calculating whp from a log - speed, time, and physics

[We're on Boost]

HX40/35+E85= So I got the turbo back from Justin. Thanks man.. Here is a log gm 3 bar doesnt read higher than 30psi the boost gauge is showing 32-33 psi.. The clutch starts to slip any more boost than 35psi plus Im out of injector aswell. Link loging 60lbs not bad for the old diesel turbo.. spining the tires rolling into 2nd and the rear end come out is fun!!!

I thought it would be a fun thing to estimate the max whp made by dsmmatt's car during this pull. Wanted to see what kind of troubles there would be with this method, and see how it comes out relative to the airflow rule of thumb.
First off, Pull has to be on a fairly level road, not much wind, not too bumpy a road surface.
Need the car speeds at the start and end of a short interval of time in the peak power part of the pull. From that you can get the average acceleration during that time interval, from that the force necessary to produce that acceleration, then with force and speed you have horsepower. Easy hey?
So what we'll wind up with is actually the average whp during that short interval of time.
Log shows airflow is consistently near maximum from 14.4 sec to 15.33 sec - anyway that's the interval I picked out. Airflow ranges about 420 to 450 gm/sec through most of this range (~ 60 lbs/sec). Could have picked a shorter interval but the shorter the interval, the more sensitive the calculation is to little errors in the logged speed.
Ha Ha, the logged speed in these cars is a JOKE. It's all over the place, up and down. So I used RPM to correct the speed. 6466rpm at start of interval, 7239 at the end. That's an rpm delta of 11.95%. Get the same delta% on mph, you have 89.3mph to 100mph for the speeds. Hopefully the logged rpm values are accurate. If not this whole thing is in the toilet!
Ok, our speed change during the interval is 10.7mph. That is 15.693 ft/sec.
Our time interval is 0.93 sec. So our speed went up 15.693 ft/sec in 0.93 seconds, that is the same as speeding up (calculator) 16.874 ft/sec in 1 sec. That is our average acceleration during the interval ~ 16.874 feet per second per second. No I am not repeating myself .
Assume 3400 lbs is the weight of the car with driver. If we actually knew what the car weighed it would be better!
Now we're ready for F=MA Yahoo! the best equation ever!
F (force in pounds, pushing the car forward)
M (mass in slugs - yeah, slugs, which is pounds divided by 32.174) so this is 3400/32.174 = 105.59 slugs
A which we already figured out is 16.874 ft/sec-sec.

So, F = 105.59 times 16.874 = 1783 pounds force

See how much fun this is?

going to make some coffee ................

 

[TunaTalon]

Your mass was 3400 pounds.

Your acceleration rate was 16.874 feet per second per second.

Force in Gs = 16.874/32 = .527Gs

Force = .527 * 3400lbs = 1791.8 lbs (checks pretty close to OP)

Speed was 95 MPH(average) = 139 Ft/sec

Power = Force * Speed

Power = 1791.8 Lbs * 139 Ft/sec = 249060 Ft Lbs/sec

One horsepower = 550 foot-pounds per second.

249060 Ft Lbs/sec /550 Ft Lbs/sec = 452.8 Horsepower.

Hmmmm over 450 WHP; that must have been quite a ride.

 

[1gDSM4g63]

Yeah than seem quite low boosting 32 psi on a HX40 but then again the car needs to shave some weight down. 3400 lbs is a lot.

 

[twdorris]

The ECMLink V3 application has an integrated torque and HP function that work extremely well if you configure them for your gear ratios properly. We've calibrated them several times with real g sensors on flat stretches of road and we've added support inside the ECU to calculate this as accurately as possible using data from every single firing event.

So just add Torque and HP to your log and then tweak the settings a bit. I keep meaning to put up a "how to" on that, but it's really not too hard. Just play around with the smoothing factors a bit to get a plot without too much jaggedness to it and I think you'll be surprised at how well it does.

EDIT: You'll use the RPM/speed display item for the gear ratios. Just select each gear one at a time in the datalog and then do a "selection->average" to get an average RPM/speed value to use for that gear. You'll then put those numbers into the Torque (or HP...they both use the same "G" function internally) properties.

Also, the latest application (3.20.52) includes an intermediate "G" calculation you can display too.

Ha Ha, the logged speed in these cars is a JOKE. It's all over the place, up and down. So I used RPM to correct the speed.

FWIW, that's exactly what the "G" routine inside ECMLink does too. That's why you need to give it a gear ratio...so we can calculate a reasonable speed from a smoothed RPM signal (exponential averaging).

