[California Saab Tuner]

I've spent a lot of time recently going over dyno charts with various friends. There is far far too much variation. 246 whp Evo8's shouldn't be slower than 213 whp Saab Viggens. This isn't right.

I began to notice that dyno numbers on performance cars (like the Evo and STi) are nearly ALWAYS exaggerated. So I thought of two methods to at least get some idea how much of the power the dyno claims you have, is actually making it to the ground.

The reason dynos are always trotted out as "accurate" machines has to do with their repeatability. However this has absolutely NOTHING to do with accuracy. Repeatability is precision, and is not necessarily even close to the putative "true" and accurate value. The methods I'm going to discuss here are not nearly as precise, but if done properly, they are considerably more accurate, because after all, how can you truly be sure you're measuring the power to the ground, if you're not ON the ground?

A g-tech (ignoring the hp calculation feature) will give you reasonably reliable g-force figures if setup properly, but it's still not quite accurate enough. Ironically I found the most reliable method is a stopwatch and video camera.

This will NOT give you peak horsepower on its own, and you will still need to visit a dyno to generate a proper curve and peak. But this CAN give you some measure of a correction factor in case your dyno was either optimistic or pessimistic.

The second method, which I will put in another post is intended for us american folks who have easies access to a drag strip than a dyno house. It's NOT accurate at determining horsepower, but it will tell you which of two cars has MORE power, and which has less, even if it cannot come close to an actual figure.

So here goes the first idea ... (I originally posted this on a WRX forum)

#1 ... this is the method for measuring power that is best used as a simple correction factor to your dyno runs.

First ... this works best if you can find conditions similar to your dyno run. Temperature, Barometric pressure and Humidity if possible.

Second ... get out a video camera and a stopwatch. I know it sounds ghetto, but try to get a good shot of the stopwatch AND the speedometer. Also, be considerate and do this in a SAFE location.

Third ... pick two speeds to accellerate between. Try to find a level spot in your torque band, and for you WRX'ers 3rd gear will be a nice sweet spot. Try between 40-75 mph. Pick out where your peak torque should be. We're not after an accurate peak horsepower measurement, just a good torque measurement. If the torque is 15% off your dyno run, the correction can be applied to your peak horsepower as well. This allows us to measure at a lower speed, but in a high gear, which will give the most accurate reading.

Start the stopwatch and don't worry about starting it and the camera at the same time. The absolute time on the stopwatch isn't important, we're going to subract times to find acceleration. The stopwatch must just be running to give an accurate time index.

Start the video camera and record your car accellerating at full throttle from whatever two speeds you chose.

You've collected your video data; now comes the fun part ... analyzing that data!

Try to find two clear shots in the data at around 60 mph. Sometimes you may not get any clear shots, and you'll have to record another run. If you have two clear shots, record the time on the stopwatch, and the speed in each shot. Try to get the two shots as close as you can, but NOT back to back for an accurate torque reading. For instance get two shots about 5 mph apart.

Here's the icky math part ... be prepared to puke ...

formula for horsepower in US units: horsepower = (mass x speed x accelleration)/550

alternate formula for horsepower in US units: horsepower = (weight x speed x g's)/55o

formula for converting hp to torque: torque = horsepower x 5252 / RPM

Approximation for converting speed in MPH to speed in ft/sec: ft/sec = mph x 1.466

So now let's say your Rex went from 55-60 mph in exactly .76 seconds ...

To find the average accelleration in ft/sec we take the difference between the two speeds, multiply it by 1.466 and divide by the time it took.

5 mph x 1.466 / .76 = 9.645 ft/sec/sec

Now let's say you had a driver and a passenger. I'll estimake 3500 lbs as the weight. This however is NOT mass, in the US system mass is measure in Slugs. One slug at sea level weighs 32 lbs ...

3,500 pounds / 32 = 109.4 slugs.

Now to find horsepower, and remember ... this isn't your PEAK horsepower, only one step in finding torque. You'll need to know your average speed as well ... (55 + 60) / 2 = 57.5 mph x 1.466 = 84.30 ft/sec

hp = 109.4 x 84.30 x 9.645 / 550 = 161.72 horsepower ... BUT WAIT

We're still looking for torque ... to find it we need to know RPM. Now you could look at your tach, but a better way is to find out where your redline is, speed wise, and thus, what RPM you'd be running at your earlier calculated average speed of 57.5 mph.

