Another torque versus horsepower question for a track car
#31
Addict
Lifetime Rennlist
Member
Lifetime Rennlist
Member
I agree with much of the above, net net , you need a datalogger to tell you what is happening (whether with exhaust change or not), where and when it is happening, and then you are guaranteed to improve your lap times by making changes to driving habits and hardware, even if they are small.
I know this is very repetitive, but unless one has gone through a thorough analysis of his own runs turn by turn and sector by sector and then test different approaches and vs. other faster drivers, no theory can tell you what is better for you, that's the reason even the brightest pro Supercup drivers still sit down and listen to their race engineers telling them what they are doing right or wrong after each session.
In the strict terms of power (torque or HP) gains per the question you posted, the graph above is more accurate than a chassis or butt dyno, and then comes where you stand on the RPM range to make the best use of your particular car that is limited by the class regulations.
I know this is very repetitive, but unless one has gone through a thorough analysis of his own runs turn by turn and sector by sector and then test different approaches and vs. other faster drivers, no theory can tell you what is better for you, that's the reason even the brightest pro Supercup drivers still sit down and listen to their race engineers telling them what they are doing right or wrong after each session.
In the strict terms of power (torque or HP) gains per the question you posted, the graph above is more accurate than a chassis or butt dyno, and then comes where you stand on the RPM range to make the best use of your particular car that is limited by the class regulations.
#32
Rennlist Member
Thread Starter
First, thanks to everyone for all of the responses. I can always count on Rennlist to help me out. I especially appreciate the links, graphs and detailed explanations from Toby, Jean, Mark Kibort, Bill Verburg and Mark Dahlen. Also, thanks to Steve Weiner for the tremendous amount of time he spent talking to me about this mystifying subject.
Mark,
This is really helpful, makes a ton of sense and is right in line with what Steve Weiner and I talked about. Since I am not gifted with an F1 driver's talents, I think I'll err on the side of, well, "forgiveness" and choose an approach that works best for my skill level. One thing that I have learned is that, in my hands, a forgiving car is a faster car.
However, Chris Walrod (and Steve Weiner) brought up an interesting point. My car has never even had the timing checked, so this is an area that could use optimizing, torque versus horsepower issues related to cat bypasses notwithstanding. But what if I can optimize the timing of the cams to emphasize low end torque while the upper end is enhanced by the freer flow of the cat bypass? Would this help on the tighter tracks or help the car be more forgiving if I lose momentum in corners through mistakes?
I'm coming to the conclusion that Jean's suggestion of getting a data logger and looking at the data on various corners with and without a cat bypass (or other mod or adjustment) is the only way to go. At some point, I'm going to have to take this approach if I want to get even faster, but, frankly, I often barely have time to look at my damn video tapes, let alone spend hours comparing data.
I am contemplating getting a G-Timer to see how timed runs are affected by mod changes such as having a cat on or off and/or tuning my Supercups with removable baffles in a way that is similar to SuperTrapp mufflers but am now not convinced that this is an efficient effort in the right direction. The G-Timer can calculate peak horsepower, which I know is the crux of this whole discussion, but can also log 0-60 times, which may be helpful in showing real world performance. Thoughts?
Also, it's really clear that someone needs to put these (or various other bypasses) on a dyno to see where the torque and horsepower changes are occurring.
Again, a great thanks to everyone who posted and all of the superb suggestions and links.
Let's get to YOUR question, not F1 engine design or what works better when you can rev to 10,000 rpm. You have a car with an engine. You are not asking about major changes. The weight will remain about the same, the usable RPM will be the same and the gear ratios remain the same. So what is the impact of losing torque (and therefore HP) at lower RPM in order to gain at higher RPM?
You do not want to give up power in the normal range you run in. If your gearing is such that your lowest RPM on track is 4500, then losing torque/HP at 4700 is probably a bad trade off. It will hurt you a lot coming off corners.
But what if the trade off region is below your normal RPM range? Let's say you lose at 4000 to gain above 5500. In an error free lap, you will be faster. But here is where the trade off is - when you make a mistake and the revs fall to 3500, you will pay a much larger penalty for that small mistake than had you not modified the torque curve.
I have seen this a lot in DSR. A newer driver opts for a built engine, making more HP but it is peakier. The result is slower laps because of the mistakes.
So, you need to understand the RPM range you normally use and then decide if you are willing to pay a bigger price for mistakes on track or want a more forgiving car with better torque at lower RPM's.
Hope this help cut through the noise on this subject.
You do not want to give up power in the normal range you run in. If your gearing is such that your lowest RPM on track is 4500, then losing torque/HP at 4700 is probably a bad trade off. It will hurt you a lot coming off corners.