Thomas Dorris

 

[We're on Boost]

Ok now to finish this thing up!

average speed (middle of the interval) = 95 mph = 139.3 ft/sec

hp is basically force times speed - we're going to get units for power of pounds-feet per second like this:

1783 lbs times 139.3 ft/sec = 248,372 lbs-ft/sec

with a little help from the "convert almost anything to almost anything else" web page, this hugeongous number comes out to 451 hp at the wheels.

Now we need to either make a wild assed guess at how much hp goes into aerodynamic drag at 95 mph, or look up what other people have come up with for the aero drag of various cars. My wild assed guess would have been 80 hp. But after looking around for awhile, I like what I see on this page:

Aerodynamic Drag - Craig's Website at Backfire.ca

On this page, about 3/4 of the way down, Craig calculates the approximate hp drag of his dad's 1987 Supra. 50 hp at 100 mph. Cool! I like his number better than my guess, but while I think my guess is high, it could be that Craig's number is a little low. I'm back to wild assed guessing I guess, but I'm going to say the real number is 60 whp for aero drag at 95 mph. If anybody has better info on the aero drag of our cars I would appreciate it. Cd and A would be nice numbers to have.

451 + 60 = 511 whp is my estimate

If the number sounds a little low to people familiar with this car, well, keep in mind it is the average whp from 6466 to 7239 rpm, not a peak number from the top of some little bump in the curve like you would see printed on a typical dyno sheet.

Thanks for the comments you guys!

Go Saints and Colts! I don't even care who wins - just go already!

 

[jrohner]

There's really good excel spreadsheet for this on EvolutionM.net. I got mine to be nearly identical to my dynojet run.

 

[the_mork]

Virtual Dyno Room-Dyno Simulator - evolutionm.net

 

[dsmmatt28]

Well here is the scope on the car spare gone; all ac gone; front rear bumper supports gone. Theres a few other things that I have saved weight on last time I had the car weight was 3183+188. So 3371 thats pretty close to 3400. I can say the car was pulling extremely hard. I can say I have a fast dsm. I came from a bws259ETT and this Holset pulls harder. I have made previous 483 that was on 30psi stock intake mani and dsm chip. Now im on JMF simm and ecmlink and hx40. This turbo hits way harder and car just squats when the boost hits. The car is much faster than when I made the 483 hp. You know that feeling you get inside you stomach when you go down the first hill on a roller coaster ride ?? Imagine that all the time when the turbo is working.. I have a 4bolt that needs to go in and the fix the tranny. Track time next month. I think the car is a sub 11sec maybe a high ten if I can launch good and use the NLTF. Will see soon.

 

[TunaTalon]

On this page, about 3/4 of the way down, Craig calculates the approximate hp drag of his dad's 1987 Supra. 50 hp at 100 mph. Cool! I like his number better than my guess, but while I think my guess is high, it could be that Craig's number is a little low. I'm back to wild assed guessing I guess, but I'm going to say the real number is 60 whp for aero drag at 95 mph. If anybody has better info on the aero drag of our cars I would appreciate it. Cd and A would be nice numbers to have.

451 + 60 = 511 whp is my estimate

If only the 60 horsepower from Craig’s site were required to overcome drag at 95 MPH then the top specified speed of 140MPH would require only:
140/95 = 1.47 times faster than reference
1.47 * 1.47 = 2.16 times the power at 95MPH
2.16 * 60 = 129.6 HP for aero drag at 140MPH

If the Talon needed only 130HP for aero drag at 140MPH it would not have been aero limited below red line. Let’s go back 19 years in my memory. Caution some fading may have occurred.

As I remember my 1991 Talon’s specs it was aero limited at 140 MPH which corresponded closely to the peak horsepower of 195 at 6000 RPM.

If we ASSUME that 90% of the power was used to overcome the drag then we have 175.5 HP at 140MPH. Since aero drag increases as the square of the speed we can calculate horsepower to overcome drag at 95MPH as:

95/140 = .678 of reference speed
.678 ^ 2 = .459
.459 * 175.5 = 80.5 HP to overcome drag at 95MPH.

This is not exact but should be close.

This would have your wheel HP (using my calculations) as 452.8 + 80.5 for a grand total of 533. This is pretty close to your estimate.

I like this technique because it requires only the logging of two speeds a few seconds apart and the cars weight. No complications from gearing or torque calculations. If the pull is made in a lower gear then aero drag could be ignored while staying within legal speeds.

