I'll let Professor Jan and Big Howe answer your question Jason, about time they took a turn.

 

Our back straight at Bedford Autodrome is 1.2 slow exit 40-50, 128mph with stock cyl head, might give it a shot of nitrous next ttime to keep those GT3's at bay

If you got avideo pleas post a link because I do enjoy them.

 

Different pulley, different exhaust, who knows what else.

 

I don't know Paul, you might ask Jan.
He told me why and I don't remember.

 

Correct on the first 2
My car weighs <2500 lbs though

 

Kewl, is that with you in it ?

 

Ok, Jan how does a difference of 22.5 ft lb at 7600 on MOTORN (157.5 ft lb)and Big Howe (180 ft lb) make similar horsepower figures 253 and 259 respectivley

https://www.northamericanmotoring.co...9&d=1210642747

https://www.northamericanmotoring.co...9&d=1225460810

rpm x torque x 5252 = HP

Or is this old as the hills formula wrong as well

 

I wish. Nope it is a street car as delivered. ( AC, carpet, radio, etc)
big howe is at 2820

 

Why did you have your car weighed Big Howe?

 

Because there are some certified scales next to one of my customers plant, and being a cabrio I was curious as the weight compared to the coupes. I was surprised it was as low as it was after hearing what some of the coupes weigh. 2820 with a 1/4 tank, no driver.

 

FYI, I came across this article some time ago that has a pretty good description of the differences between hub and wheel dyno measurements. As you would expect, the overriding difference appears to be due to the impact of the wheel's Moment of Inertia on the torque calculations:

http://home.earthlink.net/~spchurch/churchautomotivetesting/id12.html

The difference between the two approaches increases with higher horsepower cars. For the hp/torque levels produced by Minis it seems like (very) roughly a 3-5% difference.

 

So how much do you weigh and what mph did you cross the 1/4 mile line at?

 

???

 

How much do you personally weigh so we can add that to the car's weight and if you have taken the car down the strip, what mph did the car cross the line at?

 

Right, and as per the last few pages of this thread, my car has never been down the strip. Hence the question from Paul, why has my car been weighed?

 

Thanks for that link, it explains quite a bit.

Jason

 

Still waiting for Paul to explain how your car can have 250bhp yet all the other dynapak cars are barely over the whp calculations

 

I've explained it, its in post 232, I look forward to your answer.

 

My turn to ask a few questions (not directed at any one person):
Back in post 91 I asked the question about wheel hp vs hub HP....obviously there are some feelings that they do differ. What sort of % are we talking? We assume something like 15% loss for WHP, what do we assume for Hub HP? My wild guess is maybe 1 or 2% less loss, seems like most of the friction losses are in the bearings/ and gears, with no a whole lot of loss in the tires themselves.

The best way I can describe it Jason is your engine will make xxx amount of horsepower at teh flywheel, a hub dyno will show the closest to a flywheel as it only has the gearbox driveshaft and wheel bearings as drag, I reckon as a figure 7.5% on Churches, but if your dynapack has been re-caliberated to read like normal rolling roads then yes it could be 14-18%


Additionally, I'm not exactly sure what the SAE correction factor is for my dyno numbers, but I'm assuming this corrects the numbers to be at some standard temperature and pressure. Now, what happens when I make my 1/4 mile pass under better conditions? (mid 40's F and with a intercooler that was chilled on ice prior to the run). What are these standard conditions that are corrected for?

Obviously in the USA you have different elevation and range of temps depending what state you are in. Your best way is to dyno it then go to the drags weigh it and run it to get a relation between the dyno and strip. But remember to take IAT readings on the dyno, going into stage and crossing the line, again this will give you a relationship between dyno and track. Same as when you auto x record what you see and try to recreate it.


Obviously, if I would've dyno'ed with these cooler conditions I could've put down more power (ignoring correction factors...). Temp seems to be such a HUGE variable with my car, the same night I ran my 13.8, my MPH ranged from 98.x to 100.3 depending on how much time I had to cool things down before I ran. If I would've ran the 1/4 mile the same afternoon as my tune (mid to upper 80s F I think), there is no way I would've pulled off a 13.8x.

