How do you calculate HP from data logs?
#21
Registered User
Thread Starter
iTrader: (4)
Join Date: Jun 2006
Location: Dearborn, MI
Posts: 306
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by Q45tech
Dynos usually sample the drum speed at 1/1000 second intervals and calculate drum acceleration over this interval. The drum mass [weight] is known very accurately.
Originally Posted by Q45tech
Due to slippage in drive train the engine rpm is not tied to the vehicle speed precisely.
A larger # of samples per second would only increase resolution, or accuracy, of each data point. If a 5th gear run to redline takes, i don't know, 10 seconds, 160 data points (16 per second) should be plenty to show a HP and torque curve over the RPM range.
#22
Registered User
iTrader: (4)
Join Date: Sep 2005
Location: SoFla
Posts: 437
Likes: 0
Received 0 Likes
on
0 Posts
do you know:
precise mass of vehical? Think fluids, driver, any mods, etc.
drag coefficient of your car? consider altitude of your test, temperature, humidity, wind speed and direction
how was your traction? did you floor it without spinning the tires at all?
These are just some of the points where a dyno has an advantage. I'd say go spend the $60-70 and do a dyno pull.
precise mass of vehical? Think fluids, driver, any mods, etc.
drag coefficient of your car? consider altitude of your test, temperature, humidity, wind speed and direction
how was your traction? did you floor it without spinning the tires at all?
These are just some of the points where a dyno has an advantage. I'd say go spend the $60-70 and do a dyno pull.
#23
New Member
iTrader: (10)
Join Date: May 2005
Location: Jersey, New
Posts: 7,146
Likes: 0
Received 0 Likes
on
0 Posts
I guess you don't understand how a dyno works. It uses electrical signals to measure the acceleration of the drum. Its not average acceleration like you are measuring, but instantaneous acceleration. It knows the mass and radius of the massive drum, finds the amount of torque applied to the drum, and then finds the HP in correlation to RPM. HP is a function of torque and RPM, so you need to know both. If you have 16 data points/sec, and a dyno run takes 10 seconds to accelerate from 2000-7000 rpm, then youll have 160 data points, at each of which you will need to find acceleration (320 calculations), and at each of those points, find the torque (320 more calculations), and then at each of thosepoints, find the HP (320 more calculation). Then graph the data, and pick the highest data point. Thats ur peak HP. Its not hard, and I dont know why you keep asking this stupid question
#24
New Member
iTrader: (2)
"Instantaneous" acceleration is just a fancy term for "acceleration over a very short period of time." The shorter the time period, the more accurate the calculation.
So, it should be possible to compute HP from acceleration, weight, and RPM. Hey, my GTechPro calculates it, even though their numbers will be a bit lower than a stock dyno because of the addition of air resistance (drag).
So, it should be possible to compute HP from acceleration, weight, and RPM. Hey, my GTechPro calculates it, even though their numbers will be a bit lower than a stock dyno because of the addition of air resistance (drag).
#25
Registered User
Thread Starter
iTrader: (4)
Join Date: Jun 2006
Location: Dearborn, MI
Posts: 306
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by plumpzz
I guess you don't understand how a dyno works. It uses electrical signals to measure the acceleration of the drum. Its not average acceleration like you are measuring, but instantaneous acceleration. It knows the mass and radius of the massive drum, finds the amount of torque applied to the drum, and then finds the HP in correlation to RPM. HP is a function of torque and RPM, so you need to know both. If you have 16 data points/sec, and a dyno run takes 10 seconds to accelerate from 2000-7000 rpm, then youll have 160 data points, at each of which you will need to find acceleration (320 calculations), and at each of those points, find the torque (320 more calculations), and then at each of thosepoints, find the HP (320 more calculation). Then graph the data, and pick the highest data point. Thats ur peak HP. Its not hard, and I dont know why you keep asking this stupid question
For everyone else: I'm positive that a HP vs RPM curve can be calculated from the data included in the data logs. I'm looking for someone with enough knowledge of kinematics equations that can help figure out which formulas can be used to calculate torque at each time interval in the log.
#26
New Member
iTrader: (10)
Join Date: May 2005
Location: Jersey, New
Posts: 7,146
Likes: 0
Received 0 Likes
on
0 Posts
You're right, I dont. Its just the thing every student in every science major covers 2 weeks into physics 1. You have the formulas fool, find data points, and calculate it yourself.
And wtf is a background in kinematics? Ive taken quite a few semesters of physics, and I don't know what that is. Kinematics is just that, 2d motion. So stop using big words to make urself feel smarter. What you have to do is find instantaneous acceleration at each point, then find the torque on the wheels from the 1:1 ratio in the transmission and then find HP after that, and find the maximum HP for the entire RPM band. Ive said that twice, its how its done, wtf is wrong with you?