I believe for a Rex your redline in 3rd is 96 mph @ 7,000 RPM. To find the RPM at 57.5 mph ...

7,000 rpm / 96 mph * 57.5 mph = 4,193 rpm @ 57.5 mph

Now to find torque ...

Torque = 162.72 hp x 5252 / 4,193 rpm = 203.8 lb ft to the wheels.

Now we need to account (at least a little bit) for wind resistance. At about 60 mph there is between 6-9% torque lost to wind. My car lost 5% torque, but it's fairly aerodynamic. Let's say the Rex loses a mere 6% to wind.

203.8 lb ft / (1 - .06) = 216.8 lb ft

Let's say your dyno run yeilded 201.3 lb ft to the wheels at the SAME RPM and 200.6 horsepower (peak) ... to find what your peak horses SHOULD be ... assuming you did everything correctly ...

216.8 / 201.3 * 200.6 = 216.0 horsepower

You may find dynos were optimistic, or pessimistic, but please try not to use this as a bragging right. It's possible you did something wrong, or the road wasn't level, this is FOR YOUR INFORMATION ONLY. It is worthwhile to share with others what you learned, but if you got numbers WAY higher than a dyno you probably did something wrong. Dynos tend to be optimistic, I've never seen one that gave significantly lower than expected numbers unless something was wrong with the car. Mine came out just about perfect.

Use with care ...

Dubbya~

 

[California Saab Tuner]

Here is the second method. Remember this is really for us Americans who have easy access to quarter mile strips. Again it is not intended to find actual power. Basically the idea is as follows: Two guys are at the strip, one clearly runs a faster time, but the other guy thinks his car in reality has more power, and that he just cannot get traction. This is a good, and reasonable, way to settle the debate. It is not accurate at determining actual peak horsepower, but it will tell you who averaged more power between the 1/8 and 1/4 mile ... odds are this person also has a higher peak hp, though sometimes if the two cars are close, a higher peak averages lower because of a narrow powerband ...

Generally a trip down the quarter mile gives you at least the following information:

1. Reaction time (this does not affect ET)

2. 60' time (useful when talking about traction and driver skill, but not much else)

3. 1/8 mile time and speed @ 1/8 mile

4. 1/4 mile time and speed @ 1/4 mile (it's actually an average speed over the last few feet)

To find average horsepower all you need to know is average accelleration, average speed, and weight. Out of all the previously mentioned data, the best comparison for horsepower will be the speed + time between the 1/8 and 1/4 mile marks. This is ideal because hopefully traction will not be a problem, and the wind resistance for both cars will be about the same. We won't get any "real world" horsepower figures, but we can figure out pretty well which one has more, and which one has less.

To show you that this can be reasonably accurate even between VERY different cars in VERY different conditions I will compare a friend's lightly tuned Rex timeslip to a timeslip I got in my Saab at high altitude.

Subaru/ Saab

60' 1.907/ 2.43

1/8 (time) 8.879/ 10.162

1/8 (speed) 78.34/ 73.41

1/4 (time) 13.928/ 15.383

1/4 (speed) 96.81/ 96.38

Since accelleration is (difference in speed) / (difference in time) ...

For the scooby we get ... (after converting mph to ft/sec by using 1.466)

(141.9 - 114.85) / (13.928 - 8.879) = 5.357 AVERAGE accelleration

For the Saab we get ... (after converting same as the scooby)

(141.3 - 107.62) / (15.383 - 10.162) = 6.45 AVERAGE accelleration

This isn't where the story ends though. To find horsepower we need average speed as well. This will be (1/4 speed + 1/8 mile speed) / 2

For the scooby we get 128.38 ft/sec (87.575 mph)

For the Saab we get 124.46 ft/sec (84.895 mph)

To find average, uncorrected horsepower between the 1/8 and 1/4 miles we use the afformentioned formula. ( hp = mass x speed x accelleration / 550) Since both cars weigh the about the same. I'll assume with one driver they're both about 3350 lbs. (104.69 slugs)

hp = 104.69 x 128.38 x 5.357 / 550 = 131 AVERAGE and UNCORRECTED horsepower for the Scooby.