The dyno should factor it to an SAE number. Though the colder it is the lower the IAT which will effect the readings.
Again your IAT will effect your mph depending whether you going over 50 deg C up the strip


Lucky man running a 100.3

 

Hello all, this is Shawn from Church Automotive Testing. It came to my attention that there was a lot of discussion on hub vs. wheel hp here and a lot of testing from our equipment was being used as a reference point. I thought I'd come over and offer a few tidbits of info to gnaw on and perhaps answer a couple questions if necessary.

First off, I've noticed that someone has already found our rough inertia calculations from the website. While it isn't useful for everyone, if you really want to begin understanding differences between the dyno types, I recommend reading that short piece. For those that don't really feel like getting into the math, the one thing you might want to take away from it is that while power losses on inertial roller dynos (i.e. Dynojet) vary relatively consistently with power (IOW, are best expressed as a percentage loss of estimated flywheel hp), hub type load dyno losses are best expressed as a fixed loss.

Briefly, we find this holds true for several reasons. First of all, the acceleration time on a Dynapack is set at a fixed rate, and therefore the inertial losses due to flywheel mass, brake rotor mass, transmission mass, etc. should stay largely fixed as well - regardless of whether you are making 100 hp or 900 hp. On an inertial roller dyno, the acceleration time will decrease (increasing inertial losses) as power levels go up. Second, the only other major non-inertial sources of losses when dyno testing a vehicle on a Dynapack will be from fluid drag, bearing losses and gear to gear friction. Fluid drag should not change noticeably for a fixed setup (same oil level, pump, etc.) because the engine speed and acceleration rate remain unchanged. Bearing losses (outside the engine) are also relatively insensitive to power levels since they are not subjected to very large axial loads in most cases. Gear losses, especially those related to helical cut gears, will vary somewhat with power, but we are talking about losses on the order of 1-3%. Thus, a truly accurate expression for Dynapack losses might by [X hp + (Y% * Flywheel hp)], but we find that approximating with a fixed value is simpler for most people, and if anything underestimates maximum power and gains as power levels go up (IOW, it underestimates the benefit of any tuning we do - a safe bet if you must choose to err one way or the other).

Next, despite that long explanation of losses, I would also point out that we feel trying to calculate back to a precise flywheel power level is somewhat pointless. It is next to impossible to accurately measure drivetrain losses on a chassis dyno, even one that eliminates so many variables like a Dynapack. It is even more pointless on a roller dyno. There are just too many variables, and coast down testing does not accurately simulate the losses present in a normally loaded driveline (instead of one being driven by the inertia of the rollers). Nonetheless, people like to ask about our guesses as to flywheel hp, so we give them a brief disclaimer and let fly anyways.

Before I move on to a discussion of equating dyno power readings with dragstrip results (which is really the key topic in this thread), I would also suggest that anyone who owns a dyno or regularly uses one purchase this SAE paper post-haste. And if you are at all interested in how you are measuring your modification results on a dyno, you may want to spend the $12 too. Just go to www.sae.org and search the library for:

#2002-01-0887
Listening to the Voice of the Customer: Inertia Dyno Horsepower Versus OEM Rated Net Horsepower

This is an outstanding work by the researchers at the Southwest Research Institute (SWRI) and includes detailed analyses of the various drivetrain losses of a vehicle, including transmission losses, tire/road losses, losses induced by strap down tension, etc. Feel free to ask questions and I'll pull from the piece if the data is in there, but definitely buy the paper if this topic interests you at all.

A few interesting tidbits from the paper that some of you might find interesting:

1) Variations in strap down tension on a roller dyno can increase losses by 5%-7% or more versus an optimal (meaning minimum force necessary to maintain stability and traction) strap tension setup.

2) It takes upwards of 20 hp just to overcome rolling resistance at higher wheel speeds on a roller dyno - and that is assuming a rolling resistance model of a tire on a normal road. On a roller dyno where there is additional loading (through the tie down straps), this loss will go up. Furthermore, this loss varies with tire speed, so the losses at 100 mph will be more than triple those at 50 mph.

