How much does horsepower weigh?
#1
03-05-2009, 11:24 AM
Join Date: Oct 2008
Location: Columbus, OH
Posts: 138
Rep Power: 23 
How much does horsepower weigh?
So, I have been thinking about modding my new/used 997. It seems like HP is fairly expensive for the 997, and everyone typically talks about adding HP-related mods. Due to my analyst background, I have this spreadsheet in my head that calculates HP per dollar spent. Of course, this would require dyno'ing the car after every upgrade. While it would not be perfect, it would provide some sort of data-based answer to the common questions 'What is the Best Mod for my 997?' or 'How would you spend $1000?'. So, I would like your thoughts on this approach. If enough people submitted their gains, we could start to hone in an decent answers.
The more I thought about this idea, the more I realized that speed is also a function of weight. If HP:Weight is the key metric, I wonder why I don't see a lot of 'weight reduction' threads here. That's when I started to wonder how much HP weighs! How many pounds would I have to remove from my 997 to equal the acceleration gain of 10HP?
Am I way off base here, or does this make any sense? Thoughts?
Nick
The more I thought about this idea, the more I realized that speed is also a function of weight. If HP:Weight is the key metric, I wonder why I don't see a lot of 'weight reduction' threads here. That's when I started to wonder how much HP weighs! How many pounds would I have to remove from my 997 to equal the acceleration gain of 10HP?
Am I way off base here, or does this make any sense? Thoughts?
Nick
#2
03-05-2009, 11:33 AM
Registered User
Join Date: Nov 2005
Location: Vancouver, BC, Canada
Posts: 1,551
Rep Power: 114 
Nick,
very interesting question that I have no answer for.
from my personal experience of owning a PCCB-equipped 997S before (forgot how much weight saving there), I don't find that it accelerates faster than another 997S with steel brakes.
now that I have a club coupe with 381hp, I find that it may be slightly faster than a 997S in the mid-range, but not by much.
I think the weight has to be significantly different for you to feel faster (at least 50 kg or more), and hp increase has to be more than 30 if not more to be noticable.
I remembered one time when I sat in my instructor's car going around the Mission track, the instructor complained that he felt slower with me in his car (I weigh about 180 pounds).
very interesting question that I have no answer for.
from my personal experience of owning a PCCB-equipped 997S before (forgot how much weight saving there), I don't find that it accelerates faster than another 997S with steel brakes.
now that I have a club coupe with 381hp, I find that it may be slightly faster than a 997S in the mid-range, but not by much.
I think the weight has to be significantly different for you to feel faster (at least 50 kg or more), and hp increase has to be more than 30 if not more to be noticable.
I remembered one time when I sat in my instructor's car going around the Mission track, the instructor complained that he felt slower with me in his car (I weigh about 180 pounds).
#3
03-05-2009, 11:39 AM
Interesting...
It sounds like a very interesting idea. I'm no auto metrics genius, but it seems that for this to work, you'd have to abstract the hp concept in to something that would be affected by power mods and/or weight mods.
I assume they would be the standard measures of 0-60, 40-70, 0-100 acceleration runs. Different metrics and ranges would be affected by hp versus torque changes.
Of course if you want to take this further, you could add stopping distance changes, which would be aided by weight reductions, but not power adds (unless they also reduce weight).
Off the top of my head, I suspect that the cost/performance slope would remain fairly low and linear until you hit the $13K range, which is about the price for a SC enhancement. That will likely put the single knee in the power-related curve.
Weight reduction is tougher. It will be interesting to hear from others how much weight needs to be reduced to knock .1sec off of the 0-60 time. I don't have the formulae at hand. The lowest cost return here is likely for you to go on a diet (if you have anything to spare)
Beyond this, I suspect this is a costly way to go.
But keep going. I'd sure like to see the results.
I assume they would be the standard measures of 0-60, 40-70, 0-100 acceleration runs. Different metrics and ranges would be affected by hp versus torque changes.
Of course if you want to take this further, you could add stopping distance changes, which would be aided by weight reductions, but not power adds (unless they also reduce weight).
Off the top of my head, I suspect that the cost/performance slope would remain fairly low and linear until you hit the $13K range, which is about the price for a SC enhancement. That will likely put the single knee in the power-related curve.
Weight reduction is tougher. It will be interesting to hear from others how much weight needs to be reduced to knock .1sec off of the 0-60 time. I don't have the formulae at hand. The lowest cost return here is likely for you to go on a diet (if you have anything to spare)
Beyond this, I suspect this is a costly way to go.But keep going. I'd sure like to see the results.