But what if the trade off region is below your normal RPM range? Let's say you lose at 4000 to gain above 5500. In an error free lap, you will be faster. But here is where the trade off is - when you make a mistake and the revs fall to 3500, you will pay a much larger penalty for that small mistake than had you not modified the torque curve.
I have seen this a lot in DSR. A newer driver opts for a built engine, making more HP but it is peakier. The result is slower laps because of the mistakes.
So, you need to understand the RPM range you normally use and then decide if you are willing to pay a bigger price for mistakes on track or want a more forgiving car with better torque at lower RPM's.
Hope this help cut through the noise on this subject.
This is really helpful, makes a ton of sense and is right in line with what Steve Weiner and I talked about. Since I am not gifted with an F1 driver's talents, I think I'll err on the side of, well, "forgiveness" and choose an approach that works best for my skill level. One thing that I have learned is that, in my hands, a forgiving car is a faster car.
However, Chris Walrod (and Steve Weiner) brought up an interesting point. My car has never even had the timing checked, so this is an area that could use optimizing, torque versus horsepower issues related to cat bypasses notwithstanding. But what if I can optimize the timing of the cams to emphasize low end torque while the upper end is enhanced by the freer flow of the cat bypass? Would this help on the tighter tracks or help the car be more forgiving if I lose momentum in corners through mistakes?
This, in addition to, potential weight savings at the heavy end of the car.
With the power gains of the cat bypass, wonder what advancing the cams -some- would do in terms of the overall benefit? Its believed the bypass nets an 8hp gain. Advancing the cams to bump the lower end of the power curve might get the best of both worlds. In other words, install the bypasses, advance the cam and net the same factory original 280hp with a healthier torque curve all with a weight savings.
With the power gains of the cat bypass, wonder what advancing the cams -some- would do in terms of the overall benefit? Its believed the bypass nets an 8hp gain. Advancing the cams to bump the lower end of the power curve might get the best of both worlds. In other words, install the bypasses, advance the cam and net the same factory original 280hp with a healthier torque curve all with a weight savings.
I am contemplating getting a G-Timer to see how timed runs are affected by mod changes such as having a cat on or off and/or tuning my Supercups with removable baffles in a way that is similar to SuperTrapp mufflers but am now not convinced that this is an efficient effort in the right direction. The G-Timer can calculate peak horsepower, which I know is the crux of this whole discussion, but can also log 0-60 times, which may be helpful in showing real world performance. Thoughts?
Also, it's really clear that someone needs to put these (or various other bypasses) on a dyno to see where the torque and horsepower changes are occurring.
Again, a great thanks to everyone who posted and all of the superb suggestions and links.
#33
Rennlist Member
Bill this is also misleading at first glance due to the trade off in any speed range. For example, 60-100 could be best for the taller ratio where 50 to 90 might be better for the lower gear set. Using the graph that was posted showing time spent in a gear at a speed range, you can see the comparison of the efffects of greater power for longer periods of time over a speed range.
(in that example, it shows a gain of HP in two areas, one for 1 second and a second area for 3 seconds, while the HP gain of the mod compared, only gained in a small 1 second area)
Gears dont make HP, and HP is the determinant factor in acceleration. It determines the rear wheel hp at any vehicle speed. It just incorporates more. (HP incorporates speed and torque)
again, it really boils down to maximizing HP-seconds (or time spent maximzing the max hp available at any speed), and this is true over several gear shifts.
So, in the end yes, the rear wheel torque determines the "push" in a form of linear force, and it is determined by HP. Engine torque is meaningless, unless you tie a RPM to it.
mk
(in that example, it shows a gain of HP in two areas, one for 1 second and a second area for 3 seconds, while the HP gain of the mod compared, only gained in a small 1 second area)
Gears dont make HP, and HP is the determinant factor in acceleration. It determines the rear wheel hp at any vehicle speed. It just incorporates more. (HP incorporates speed and torque)
again, it really boils down to maximizing HP-seconds (or time spent maximzing the max hp available at any speed), and this is true over several gear shifts.
So, in the end yes, the rear wheel torque determines the "push" in a form of linear force, and it is determined by HP. Engine torque is meaningless, unless you tie a RPM to it.
mk
the only thing that gives you a push to accelerate is torque. Hp is just a quantity derived from torque and time.
That said and hopefully understood, it is best to have torque at high rpm because then you can then use gearing to multiply the torque(and transform it from rotary motion to linear aceleration.
Again the thing that cause acceleration is the forward thrust(measured in pounds in the US) - air & mechanical resistance. Linear thrust can be calculated for a known torque curve(measured in ft-lb) and the gear ratios and the drive wheel loaded radius.