I realize that seems low, but remember that's not peak hp, it's average. And it's averaged from 80-100 mph. It's not gonna be nearly as high as peak, and don't forget the wind resistance. Now the Saab ...

hp = 104.69 x 124.46 x 6.45 / 550 = 152.8 AVERAGE and UNCORRECTED horsepower for the Saab

Now you can see why just looking at E.T. and even Trap speed isn't always the best method for comparing horses. Despite being much slower in time, and a little slower in MPH, between the 1/8 and 1/4 mile the Saab picks up much more mph, as a result of a very very wide powerband. (HP to the wheels was 213 @ 4,900 RPM and 251 lb ft @ 3,000 rpm)

This still won't necessarily give you accurate comparison for peak horses. But it is usefull. This will not correct bad dyno figures either, but if you have time slips, you can figure out from this who really has more power, and who is (even more impressive perhaps) a better driver and/or has more traction.

Remember though, this is just another tool. Use with care and caution.

Dubbya~

 

[StanleyB]

Interesting question Adrian, and one I already questioned HERE?
The guys are out today enjoying themselves on a RR meet, so we'll get a chance to compare before and after figures from the various trips.

 

[wrighar]

Originally posted by StanleyB:
[qb]Interesting question Adrian, and one I already questioned HERE?
The guys are out today enjoying themselves on a RR meet, so we'll get a chance to compare before and after figures from the various trips. [/qb][/b]

I have to say that I say Tandinos new (35000m) t16s on the rolls yesterday, and the figures were so close to manufactures spec that it was almost unbelieveable..

173.5 bhp
204ftlb
from memory, all at the correct RPM.

Andrew

 

[Jezzadee]

There was a 9 5 LPT rated at 150 bhp stock that got 151.5 bhp as well.

 

[California Saab Tuner]

But Hirsch's dyno seems to have been a bit tempermental.

This could be useful as a correction to that. Or perhaps as proof of accuracy should that indeed be the case. Either way, I've used it on my car, smack on every time. Same with 2 of my friends. Whereas 3 of my friends who's dyno numbers always seemed off were off when I used the first method here. So far it seems to be fairly consistent. Always worth a shot.

Dubbya~

btw ... if you have an AP22 accellerometer you can use the second formula I listed for finding the hp at the wheels at your peak torque. Then follow the rest of the instructions per the post. You'd just put in the accelleration in g-force into the formula with weight and g's instead of mass and ft/sec/sec.

 

[StanleyB]

From the results and comments so far made I would like to believe that the Emerald RR is very well callibrated. I am looking forward to seeing some more graphs and feedback from the individual owners. But as with all other RR meetings, I sometimes get the impression that not everyone remembers that it is basically a fun day out and a chance to meet new friends and catch up with old ones. Some take the RR results far too personal. Which is a great shame.
Just my two cents though. I might be reading the wrong message between the lines .

 

[jeremyfrost]

Yep, Emerald has a reputation for being one of the most accurate and consistant RR in the UK. These guys are well respected.

 

[Jezzadee]

I certainly wasn't aware of people taking it too seriously. I think a couple of people were disappointed with their results since they were below previous ones, but were pleased that they now had a diagnostic to work out what the problems were. Since we all have such different cars, some stock, some heavily modded, there's no way it could be seen as any kind of competition IMO. Personally I was just hugely relieved to find that mine was running well. I was almost certain they would detect knock, as I've seen it happen to others on a similar setup to mine, and I was expecting around 160bhp. To get 183bhp and no knock means I am very pleased, but not at anyone elses expense. It was good to meet people again, and some new - but I always forget the mild sense of hypothermia I develop at these things Thanks to JezF for organising it all too.

 

[Bussman]

Yes I totally agree its not a competition..
it was a exercise in how to meet new people/check your car for faults,& enjoy the day..
my results were down on what I was expecting but "hey its a 17 year old T8"...only started to mod it, now I have a benchmark to work from..
it was more than just RR results....
people from all over the UK & Europe turned up to a small part of Norfolk,to meet & greet new friends see some cool motors & have a laugh

Paul

 

[California Saab Tuner]

I don't think it was a competition either. I actually posted this link BEFORE I remembered that it was RR day. Kind of Ironic actually.