So, I have been thinking about modding my new/used 997. It seems like HP is fairly expensive for the 997, and everyone typically talks about adding HP-related mods. Due to my analyst background, I have this spreadsheet in my head that calculates HP per dollar spent. Of course, this would require dyno'ing the car after every upgrade. While it would not be perfect, it would provide some sort of data-based answer to the common questions 'What is the Best Mod for my 997?' or 'How would you spend $1000?'. So, I would like your thoughts on this approach. If enough people submitted their gains, we could start to hone in an decent answers.
The more I thought about this idea, the more I realized that speed is also a function of weight. If HP:Weight is the key metric, I wonder why I don't see a lot of 'weight reduction' threads here. That's when I started to wonder how much HP weighs! How many pounds would I have to remove from my 997 to equal the acceleration gain of 10HP?
Am I way off base here, or does this make any sense? Thoughts?
Nick
The more I thought about this idea, the more I realized that speed is also a function of weight. If HP:Weight is the key metric, I wonder why I don't see a lot of 'weight reduction' threads here. That's when I started to wonder how much HP weighs! How many pounds would I have to remove from my 997 to equal the acceleration gain of 10HP?
Am I way off base here, or does this make any sense? Thoughts?
Nick
#4
03-05-2009, 12:15 PM
For me I compare cars on Weight:HP ratio above all else. That drives classification and comparable performance in straight line acceleration.
Handling is much more difficult as you can really only measure that with lap time from same driver, same course, same conditions, which is even harder to find then 2 dyno's that give the same result.
Here is how it works for me. Stock 997C2S weight of 3031 and hp of 355 is an 8.82. Replace PSE with bypass pipes, save 40lbs now you are at 2991, assume the same 355hp and you are at 8.43. If we assume the bypass pipes increased hp, say 10hp, so now 2991 at 365 we have a 8.20.
Another way to do this is to establish a given acceleration rate. Find a flat stretch of road and do a 3rd gear only run from 2k to 7k and time it (Sport Chrono for the win!). Then repeat with the modified car. This is common and is an easy way to measure without the dyno cost or complications (and you get real airflow). To keep the speeds more reasonable, could do 2k to 6k, but you lose out on that top end power.
Dyno's are a waste IMO unless you use the same one on your car and establish baselines. To much value is put on "dyno results", not simple independent testing.
Handling is much more difficult as you can really only measure that with lap time from same driver, same course, same conditions, which is even harder to find then 2 dyno's that give the same result.
Here is how it works for me. Stock 997C2S weight of 3031 and hp of 355 is an 8.82. Replace PSE with bypass pipes, save 40lbs now you are at 2991, assume the same 355hp and you are at 8.43. If we assume the bypass pipes increased hp, say 10hp, so now 2991 at 365 we have a 8.20.
Another way to do this is to establish a given acceleration rate. Find a flat stretch of road and do a 3rd gear only run from 2k to 7k and time it (Sport Chrono for the win!). Then repeat with the modified car. This is common and is an easy way to measure without the dyno cost or complications (and you get real airflow). To keep the speeds more reasonable, could do 2k to 6k, but you lose out on that top end power.
Dyno's are a waste IMO unless you use the same one on your car and establish baselines. To much value is put on "dyno results", not simple independent testing.
#5
03-05-2009, 12:27 PM
Sources
The question was so intriguing it made me register after lurking for a long time. If you have not already, you could search Hot Rodder Mag. site or do a search on "unsprung weight horsepower gain". IIRC after a quick look there was a rule of thumb of 0.1 sec./100 lbs lost in the 1/4 mile. (Just quoting and have no idea how valid that number is.) I'm just guessing but I 'm not sure that weight reduction versus speed gains will be a linear relationship - or if it is it may not directly translate to performance gains (e.g. you lighten up the rear end of a 911 and you get a faster car - the same hp moving lesser weight - but this might adversely effect traction or some other variable). Hopefully others with more knowledge will reply.
#6
03-05-2009, 12:50 PM
As a former Lotus owner, I can attest to the fact that THE best way to improve performance is a weight reduction program. Decreasing weight will improve acceleration, braking and cornering. Increasing HP on the other hand, only improves acceleration. As far as weight is concerned go for the unsprung weight like wheels. I had 16/17 in Volk TE 37's and they weighed 12 and 15 pounds. The Turbo wheels on my car are a porky 28 pounds...