Here are 2 thrust curves, identical engines and tires for both, but w/ different gearing. the red is a g50/20 the blue a g50/30. The y axis is in pounds of thrust(push) the higher the more acceleration. Of corse aero resistance increasingly becomes a factor as speed goes up. That is not considered here.
That said and hopefully understood, it is best to have torque at high rpm because then you can then use gearing to multiply the torque(and transform it from rotary motion to linear aceleration.
Again the thing that cause acceleration is the forward thrust(measured in pounds in the US) - air & mechanical resistance. Linear thrust can be calculated for a known torque curve(measured in ft-lb) and the gear ratios and the drive wheel loaded radius.
Here are 2 thrust curves, identical engines and tires for both, but w/ different gearing. the red is a g50/20 the blue a g50/30. The y axis is in pounds of thrust(push) the higher the more acceleration. Of corse aero resistance increasingly becomes a factor as speed goes up. That is not considered here.
#34
Rennlist Member
One of the best HP / torque comparisons I've seen yet. Only one area that could be clarified, it was the area of shift points. Overall, the message was good. Its all about maximizing HP.
mk
mk
Mark,
Attached link to an interesting article about the relationship between hp, torque, acceleration from standing start, and acceleration at speed. (At the bottom of the page, there are a couple more links re: how to compare power curves, etc.)
http://craig.backfire.ca/pages/autos/horsepower
Cheers,
Toby
Attached link to an interesting article about the relationship between hp, torque, acceleration from standing start, and acceleration at speed. (At the bottom of the page, there are a couple more links re: how to compare power curves, etc.)
http://craig.backfire.ca/pages/autos/horsepower
Cheers,
Toby
#35
again, it really boils down to maximizing HP-seconds (or time spent maximizing the max hp available at any speed), and this is true over several gear shifts.
So, in the end yes, the rear wheel torque determines the "push" in a form of linear force, and it is determined by HP. Engine torque is meaningless, unless you tie a RPM to it.
mk
So, in the end yes, the rear wheel torque determines the "push" in a form of linear force, and it is determined by HP. Engine torque is meaningless, unless you tie a RPM to it.
mk
high hp comes from having torque at high rpm
your concept of hp-sec is probably a good one but harder to visualize
I(being a burnt piece of toast leftover from the 60s) prefer the more basic third grade concept of push at what ever rpm or speed, then consider how that interacts w/ the individual track situation.
I agree w/ Jean that accurate data logging is going to give you the most use-able picture of what is going on.
#36
Rennlist Member
Mark, I added the bypass pipes immediately to avoid burning out the cat and dropping the rear weight. I am running a Fabspeed full exhaust now and it did help top end, especially at the Glen. When we were at the Club race, I was having problems with my computer ( it had gotten wet at VIR), so I never saw great laps. I cleaned the computer with tuner wash and the ECU stop crapping out a week later. Anyways, while I was changing a collapsed lifter, I upgraded the exhaust and my times improved by almost a full second at the Glen.
#37
Rennlist Member
Its kind of a chicken and egg thing. Actually, it starts with energy (potential energy, like the fuel), and works its way down to the actual forces. Power is the potential and value for acceleration, and it creates the force that does the work, and does it at a rate based on the HP value. after all, HP is the rate of doing work.
HP is a rate of change of kinetic energy. It doesnt really have to be calculated or derived from torque and rpms, or force and speed. For example, a rolling chassis dyno can just measure the speed change (acceleration) of the drums and give a Hp value without measuring or knowing torque.
mk
HP is a rate of change of kinetic energy. It doesnt really have to be calculated or derived from torque and rpms, or force and speed. For example, a rolling chassis dyno can just measure the speed change (acceleration) of the drums and give a Hp value without measuring or knowing torque.
mk
#38
HP is a rate of change of kinetic energy. It doesnt really have to be calculated or derived from torque and rpms, or force and speed. For example, a rolling chassis dyno can just measure the speed change (acceleration) of the drums and give a Hp value without measuring or knowing torque.
mk
mk
mathematically work and KE and torque are all the same. They all use the same metric which in the Engineering system is measured in ft-lbs, dimensionally that is a distance x a force, again using enginering metrics
work in ft-lb = KE in ft-lbs = torque in ft-lbs = feet x pounds
hp in that same system is work/time. dimensionally that is a (distance x a force)/ time, in Engineering units hp = {(feet x pounds)/ sec} x a constant to change it into standard hp
so to continue the dimensional analysis your concept of hp-sec dimensionally simpfies into
hp x sec = {(feet x pounds)/ sec} x sec where the sec in the numerator cancels w/ the sec in the denominator leaving ft-lbs ie torque
#42
Rennlist Member
yes, hp-seconds, just like killowatt-hours, is a unit measure of work.