There is lots more interesing information from the paper if you are so inclined.

Anyways, that brings us onto the estimation of 1/4 mile results from dyno curves, or estimating power from 1/4 mile results. Let me first say that if you control (or at least understand) all the variables and put together a detailed enough simulation, you can get a pretty good approximation going either way on such a calculation. As an trained engineer I firmly believe this to be true.

However, as a trained engineer who has had to work in the realities of the real world, I must also say that the attempts made by most people to mathematically model such a result are sorely lacking in resolution and accounting for enough significant variables.

From what I can tell, most folks use simulations that attempt to calculate a peak power number at the flywheel by using empirical results to establish a best fit curve between rated flywheel horsepower and observed drag strip results. Those attempting to fit this to dynamometer results are probably using an adaption of this model. If anyone is going more in depth, please by all means share. The main problem with such calculations, as I mentioned earlier, is that they are usually lacking in taking into account a number of variables. Among the most critical are the shape of the power and torque curve, tire losses, aero losses, atmospheric conditions, available traction, etc. Furthermore, because they are using an empirical fit method to develop the model, said model is often lacking in proper physical modelling of the various forces at play - for example, a proper tire dynamics model which will be able to take certain raw tire variables (rubber compound, sidewall construction, air pressure, etc.) and accurately project coefficient of friction, sidewall deflection losses, deformation, etc. (interestingly, this is the same problem that climate models run into - too much empirical fitting and not enough understanding of actual physical processes)

Since this is a Mini forum, let me point out that the first issue to pop up is that of traction. Let's face it, FWD is not exactly optimal for transferring power to the pavement under most circumstances. For a relatively low powered car, that isn't too big of a deal, because after first gear traction limitations are gone. But for a car with lots more power and traction issues into 2nd or 3rd gear, you are limited in how much power can be put to the ground. The bigger disparity between available power and the power you can put to the ground (traction), the harder it becomes to model quarter mile results vs. power numbers - even trap speed. To disprove the trap speed to hp calculation model only takes one example. In testing a couple of years ago Motor Trend was testing some very high hp hardware. In particular a turbocharged Corvette which ran 11.9@134 on street tires, but 11.1@151 mph on drag radials. A similarly powered Viper went from 138 mph to 152 mph (and even faster traps on slicks). All with the same power. Thus, traction plays a role in trap speed, increasingly so as power builds.

A second issue that I find particularly relevant as a dyno tuner is the attempt to calculate a peak power number when what the quarter mile and trap speed really measure is average horsepower. So what you're really doing is integrating the average horsepower applied to the track over the course of a run, while attempting to figure out how much power had to be applied to overcome aerodynamic drag, how much time was spent not accelerating (shifts), and how large your inertial losses were while hoping that there were minimal amounts of time where you were not traction limited in applying power. To calculate a peak power number from that is difficult without a very complex simulation. For example, a car with high average power (say a turbocharged car with 200 whp and 200 lbs-ft of torque) and the same power to weight ratio as a 200 whp naturally aspirated car will usually return a higher trap speed because the average power is higher.

Finally, I don't think enough people take atmospheric conditions into account in enough detail to matter. Density altitude (or DA) is a measurement top drag racers use to predict their performance on any given day on any given track. They typically know what their car will run with good traction at a range of DAs, so they can make a rough estimate of how changes in DA will affect their times.

A list of how various density altitudes affect times is here (and in many other places:

http://www.smokemup.com/auto_math/nh...on_factors.php

A means of calculating your density altitude can be found here (just enter 29.9 for altimeter reading to assume normal barometric pressure if you don't know it)

http://wahiduddin.net/calc/calc_da.htm

As a rough example, the track where most people get their times in Southern California is Fontana. Fontana is approximately 1200 ft of elevation. As I write this, the weather conditions are 64F, 29.9" and a dewpoint of 47F. This results in an effective DA of 1974 ft. This gives a trap speed factor (according to the NHRA) of 1.023. IOW, you would run 2.3% faster at sea level than at Fontana today.