#7
03-05-2009, 01:11 PM
This makes a lot of sense. The only question that remains is the relationship between your coefficient and performance. If the acceleration of a 8.82 'S' is 'X', what is the acceleration of an 8.43 'S'?
Even after the test you cite, what happens if you drop more weight or add more power? What's the shape of the curve?
Not criticizing, just curious.
Even after the test you cite, what happens if you drop more weight or add more power? What's the shape of the curve?
Not criticizing, just curious.
For me I compare cars on Weight:HP ratio above all else. That drives classification and comparable performance in straight line acceleration.
Handling is much more difficult as you can really only measure that with lap time from same driver, same course, same conditions, which is even harder to find then 2 dyno's that give the same result.
Here is how it works for me. Stock 997C2S weight of 3031 and hp of 355 is an 8.82. Replace PSE with bypass pipes, save 40lbs now you are at 2991, assume the same 355hp and you are at 8.43. If we assume the bypass pipes increased hp, say 10hp, so now 2991 at 365 we have a 8.20.
Another way to do this is to establish a given acceleration rate. Find a flat stretch of road and do a 3rd gear only run from 2k to 7k and time it (Sport Chrono for the win!). Then repeat with the modified car. This is common and is an easy way to measure without the dyno cost or complications (and you get real airflow). To keep the speeds more reasonable, could do 2k to 6k, but you lose out on that top end power.
Dyno's are a waste IMO unless you use the same one on your car and establish baselines. To much value is put on "dyno results", not simple independent testing.
Handling is much more difficult as you can really only measure that with lap time from same driver, same course, same conditions, which is even harder to find then 2 dyno's that give the same result.
Here is how it works for me. Stock 997C2S weight of 3031 and hp of 355 is an 8.82. Replace PSE with bypass pipes, save 40lbs now you are at 2991, assume the same 355hp and you are at 8.43. If we assume the bypass pipes increased hp, say 10hp, so now 2991 at 365 we have a 8.20.
Another way to do this is to establish a given acceleration rate. Find a flat stretch of road and do a 3rd gear only run from 2k to 7k and time it (Sport Chrono for the win!). Then repeat with the modified car. This is common and is an easy way to measure without the dyno cost or complications (and you get real airflow). To keep the speeds more reasonable, could do 2k to 6k, but you lose out on that top end power.
Dyno's are a waste IMO unless you use the same one on your car and establish baselines. To much value is put on "dyno results", not simple independent testing.
Trending Topics
#8
03-05-2009, 01:41 PM
Nick,
This is such a great question but I am afraid it is impossible to answer unless you have a standardized form for testing or an equation. I saw the .01 per 100lbs in a 1/4 mile, which I have heard before. But does the distance have turns or in a straight line. In our Koni Challenge cars 100 pounds is a big deal. The other factor is weight placement; high weight hurt us more than low weight.
Grand Am dictated where our weight went on the Koni cars, which was in a box where the passenger seat would be. Rather high weight proved to be damaging; and when we complained about having too much weight thrown at us we would take it out to show the officials the time difference which was usually about 3 seconds.
On our Rolex GT3 they do not mandate where we put the weight so we place it in sheet form under the fuel cell. That proves to be good weight that helps the cars performance with the proper set up.
Not trying to complicate things but just trying to say there are many variables in trying to answer this great question. But I will say the biggest free weight out there is for the driver of the car to monitor their own weight. Love the question, thanks for asking it.
This is such a great question but I am afraid it is impossible to answer unless you have a standardized form for testing or an equation. I saw the .01 per 100lbs in a 1/4 mile, which I have heard before. But does the distance have turns or in a straight line. In our Koni Challenge cars 100 pounds is a big deal. The other factor is weight placement; high weight hurt us more than low weight.
Grand Am dictated where our weight went on the Koni cars, which was in a box where the passenger seat would be. Rather high weight proved to be damaging; and when we complained about having too much weight thrown at us we would take it out to show the officials the time difference which was usually about 3 seconds.
On our Rolex GT3 they do not mandate where we put the weight so we place it in sheet form under the fuel cell. That proves to be good weight that helps the cars performance with the proper set up.
Not trying to complicate things but just trying to say there are many variables in trying to answer this great question. But I will say the biggest free weight out there is for the driver of the car to monitor their own weight. Love the question, thanks for asking it.
#9
03-05-2009, 01:49 PM
This makes a lot of sense. The only question that remains is the relationship between your coefficient and performance. If the acceleration of a 8.82 'S' is 'X', what is the acceleration of an 8.43 'S'?
