larger, but lets disregard that here.
If you extend the plot above, the engine will be making about 160 ft/lb of torque, but the torque mulitplication of the transmission increases that to about 552.64 ft lb at the tail shaft. Now looking at second gear its torque output is about 217 ft-lb at 4000 rpm, so multiply by 1.947 you get 422.499 ft-lb of torque at the tail shaft in second gear. This means the actual accelerating force at the tire contact patch will also drop by the same ratio. With a final drive ratio or 3.90, and using a 1 ft lever arm at the tire your actual accelerative thrust at the tire patch, in low gear at 7000 rpm will be 2155 lbs of thrust. When you complete the shift into second, the thrust at the tire patch will instantly drop to 1647.7 lbs. That means shifting to second gear from low gear drops your potential acceleration to only 76% of the acceleration you had in low gear at 7000 rpm.
If you compute the power at the axle you get about the same numbers.
Low gear
7000/3.454=2026.6 rpm
2026.6/3.9=519.6 rpm
2026.6/3.9=
519.6 rpm
hp = torque*rpm/5252
(519.6*2155)/5252 = 213.2 hp
Second gear
4000/1.947=2054rpm at the tail shaft
2054/3.9 = 526.8rpm at 1647.7ft-lbs
hp = (torque x rpm)/5252=165.3hp as you complete the shift to second gear.
That is why you need to hold low gear to the highest possible rpm without hitting fuel cut. The gear split on 1-2 shift is too large for the stock WRX rpm range, and you fall out of the torque curve as you go into second and deliver only 3/4 the power to the wheels at 4000 rpm in second gear, as you do at 7000 rpm in low.
This is also why, a laggy turbo that does not build boost quickly at low rpm hurts you so badly. On the stock WRX if you really push up the boost at the mid range rpm's [b]and[/b] have enough octane to avoid detonation, it significantly improves your drag strip time. By getting lots of torque at 4000 rpm you can fill up that torque/hp hole at the bottom of second gear right after the shift.
The same sort of thing happens in the other gears but I'll leave the computations to the reader.
Larry
[edit correction to second gear power output thanks to Patrick Olsen's proof reading]
| Patrick Olsen | 07-23-2006 09:18 AM |
[QUOTE=kwak]I mistakingly assume we were talking about the red curve, not stock power numbers on the blue curve. In that context your 5000 rpm shift point example obviously makes no sense.[/quote]
Despite the fact that I specifically said "and I'm considering the blue, stock curves"? Regardless of which set of curves I pick off that graph, shifting at 5000rpm in an attempt to optimize acceleration is absolutely wrong!
[quote=kwak]Since you brought up transmission output power you should know that you want to shift when and if the transmission output power curve for the current gear crosses the curve for the next gear. Right?[/quote]
Now you're actually saying something that makes sense.
[quote=kwak]Here is an example calculated from an actual dyno sheet. The examples show the crossover points between the two gears.
Transmission output hp:
4th gear at 6000 engine rpm = 180
5th gear at 4800 engine rpm = 180
Transmission output torque:
4th gear at 5300 engine rpm = 200
5th gear at 3900 engine rpm = 200
Should we shift when the hp curves cross? When the torque curves cross? Some combination?[/quote]
When the HP curves cross, as I've already said a couple of times.
[quote=kwak]It is possible to look at torque and hp, and the relationship between their maximum values and the shape of the curves, to get a good idea of whether you should run to redline or not without going through all the math. The simple rule of thumb is if hp >= tq then just run to redline. It's when tq > hp that the issue needs to be looked at more carefully.[/quote]
The only curve that matters for optimizing acceleration is the horsepower curve.
[quote=renhoek]Geez Pat. Now I feel stupid for asking, but I'll put my hand up and say I thought Yes! :D
And on this topic, there is a suggestion by Kwak (at least that's how i read it) that for a basically stock WRX, shifting gear to keep TQ high (@5000rpm in your example), would provide faster acceleration going up thru the gears.