And wtf is a background in kinematics? Ive taken quite a few semesters of physics, and I don't know what that is. Kinematics is just that, 2d motion. So stop using big words to make urself feel smarter. What you have to do is find instantaneous acceleration at each point, then find the torque on the wheels from the 1:1 ratio in the transmission and then find HP after that, and find the maximum HP for the entire RPM band. Ive said that twice, its how its done, wtf is wrong with you?
Last edited by plumpzz; 04-15-2007 at 06:47 PM.
#27
New Member
iTrader: (10)
Join Date: May 2005
Location: Jersey, New
Posts: 7,146
Likes: 0
Received 0 Likes
on
0 Posts
119 hp to the ground at 130mph. Happy? Notice that number means nothing. Its not how much HP the engine makes, or how much HP actually gets to the wheels. its the net work done by the car including friction, fluid and solid.
#28
Registered User
Thread Starter
iTrader: (4)
Join Date: Jun 2006
Location: Dearborn, MI
Posts: 306
Likes: 0
Received 0 Likes
on
0 Posts
You obviously lack reading comprehension skills too. That's great that you fumbled your way through two semesters of high school level physics classes without learning basic terminology, but I'm looking for answers from people who are a bit more educated.
Do you even have calculations for how you came up with your number? If it's as simple as you make it sound, post your spreadsheet with the calculation columns, and your graphs of Torque vs RPM and HP vs RPM.
Maybe you should just stop posting and save yourself further embarrassment. You obviously don't understand the question that is being asked.
Do you even have calculations for how you came up with your number? If it's as simple as you make it sound, post your spreadsheet with the calculation columns, and your graphs of Torque vs RPM and HP vs RPM.
Maybe you should just stop posting and save yourself further embarrassment. You obviously don't understand the question that is being asked.
#30
Registered User
iTrader: (4)
Join Date: Sep 2005
Location: SoFla
Posts: 437
Likes: 0
Received 0 Likes
on
0 Posts
HP = 745.69987158227022 (Watt)
Watt = (kg x m^2) / s^3
kg = 2.20462262 pounds
m = 3.2808399 feet
HP = (745.69987158227022)(lbs)(ft^2) / (s^3)(2.20462262)(3.2808399^2)
HP = (31.4238746)(lbs)(ft^2) / (s^3)
Now, here are your problems:
1) you do not know precise enough mass of vehical, fluids, driver, dirt on car, etc.
2) For best accuracy, ft => 0. Makes accurate timing difficult lol.
3) you have to do a number of points to try and flesh out a graph
4) it does not account for various losses like air resistance, drivetrain loss, traction issues, etc
but you ought to be able to get somewhere near what a mustang dyno would read. I think. Only one way to test it, and that requires you to hit the dyno anyways lol.
Watt = (kg x m^2) / s^3
kg = 2.20462262 pounds
m = 3.2808399 feet
HP = (745.69987158227022)(lbs)(ft^2) / (s^3)(2.20462262)(3.2808399^2)
HP = (31.4238746)(lbs)(ft^2) / (s^3)
Now, here are your problems:
1) you do not know precise enough mass of vehical, fluids, driver, dirt on car, etc.
2) For best accuracy, ft => 0. Makes accurate timing difficult lol.
3) you have to do a number of points to try and flesh out a graph
4) it does not account for various losses like air resistance, drivetrain loss, traction issues, etc
but you ought to be able to get somewhere near what a mustang dyno would read. I think. Only one way to test it, and that requires you to hit the dyno anyways lol.
#31
Registered User
iTrader: (4)
Join Date: Sep 2005
Location: SoFla
Posts: 437
Likes: 0
Received 0 Likes
on
0 Posts
actually, here may be a better way now that I think about it:
F = m x a
Power = (F x d) / t
P = (m x a x d) / t
P = m x a x v
Power = mass x acceleration x velocity
would be a good bit easier based on the data you have now that I think of it. I'll leave all the unit conversions to you though.
F = m x a
Power = (F x d) / t
P = (m x a x d) / t
P = m x a x v
Power = mass x acceleration x velocity
would be a good bit easier based on the data you have now that I think of it. I'll leave all the unit conversions to you though.
#33
Registered User
That's what I've been sayin'....
Energy Method:
Given - velocity, mass, time
K = 0.5*m*v^2
P = ΔK / Δt = (Kf - Ki) / (tf - ti) (Watts)
HP = P(Watts) / 745.6999
Work Method:
Given - velocity, mass, time
a = Δv / Δt = (vf - vi) / (tf - ti)
F = m*a
d = (vi*Δt) + 0.5*Δa*Δt
Work = F*d
P = Work / Δt (Watts)
HP = P (Watts) / 745.6999
Newton’s Law Method:
Given – velocity, mass, time, gravitational constant
a = Δv / Δt = (vf - vi) / (tf - ti)
F = m*a
a = F / m = (550*g*HP) / (v*m ) (ft/sec^2)
HP = (a*v*m) / (550*g)
But none of these methods give us a good #?????????????????