At any rate I only wanted to post it in the event that RR figures ever seemed questionable. It's a good method just in case. Not meant as competition, just verification for your own informational purposes.

Dubbya~

 

[StanleyB]

There are 3 types of rolling road dyno which I am familiar with.
1. - Clayton type water brake dyno's. 2 rollers in the normal way and an internal viscous type torque converter to measure the applied force to the driven roller.
2. - Electric eddy current dyno's - Hoffman and Sun are the usual makes.
3. - Massive flywheel Bosch system.

All have their good and bad points but IMO they are all capable of reasonably accurate results if calibrated and used properly. I have a beef with systems with small diameter rollers that seem to suffer from tyre slip more often. The Claytons fall into this category.
My BIG bugbear though is with so called transmission losses measured on the overrun. Most dyno firms insist on giving "flywheel" bhp printouts and often the measured wheel figure is never even shown. IMHO it is IMPOSSIBLE to measure trans losses on a rolling road dyno.
What happens is this - you run the car under load until just after peak power then the operator drops the clutch and lets the car freewheel down against the roller which supposedly measures the "negative" power absorbed and treats this as a trans loss.
Now one thing is for certain - trans loss is a percentage of power input to the system. If the car is in neutral there is no power being fed in. Also the gears are now backlashed against the wrong face of the gear teeth as the roller is driving the car not the other way round. Under these circumstances I see no way in which the trans loss shown can bear any relationship to actual losses when the car is under power. Tweaking the electronics to generate big overrun losses or just using a touch of brake pedal while the car is winding down makes for big trans losses and a nice fat flywheel power curve to keep the punters happy.
On most Front Wheel drive cars trans losses are between 15% and 17% of the flywheel figure. VW themselves quote 15% as being an average transmission loss for their cars. (by trans loss I mean all losses between the flywheel and the road so it includes gearbox, final drive and tyre losses).
Trans losses will vary with the gear in which the car is tested. This is mainly due to the higher wheel speed in a higher gear leading to greater tyre losses. Always use the same gear and same tyre pressures to make comparisons meaningful for power runs taken at different times.

The ONLY WAY to accurately know flywheel bhp is to take out the engine and put it on an engine dyno. I believe that if the dyno shows much higher losses than this then you are being told porkies. They give a decent guide to power and are of course mainly of use in setting up fuel and ignition curves. For accurate work you need an engine dyno where you can control oil and water temps to get repeatable results.

 

[California Saab Tuner]

I agree Stanley ... but that is a little off the topic I had intended for. Perhaps the title of the thread was a little miss-leading, for which I apologize.

Really I would like to focus on whether the AT THE WHEELS figures are accurate or not. I generally use a 15-17% loss for FWD as well, and sometimes I wonder if the dyno operators don't already know what the car is "supposed to have" in the first place, and adjust accordingly.

Worse still, frequently with cars like the new USDM Evo and STi I find that many shops find ways even to exaggerate the at the wheels figures. Which was the reason I started this thread. Even these are subject to question. Emerald seems to be within reasonable limits, but other dynos are certainly not.

I'm hoping soon to get a scan tool which will read the mass airflow into the engine, I believe this will be ironically far more accurate than actually measuring the power output, though not as precise. Since for similar tuned engines any given amount of air will make a given amount of power (assuming it's running at full potential with no knock) ... if the airflow is consistent with dyno results I'll be satisfied.

Even more interestingly I'm looking into mods that increase power without increasing airflow or fuel rates. Like ceramic and moly coatings. This would allow more power, without upgrading the ECU, turbo, or fuel system. They also should increase mileage as the same fuel makes more power, and thus you need less fuel to drive the vehicle on the road. But again this is a bit off topic ...

Has anyone attempted the above method? I know a few members around here have the AP22 unit ... make sure to let us know if you give it a go.

Dubbya~

edit: BTW the type of dyno I was on was just a single roller in the ground. I'm told it was good to upwards of 800 hp! He certainly strapped the car down accordingly.

It has a special traction material on the roller as well. I think this would be the most accurate method outside an actual engine dynamometer.