MK
MK
I just can't let this go
mathematically work and KE and torque are all the same. They all use the same metric which in the Engineering system is measured in ft-lbs, dimensionally that is a distance x a force, again using enginering metrics
work in ft-lb = KE in ft-lbs = torque in ft-lbs = feet x pounds
hp in that same system is work/time. dimensionally that is a (distance x a force)/ time, in Engineering units hp = {(feet x pounds)/ sec} x a constant to change it into standard hp
so to continue the dimensional analysis your concept of hp-sec dimensionally simpfies into
hp x sec = {(feet x pounds)/ sec} x sec where the sec in the numerator cancels w/ the sec in the denominator leaving ft-lbs ie torque
mathematically work and KE and torque are all the same. They all use the same metric which in the Engineering system is measured in ft-lbs, dimensionally that is a distance x a force, again using enginering metrics
work in ft-lb = KE in ft-lbs = torque in ft-lbs = feet x pounds
hp in that same system is work/time. dimensionally that is a (distance x a force)/ time, in Engineering units hp = {(feet x pounds)/ sec} x a constant to change it into standard hp
so to continue the dimensional analysis your concept of hp-sec dimensionally simpfies into
hp x sec = {(feet x pounds)/ sec} x sec where the sec in the numerator cancels w/ the sec in the denominator leaving ft-lbs ie torque
#43
kW is a unit of power as is hp
kw-hr a unit that measures the amount of energy used and can equivalently be measured in ft-lbs, joules, newton-meters, dyne-cm, ergs etc.
kw-hr is dimensionally equivalent to ft-lb, it is the same thing by a different name
#44
Burning Brakes
Join Date: Nov 2002
Location: 2 ends of the Pacific
Posts: 761
Likes: 0
Received 0 Likes
on
0 Posts
If you are running the bypass pipes as separate pipes instead of merging them (as in X-pipe) then you will lose so low end torque because you "lose" that tuning point that was present in the 993 cats which actually merges the two banks of cylinders.
Size of bupass pipes also determine if you lose low end torque (ie. too large a diameter).
For street type motors I would always try to merge to two banks with a cross pipe. This is especially so for 993 stock headers which are too short IMHO.
But I agree that if you do not remap then alot of the exhaust tuning that you do will not be optimised and it then becomes a sound tuning exercise not truly one of torque/power
Size of bupass pipes also determine if you lose low end torque (ie. too large a diameter).
For street type motors I would always try to merge to two banks with a cross pipe. This is especially so for 993 stock headers which are too short IMHO.
But I agree that if you do not remap then alot of the exhaust tuning that you do will not be optimised and it then becomes a sound tuning exercise not truly one of torque/power
#45
If you are running the bypass pipes as separate pipes instead of merging them (as in X-pipe) then you will lose so low end torque because you "lose" that tuning point that was present in the 993 cats which actually merges the two banks of cylinders.
Size of bupass pipes also determine if you lose low end torque (ie. too large a diameter).
For street type motors I would always try to merge to two banks with a cross pipe. This is especially so for 993 stock headers which are too short IMHO.
But I agree that if you do not remap then alot of the exhaust tuning that you do will not be optimised and it then becomes a sound tuning exercise not truly one of torque/power
Size of bupass pipes also determine if you lose low end torque (ie. too large a diameter).
For street type motors I would always try to merge to two banks with a cross pipe. This is especially so for 993 stock headers which are too short IMHO.
But I agree that if you do not remap then alot of the exhaust tuning that you do will not be optimised and it then becomes a sound tuning exercise not truly one of torque/power
On an even fire engine such as a 911 the goal to obtain max perfomance is to keep the exhaust gas flowing as fast as possible, part of that schema is maintaining max pulse separation, w/ separate banks the pulses are kept a 240 crankshaft degrees apart in a given bank at all times.
On an odd fire engine which does benefit from joining the banks the conection is there solely to relieve overloading inherent in the fireing order where 2 left and 2 right bank cylinders fire consecutively into the same collector. A US 90* V8 w/ a 90* crankshaft is an example of an odd fire engine, ideally the pulses in a given bank are kept 180 crankshaft degrees apart and either need a 180 degree exhaust(ala Ford GT40) which does keep the pulses 180 crankshaft degrees apart or a x - over exhaust which relieves the collector overpressure from the consecutive LL and RR pulses in a given collector (alternately the collectors can be made far larger but this undesireable from a packaging perspective
Ferrari and Indy cars use a an even fire V8 configuration by twisting the crankshaft instead of the exhaust pipes. Their flat 180 degree crank always keeps the exhaust pulses in a bank 180 degrees apart.