The Talon’s Cd is on a brochure somewhere in the house but the area is a mystery to me. I remember spending a couple of hours last year looking for the frontal area of a DSM but came up empty.

 

[We're on Boost]

Yeah those are some good thoughts.
I keep thinking that Craig's drag hp numbers are a little low too. But I don't think they are low by very much. For one thing, while drag Force is proportional to V squared, drag HP is proportional to V cubed! Here is a link for that which is a lot more clear about it than Craig:

Drag (physics) - Wikipedia, the free encyclopedia

So when you get to calculating the drag HP of a car at say 150 mph, it is quite a bit.
I have no idea what the real top speed is of a stock 87 supra, or for that matter of my own Talon. My best top speed experience was with low powered cars that couldn't go very fast! Like my 1970 Peugeot 504. A long time ago when it was easy to find long stretches of road in places like Montana, Oregon, some of the other Western states, where you could drive flat out for minutes at a time without fear of going to jail. So this Peugeot - flywheel hp in the mid 90's, top speed about 98 mph or so. I figured a little over 15% in drive train losses, so maybe 80 hp at the wheels, all of it going into aero drag at top speed. That's where my wild guess came from.
Anyway, I like the method too. I think the weakness of it is probably that it would give you a slightly different answer if you pick a slightly different interval start or end point. This would depend a lot on if the logged rpm values are jittery at all, like from bumps in the road or whatever. So I like Tom Dorris's idea of smoothing the RPM signal.
I hope more people get into road dyno usage and/or calculating their own hp numbers from the raw data. Then, in addition to everybody arguing about Mustang vs Dynojet vs DynoDynamics - we can also compare the different road dyno techniques!
Nah, seriously, when you do something like this with just the raw data, your calculator, and Isaac Newton, what could be better? Ah yeah, and a cup of great coffee.

 

[TunaTalon]

For one thing, while drag Force is proportional to V squared, drag HP is proportional to V cubed! Here is a link for that which is a lot more clear about it than Craig:

Drag (physics) - Wikipedia, the free encyclopedia

Nah, seriously, when you do something like this with just the raw data, your calculator, and Isaac Newton, what could be better? Ah yeah, and a cup of great coffee.

Beware of Wikipedia.

Where: Power = Force * Speed and force is a square function of speed then Power at any given constant speed is a square function of the speed.

Wikipedia may be confused with the power required to accelerate, not hold speed against drag and momentum.


Power required to accelerate a mass is a square function of speed.
Power required to overcome aerodynamic drag is a square function of speed.

So to overcome drag and momentum the power required to accelerate increases with the cube of the speed.

This is explained better at Velocity & air drag.

If going on high speed we should consider, that the energy needed to accelerate the car quadruples with a doubling of the velocity and also the air drag quadruples with double velocity. Therefore, a car cruising on a highway at 50 mph (80 km/h) may require only 10 horsepower (7 kW) to overcome air drag, but that same car at 100 mph (160 km/h) requires 80 hp (60 kW). Twice the speed requires eight times the power.

Emphasis mine.

Coffee is fine for the daytime breaks but I still prefer California’s best for my evening work. Work??? Who said that?

 

[dsmmatt28]

[/Coffee is fine for the daytime breaks but I still prefer California's best for my evening work. Work??? Who said that?]

I second that..

 

[We're on Boost]

You guys with your California's best . Actually I was thinking - how about a HotforWords video that gives a step by step on how to do this whole thing? Hot chicks doing physics = PASS

You know in that Quantify link they talk about force quite a bit, and they talk about kinetic energy some which is not very helpful to us I don't think. They talk about power almost not at all. Only in the last 3 paragraphs do they talk about power. Where they finally get down to what we're talking about is when they give the example where they say that if 10 hp goes into air drag cruising at 50 mph, then 80 hp goes into air drag cruising at 100 mph. Eight times the hp when you double the speed. Figure that backwards, they are saying hp goes up with the cube of the velocity.
or:
V2/V1 = 2
2³ = 8
8 X 10hp = 80hp

The hp numbers in Craig's chart do the same thing, they go up with the cube of the speed. Craig never really explained this and never wrote an equation for Power as a function of velocity. But he does have an equation for Vcubed, which when you rearrange it for power would look like this:

hp = V³ CdA / 151130

In the Wikipedia article the guy comes right out with it in the section called POWER. He writes the equation:

Pdrag = ½ ρ V³ A Cd

You know what, writing an equation that looks right in here is really hard. If it wasn't so late I'd have some more coffee.