Also I went down to the dyno place today to talk to the operator about something. My figures are in fact corrected at the wheel figures for atmospheric conditions ... however before you jump to the "I told you so" routine ... they were corrected DOWN ... the true at the wheels figure is what appeared on the screen after the run. The highest I remember seeing was 214, and I remember being a little dissapointed when it came out to 213. It was always one or two lower on the printout than the screen. At the time he had told me the printout was "more accurate" ... at any rate I don't doubt the correction too much. It was a cool good day for a dyno run. So I don't mind being penalized only by one horsepower. It's still right where EVS said all the other Viggens were. And for the record I ran in 4th gear on the dyno. Might have been higher or lower in 3rd. We'll never know.

 

[StanleyB]

I am not trying to derail your topic Adrian. But I didn't want to start another thread on basically the same topic, as that would only confuse people. Best if we put all the relevant bits together, it makes for far better reading and discussion .

 

[Alex_S]

Wheel figures can be even more variable than trying to calculate transmission losses.

Things such as tyre pressure, make of tyre (affects the sidewall stiffness), the aspect ratio of the tyre, the gear selected, the primary ratio, even the weight sat over the wheels (if the car is having wheel spin problems) can significantly affect the measured wheel figures.

The transmission losses consist of both a constant rpm loss figure and a power related loss. The power related loss can be calculated/estimated based on the measured wheel figures and rpms. The Emerald calculations do seem to be pretty good based on figures of a large number of standard cars that have been tested there.

 

[philip hs]

Hi guys,

just wanted to add my 2 cents to the discussion concerning flywheel hp vs. wheel hp:

I usally go to a place called "Heisel Motorsport" to use their dyno. Two very well respected tuners, Hartge (BMW) and Carlsson (Mercedes-Benz) go there, too. Both have a so called "Motorprüfstand" (engine-dyno?) where they measure flywheel power of the engine before they implant it into the car. Then they drive to the Heisel Dyno to measure wheel hp. But the Heisel dyno also puts out another flywheel figure using the method Stanley described. Acording to Heisel and Carlsson, the flywheel figure of the Motorprüfstand and the Heisel Dyno are more than 98% identical... so perhaps the flywheel figures aren´t that off...

Yours,

Philip

 

[Vigge]

Hirsch and Nordic both use 12% for losses.

 

[burre]

Hirsch uses the rundown powerloss check on the dyno run. They use whatever that shows them.

 

[Eric van Spelde]

Based on a wide range of dyno sessions that I've seen data of, I'd assume about 13-15% drivetrain losses for the transverse-engined Saabs, and 17-20% for the 99/c900 with the primary chain drive and longitudinal engine. It's far from an exact science, though.

 

[California Saab Tuner]

The C900 definitely has more drivetrain loss than the newer Saabs. Not sure what either calculates out to.

I can tell my C900 certainly has more. You can feel when the gearbox gets hot when driving around town. My C900 would get hot rather quickly, makes it a little more sloppy. But the Viggen gets hot very fast as well. It gets hot just about as quickly as the C900, but is putting in much more power, so because it is not heating up more quickly one would assume less power per hp of input is being wasted into heat.

There are other things to consider as well, like wheel weight, and tire sidewall deflection under load. The latter of which would be really low on low profile tires like the Viggens have. But the Viggens wheels are heavy as heck, with tires they are MUCH heavier than the steel wheels on my C900, and get bent WAY more easily. Go figure.

I think the Hirsch rundown technique could be very effective at determining loss due to intrinsic gearbox friction (which is very little) and rotational inertia caused by the wheels and tires.

However it's debatable whether the intrinsic friction of the gearbox is proportional to the friction when 200 bhp is passing through. There is some increase in friction as power increases, but it will not be linearly proportional to the power input, or you would have zero friction during coast-down. The exact relationship would depend on each individual gearbox design, materials, and gear oil.

Again I think the best solution is just to state the wheel hp figure, and what kind of drivetrain you use. That way the people you're talking to can decide on their own how much you were probably losing to friction. Besides ... where's the fun in bragging about 320 crank hp, if only 190 makes it to the wheels? Then they could just tease you about having the world's worst gearbox.

Dubbya~