Anyway, one good thing is that it appears that going much beyond 6000rpm is a waste of time. That's cut 1000rpm off my shift point already! :D [/quote]
If the last comment is what you've gotten out of this so far, then you need to go back and re-read. You want to rev [i]beyond[/i] peak power (which is at 6000rpm) so that when you drop into the next higher gear you're maximizing the available power to accelerate - just as Larry said in his post.
You can prove this to yourself pretty easily if you've got some open highway or a dragstrip to work with. On the strip it'll be easy - just do a run shifting at 5000rpm (so you keep the engine right around peak torque), then do a run shifting at 6500-7000rpm. You'll be quicker and trap higher with the latter technique. If you've got an open highway, start at, say, 20mph in 2nd and accelerate to 90mph (insert standard warning about not exceeding the speed limit, etc etc). Time yourself doing it while shifting from 2-3 and 3-4 at 5000rpm, and then again at 6800rpm.
[quote=hotrod]If you compute the power at the axle you get about the same numbers.
Low gear
7000/3.454=2026.6 rpm
2026.6/3.9=519.6 rpm
2026.6/3.9=
519.6 rpm
hp = torque*rpm/5252
(519.6*2155)/5252 = 213.2 hp
Second gear
4000/3.454=1158 rpm at the tail shaft
1158/3.9 = 296.9 rpm at 1647.7 ft-lbs
hp= (torque x rpm)/5252=93.15 hp as you complete the shift to second gear.[/quote]
Thanks, Larry. As you said in your initial post, the drag race types already have this all figured out, and it seems like common sense to me, but obviously not everyone is familiar with all this.
Anyway, just wanted to point out your 2nd gear HP number is off because you used the 1st gear ratio rather than 2nd. It should say...
Second gear
4000/1.947=2054rpm at the tail shaft
2054/3.9 = 526.8rpm at 1647.7ft-lbs
hp = (torque x rpm)/5252=165.3hp as you complete the shift to second gear.
Note that 165hp @ 4000rpm is what you get if you look at the graph from the WRX brochure.
Pat
Despite the fact that I specifically said "and I'm considering the blue, stock curves"? Regardless of which set of curves I pick off that graph, shifting at 5000rpm in an attempt to optimize acceleration is absolutely wrong!
[quote=kwak]Since you brought up transmission output power you should know that you want to shift when and if the transmission output power curve for the current gear crosses the curve for the next gear. Right?[/quote]
Now you're actually saying something that makes sense.
[quote=kwak]Here is an example calculated from an actual dyno sheet. The examples show the crossover points between the two gears.
Transmission output hp:
4th gear at 6000 engine rpm = 180
5th gear at 4800 engine rpm = 180
Transmission output torque:
4th gear at 5300 engine rpm = 200
5th gear at 3900 engine rpm = 200
Should we shift when the hp curves cross? When the torque curves cross? Some combination?[/quote]
When the HP curves cross, as I've already said a couple of times.
[quote=kwak]It is possible to look at torque and hp, and the relationship between their maximum values and the shape of the curves, to get a good idea of whether you should run to redline or not without going through all the math. The simple rule of thumb is if hp >= tq then just run to redline. It's when tq > hp that the issue needs to be looked at more carefully.[/quote]
The only curve that matters for optimizing acceleration is the horsepower curve.
[quote=renhoek]Geez Pat. Now I feel stupid for asking, but I'll put my hand up and say I thought Yes! :D
And on this topic, there is a suggestion by Kwak (at least that's how i read it) that for a basically stock WRX, shifting gear to keep TQ high (@5000rpm in your example), would provide faster acceleration going up thru the gears.
Anyway, one good thing is that it appears that going much beyond 6000rpm is a waste of time. That's cut 1000rpm off my shift point already! :D [/quote]
If the last comment is what you've gotten out of this so far, then you need to go back and re-read. You want to rev [i]beyond[/i] peak power (which is at 6000rpm) so that when you drop into the next higher gear you're maximizing the available power to accelerate - just as Larry said in his post.