OTOH, today in Englishtown, NJ, known as one of the fastest tracks in the nation at about 100ft above sea level, the conditions are 42F, 30.11, and 22F dewpoint. This results in a DA of -1184 FT. That's right, an effective altitude over 1000 ft below sea level. This would result in a trap speed 1-2% faster than at sea level. Or, for another comparison, if you ran the same car today at Fontana and Englishtown, the trap speed would vary by 4+%. IOW, a car that would trap a nominal 100 mph at sea level would trap 102 mph at Englishtown and 98 mph at Fontana. But the car would make the same power on the dyno on both days. And good luck if you have times from a track like the old Palmdale dragstrip in summer. At 2600 ft of elevation with temps of 100+F, you would see density altitudes in excess of 6000 ft - enough to slow your 100 mph trap speed mini down by 8+ mph.

You can attempt to correct your times by looking back at the day's weather conditions after the fact, but the pros actually have their own on site mini-weather stations to get accurate data for setups and dial ins. Without such accurate data, and knowing how variable weather can be even over a short distance, your trap speed at the track becomes less valuable in predicting power without really good data acquisition.

Anyways, I know that was a mega long post, but much of it is culled from other presentations and such that I've made. I hope it has shed more light on this interesting topic and that folks continue to refine their testing methodology and keep going faster. If anyone has any questions that I'm qualified to answer I'd be more than happy to help out.

Rgds,

Shawn Church

 

Man.... all I can say is thank you !

That is some awesome info Shawn.

Longboard

 

Welcome on NAM Shawn, If there is award for best first post ever I think you just won it.

Seeing as you are a Dynapack owner and had one since they were first introduced in the USA, could you explain post 232 and what or how those figures could of been produced. (not yours obviously)

 

Hi Paul. Your question is a tough one because all the important information is impossible to see in the picture with the apparently mismatched data. That shot is from what Dynapack calls comparison mode, which I really don't like to use (I'll explain why) unless the customer is asking for some weird data comparisons such as flywheel torque vs. time instead of the usual rpm.

Comparison mode allows the operator to select any two variables for the Y-axis and time or rpm for the X-axis. It is possible to select uncorrected power on one axis and corrected on the other. My guess is that one of the values is corrected and one is uncorrected. That's why I always prefer the dual graph screen you see in the other photo. It is very clear on what you're seeing, as well as showing the TCF (transmission correction factor - which should be 1.0 for comparison), gear ratios, etc.

SC

 

Sorry for the question Shawn as you haven't obviously got all the data to look at, but I understand where you are coming from. Is correcting the raw data by 11.11% from 157.5 lb ft to 175 lb ft normal on a dynapack?

What inputs do you have to feed a dynapack to run a car on the dyno?

Could you explain how the dynapack works and calculates the figures?

Are both your dynos flat out everyday of the week ?

How many Mini's do you think you could dyno in one day using both dyno's in a shootout.

 

To disprove the trap speed to hp calculation model only takes one example. In testing a couple of years ago Motor Trend was testing some very high hp hardware. In particular a turbocharged Corvette which ran 11.9@134 on street tires, but 11.1@151 mph on drag radials. A similarly powered Viper went from 138 mph to 152 mph (and even faster traps on slicks). All with the same power. Thus, traction plays a role in trap speed, increasingly so as power builds.

Unfortunatley I have to disagree with you on this one Shawn We run a series called the Street Elinmator here in the UK, rules are Micket Thomson Sportsman Pro's (yes those dinausour block tread tires) and pump gas. All the cars have to do a 30 mile cruise after qualifying.

At the 6th round at Shakespear County Raceway we qualified with 8.962 @ 164.36 , in the 1st round the car stood up for the first time this year into a wheelie, Jeff had to lift to bring it down but soon as he nailed it again it cough and splattered then eventually found itself and went again, guess what 10.222 @ 165.33 so 1 mph difference or in a % terms 0.59%

The cars best is 8.36 @ 166

http://www.streeteliminator.com/late...ws.html#round6

If you pan down its the yellow one with flames and has a 92 inch wheel base