I'm wasted right now!
Late,
Nick
Energy Method:
Given - velocity, mass, time
K = 0.5*m*v^2
P = ΔK / Δt = (Kf - Ki) / (tf - ti) (Watts)
HP = P(Watts) / 745.6999
Work Method:
Given - velocity, mass, time
a = Δv / Δt = (vf - vi) / (tf - ti)
F = m*a
d = (vi*Δt) + 0.5*Δa*Δt
Work = F*d
P = Work / Δt (Watts)
HP = P (Watts) / 745.6999
Newton’s Law Method:
Given – velocity, mass, time, gravitational constant
a = Δv / Δt = (vf - vi) / (tf - ti)
F = m*a
a = F / m = (550*g*HP) / (v*m ) (ft/sec^2)
HP = (a*v*m) / (550*g)
But none of these methods give us a good #?????????????????
I'm wasted right now!
Late,
Nick
#35
Registered User
It's funny...these guys seem to be able to do it???
http://www.bahntech.com/products/Auto_Dyno/default.asp
h8bumps, let's write some software and sell it for buko $$$!
Late,
Nick
http://www.bahntech.com/products/Auto_Dyno/default.asp
h8bumps, let's write some software and sell it for buko $$$!
Late,
Nick
#36
Registered User
iTrader: (4)
Join Date: Sep 2005
Location: SoFla
Posts: 437
Likes: 0
Received 0 Likes
on
0 Posts
It requires you to know the same stuff I pointed out. Weight, drivetrain loss %, and coefficients of friction. Not real sure where you would find good #s unless you have run your Z on both an engine and mustang dyno, plus had time in a wind tunnel.
#38
New Member
iTrader: (2)
Originally Posted by sabrefanpc
It requires you to know the same stuff I pointed out. Weight, drivetrain loss %, and coefficients of friction. Not real sure where you would find good #s unless you have run your Z on both an engine and mustang dyno, plus had time in a wind tunnel.
You can get a good approximation of what you'd get on a static dyno, but then again the question becomes "which static dyno?" They all read differently, even on the same car on the same day with the same setup, so it's all an approximation anyway.
Besides, we're probably talking about RW numbers, so you don't need drivetrain loss %, and coefficient of friction is also a "red herring" -- especially since there is wheel slippage on any non-direct-connect dyno. An exact vehicle weight is important in the calculation, but being off by a few pounds (or even a few tens of pounds) isn't probably going to throw off the numbers THAT much.
#39
New Member
iTrader: (10)
Join Date: May 2005
Location: Jersey, New
Posts: 7,146
Likes: 0
Received 0 Likes
on
0 Posts
^ +1 , almost. Since you know the change in speed over the RPM, drive train loss, and friction is already factored into the time. The only problems is that I cant do enough calculations without setting up a program in excel to do it for me. One would have to do every calculation and graph it, and find the max, just like the dyno does. Im doing the psuedo math that a dyno does to get that number.
#40
Registered User
iTrader: (4)
Join Date: Sep 2005
Location: SoFla
Posts: 437
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by PDX_Racer
Where's the B.S. flag?
You can get a good approximation of what you'd get on a static dyno, but then again the question becomes "which static dyno?" They all read differently, even on the same car on the same day with the same setup, so it's all an approximation anyway.
Besides, we're probably talking about RW numbers, so you don't need drivetrain loss %, and coefficient of friction is also a "red herring" -- especially since there is wheel slippage on any non-direct-connect dyno. An exact vehicle weight is important in the calculation, but being off by a few pounds (or even a few tens of pounds) isn't probably going to throw off the numbers THAT much.
You can get a good approximation of what you'd get on a static dyno, but then again the question becomes "which static dyno?" They all read differently, even on the same car on the same day with the same setup, so it's all an approximation anyway.
Besides, we're probably talking about RW numbers, so you don't need drivetrain loss %, and coefficient of friction is also a "red herring" -- especially since there is wheel slippage on any non-direct-connect dyno. An exact vehicle weight is important in the calculation, but being off by a few pounds (or even a few tens of pounds) isn't probably going to throw off the numbers THAT much.
http://www.bahntech.com/products/Aut...er/default.asp
Please, no BS flag for myself, as I am simply reading and responding, not asserting anything without considering, like how "a few tens of pounds" will not make much of a difference ( 30 lbs in a 3000lb car would make about 1% difference in your power numbers, small to some I guess but still something to be considered).