You can prove this to yourself pretty easily if you've got some open highway or a dragstrip to work with. On the strip it'll be easy - just do a run shifting at 5000rpm (so you keep the engine right around peak torque), then do a run shifting at 6500-7000rpm. You'll be quicker and trap higher with the latter technique. If you've got an open highway, start at, say, 20mph in 2nd and accelerate to 90mph (insert standard warning about not exceeding the speed limit, etc etc). Time yourself doing it while shifting from 2-3 and 3-4 at 5000rpm, and then again at 6800rpm.
[quote=hotrod]If you compute the power at the axle you get about the same numbers.
Low gear
7000/3.454=2026.6 rpm
2026.6/3.9=519.6 rpm
2026.6/3.9=
519.6 rpm
hp = torque*rpm/5252
(519.6*2155)/5252 = 213.2 hp
Second gear
4000/3.454=1158 rpm at the tail shaft
1158/3.9 = 296.9 rpm at 1647.7 ft-lbs
hp= (torque x rpm)/5252=93.15 hp as you complete the shift to second gear.[/quote]
Thanks, Larry. As you said in your initial post, the drag race types already have this all figured out, and it seems like common sense to me, but obviously not everyone is familiar with all this.
Anyway, just wanted to point out your 2nd gear HP number is off because you used the 1st gear ratio rather than 2nd. It should say...
Second gear
4000/1.947=2054rpm at the tail shaft
2054/3.9 = 526.8rpm at 1647.7ft-lbs
hp = (torque x rpm)/5252=165.3hp as you complete the shift to second gear.
Note that 165hp @ 4000rpm is what you get if you look at the graph from the WRX brochure.
Pat
| hotrod | 07-23-2006 09:31 AM |
Good catch, I was composing that while working, so not surprised I made a typo.
I will correct the post.
Larry
I will correct the post.
Larry
| Mike D02 | 07-23-2006 12:11 PM |
Here is a method I've used in the past.
Measure actual acceleration using whatever tool is currently available. I used a g- analyst years ago but it is old tech now. Basically any tool that will let you track and record accel and rpm either manually or with a log. Perhaps a g-tech device. Make a graph for 2nd and 3rd plotting accel against rpm. No need to graph 1st gear as you will use all available rpm as discussed previously and no need to graph 4th unless your track allows for the use of 5th gear (sorry, forgot about 6 speed cars). Overlay this on your gear/rpm chart and where to(and not)to shift should jump out. Besides it is rather cool to gather and use your own data.
Mike
Measure actual acceleration using whatever tool is currently available. I used a g- analyst years ago but it is old tech now. Basically any tool that will let you track and record accel and rpm either manually or with a log. Perhaps a g-tech device. Make a graph for 2nd and 3rd plotting accel against rpm. No need to graph 1st gear as you will use all available rpm as discussed previously and no need to graph 4th unless your track allows for the use of 5th gear (sorry, forgot about 6 speed cars). Overlay this on your gear/rpm chart and where to(and not)to shift should jump out. Besides it is rather cool to gather and use your own data.
Mike
| Patrick Olsen | 07-23-2006 02:38 PM |
[QUOTE=Mike D02]... and no need to graph 4th unless your track allows for the use of 5th gear (sorry, forgot about 6 speed cars).[/QUOTE]
Hey, don't forget those of us who don't have those 3.90 gears and only rev to 6500rpm (or the even more restrictive 6250rpm of the SOHC 2.5L), too - I shift to 5th at about 100mph!
Pat
Hey, don't forget those of us who don't have those 3.90 gears and only rev to 6500rpm (or the even more restrictive 6250rpm of the SOHC 2.5L), too - I shift to 5th at about 100mph!
Pat
| solo-x | 07-24-2006 11:04 AM |
key words here, "area under the curve". /thread
| subieworx | 07-24-2006 03:12 PM |
I tend to leave gearing out of it for my car as I am still making increasing HP at 7000 rpm. Why shift early when power is still building.
Không có nhận xét nào:
Đăng nhận xét