Even after the test you cite, what happens if you drop more weight or add more power? What's the shape of the curve?
Not criticizing, just curious.
Even after the test you cite, what happens if you drop more weight or add more power? What's the shape of the curve?
Not criticizing, just curious.
For me I compare cars on Weight:HP ratio above all else. That drives classification and comparable performance in straight line acceleration.
Handling is much more difficult as you can really only measure that with lap time from same driver, same course, same conditions, which is even harder to find then 2 dyno's that give the same result.
Here is how it works for me. Stock 997C2S weight of 3031 and hp of 355 is an 8.82. Replace PSE with bypass pipes, save 40lbs now you are at 2991, assume the same 355hp and you are at 8.43. If we assume the bypass pipes increased hp, say 10hp, so now 2991 at 365 we have a 8.20.
Another way to do this is to establish a given acceleration rate. Find a flat stretch of road and do a 3rd gear only run from 2k to 7k and time it (Sport Chrono for the win!). Then repeat with the modified car. This is common and is an easy way to measure without the dyno cost or complications (and you get real airflow). To keep the speeds more reasonable, could do 2k to 6k, but you lose out on that top end power.
Dyno's are a waste IMO unless you use the same one on your car and establish baselines. To much value is put on "dyno results", not simple independent testing.
Handling is much more difficult as you can really only measure that with lap time from same driver, same course, same conditions, which is even harder to find then 2 dyno's that give the same result.
Here is how it works for me. Stock 997C2S weight of 3031 and hp of 355 is an 8.82. Replace PSE with bypass pipes, save 40lbs now you are at 2991, assume the same 355hp and you are at 8.43. If we assume the bypass pipes increased hp, say 10hp, so now 2991 at 365 we have a 8.20.
Another way to do this is to establish a given acceleration rate. Find a flat stretch of road and do a 3rd gear only run from 2k to 7k and time it (Sport Chrono for the win!). Then repeat with the modified car. This is common and is an easy way to measure without the dyno cost or complications (and you get real airflow). To keep the speeds more reasonable, could do 2k to 6k, but you lose out on that top end power.
Dyno's are a waste IMO unless you use the same one on your car and establish baselines. To much value is put on "dyno results", not simple independent testing.
#10
03-05-2009, 02:57 PM
Good points and to be honest I can't say as I didn't measure the acceleration of my C2S stock, so now that it's changed it's going to be hard to get a baseline on my car. I was simply offering a test method for the OP to get some type of statistics going. But to come up with a universal ratio for weight saved to converted hp gain is i think impossible, to many variables.
#11
03-05-2009, 03:04 PM
An acceleration formula
1/4 mile time = ((weight/hp)^(1/3))*5.8144
3200/385 = 8.31
8.31^(1/3) = 2.025
2.025 * 5.8144
11.77
Theoretical max presuming perfect gearing and launch. Looks about 1/2 sec fast based upon my more-or-less C2S example . . .
YMMV
1/4 mile time = ((weight/hp)^(1/3))*5.8144
3200/385 = 8.31
8.31^(1/3) = 2.025
2.025 * 5.8144
11.77
Theoretical max presuming perfect gearing and launch. Looks about 1/2 sec fast based upon my more-or-less C2S example . . .
YMMV
#14
03-05-2009, 03:56 PM
These types of questions have been debated for years here and at RennList. I've seen every possible mod debated and posed in a $ / gain method, as well as weight reduction. This is one reason why when I was shopping for a GT3 I found the perfect one - a zero option car. That meant no heavy Xenons (something like 14 lbs additional), no PCM, no power seats, etc. Many threads have talked about the effects of sprung and unsprung weight. Like those OEM GT3 wheels - they are terrible, especially the 32 lb rears. Compare the 32 to my 22 lb forged rears and that 10 lbs is like 10 HP equivalent. Rotors are HUGE, just like the wheels / tires. Some have even advocated titanium lugs because they are unsprung / rotational mass.
As to the most cost effective mods, they are all laid out here and at the sister forum RL. Air filters really do nothing and can cause damage (oil fouling the MAF), software really does work if it's good, headers / cats can improve flow, but there's always a risk of losing torque to high-end HP.
As to the most cost effective mods, they are all laid out here and at the sister forum RL. Air filters really do nothing and can cause damage (oil fouling the MAF), software really does work if it's good, headers / cats can improve flow, but there's always a risk of losing torque to high-end HP.
