Dyno numbers misleading?

[olddog]

Every dyno run I have seen on TV were, what I will call, a dynamic, as apposed to static, test. What I mean by dynamic is that the motor rpm are increasing, as the measure of torque is made. In a static test, the motor rpm would be constant, as you measured the torque, in which case the torque output of the motor would be in perfect balance with the resistance of the dyno. It is easy to see that doing a static test at every few hundred rpm would be a real PITA. Hence the use of the dynamic test.

However this brings me to my point. An engine has a lot of rotating mass. The bigger the engine the more the mass. It takes energy to accelerate this rotating mass, and if I remember my physics correctly this is called inertia. Let's say, for purposes of discussion, that a particular engine at WOT takes 0.6 seconds to free rev from 2000 to 6000 rpm and that the engine outputs a true time weighted average of 400 lb-ft of torque through this rpm range. This then means that it takes 400 lb-ft of torque to accelerate the rotating mass of this engine from 2000 to 6000 rpm in 0.6 seconds. If we cut the output to 1/10 or 40 lb-ft, it would stand to reason that it would then take 10 times as long or 6 seconds to accelerate from 2000 to 6000 rpm. Now if we put this engine on a dyno, and during the dynamic test the engine goes from 2000 to 6000 rpm in 6 seconds, what would the dyno measure? I suggest that it will measure the torque output of the engine as a time weighted average of 360 lb-ft, as 40 lb-ft will be going to accelerating the mass of the engine and 360 lb-ft will be going to turning the dyno, which adds up to the 400 lb-ft of the true torque the engine can output in a static condition.

To go one step further, if the dynamic dyno test took 12 seconds to go through the same 4000 rpm range on the same engine, then only 1/20 of the true torque would be going to accelerate the rotating mass. So in this test, 20 lb-ft would be lost to inertia and the dyno would measure 380 lb-ft. Like wise if the inertia was cut in half and the engine only required 0.3 seconds to free rev through the same 4000 rpm range and again the dynamic dyno test took 6 seconds, we would have the same 1/20 of the true torque lost to inertia and would measure 380 lb-ft.

I believe this explains why a chassis dyno measures less torque/hp at the wheels if you measure in 3rd gear than in 4th gear. The lower the gear the quicker the engine revs up and the more energy is consumed by inertia.

What do you think?

[olddog]

Ok. I never claimed to be the sharpest tool in the shed. I know some of the people (members) own and operate their own dyno. Someone has to have an opinion on this. Am I on to something, is this totally wrong, or is this already known by everyone (except me)? Help me out here.

[Anthony]

Why not email this question to one of the Dyno manufacturers, as I would think they realy understand the theory/physics of their product, and be able to explain what their readings mean. If you email them, post the answer here.

[Power Surge]

Your post made my head hurt.

What is your actual question?

If it's about the affects of different gears for a chassis dyno, I think it's pretty simple. Any type of gear reduction, will show lower numbers, because it takes less engine hp to move the inertia mass (the dyno roller) because the gear reduction "assists" it. For an accurate power reading, you want to dyno in whatever gear is closest to 1:1 because that's your true power output of the motor through all the driveline. If you have an automatic car, you want 1:1 with the converter locked to eliminate all slip/gear reduction.

[olddog]

I took Anthony's advise and read on DYNOmite Dynamometers web site. In an artical I found this paragraph.

"Even if you select a low inertia brake remember that the engine's moving components still have there own inertia. If you take readings while the engine is accelerating or decelerating, inertial energy is being subtracted or added, respectively, to what your gauges indicate. Disappointingly, unscrupulous dynamometer operators use inertia to display impressive flash power readings by suddenly cranking on the brake load. Obviously such "inertial energy augmented" numbers have nothing to do with the true horsepower capabilities of the engine. After you run a dynamometer for awhile, you can spot such shenanigans in other's printed dyno data. This is another reason engine builders get their own dynamometers."

In summary, I think they are saying that yes I am right. However they do talk about doing what I called a static test at every 250 rpm, and they did mention when talking about flywheel type dynamometers and I quote "Fortunately, high end computerized data acquisition systems provide composition algorithms to back out the effects of absorber (and crank-train) inertia from acceleration data. On a high inertia dynamometer, compensation is required even for fairly low rate sweep testing." So maybe some dyno operators are correcting for this.

I have noticed that on TV some of these dyno runs are as short as 3 seconds while others are 8 seconds or longer. This can impact the numbers. I think when considering big block engines, these errors can get very missleading.

I have read on here many times about someone very confused at how they lost so much HP from the engine dyno to the chasses dyno. When you consider the large inertia of a 4+" stroke FE, large diameter flywheel and clutch, large wheels and tires, and heavy duty rear end and tranny, a 3 second chasses dyno run is going to measure much lower numbers. This is due to so much power going into accellerating all this enertia. Let's face it your car is not going to go from 40 mph to 120 mph in 3 seconds, at a 1:1 gear ratio.

Sorry about the head ache Power Surge.

[Power Surge]

Originally Posted by olddog

I have read on here many times about someone very confused at how they lost so much HP from the engine dyno to the chasses dyno. When you consider the large inertia of a 4+" stroke FE, large diameter flywheel and clutch, large wheels and tires, and heavy duty rear end and tranny, a 3 second chasses dyno run is going to measure much lower numbers. This is due to so much power going into accellerating all this enertia. Let's face it your car is not going to go from 40 mph to 120 mph in 3 seconds, at a 1:1 gear ratio.

The length of a dyno run is based on the vehicle itself, and not so much the dyno. A more powerful vehicle will accellerate faster on a dyno, just like it would on the road. It also depends on what rpm the dyno operator starts the dyno at. If you start the dyno at 4500 and the motor only goes to 6500, that's going to be a very quick dyno. I usually start about 3000 rpm, and pull to redline, which varies with vehicle. But the average time for a pull on my dyno, I would say is about 6-7 seconds. It also depends on the TYPE of dyno. I use an inertia dyno. So the vehicle is moving a 3000 lb mass on my dyno. Obviously a more powerful car will spin the mass faster. Some people use eddy current dynos, that "load" the dyno with a specifc load, set by the operator. With these dynos, you can adjust the amount of load, which naturally will affect the numbers.

The one thing I always suggest to people, is to not get caught up in the peak numbers. The dyno is a tool, used for tuning. Where you finish hp wise is not as important as how much you gained from the tuning.

Cobras have a LOT of parasitic loss, much more than any other vehicle. IRS eats up power, and so do Cobra side pipes. I have dynoed vehicles with the same motors as in some Cobras and they don't have anywhere near the loss.

Big block, small block, it doesn't matter.

[RodgerH]

Originally Posted by olddog

It is easy to see that doing a static test at every few hundred rpm would be a real PITA.

Hey olddog,

I don't know if you have ever watched an engine dyno pull, but when you are "up close and personal" it gets the adrenalin flowing - especially if it is your engine!

I watched my engine go thru several pulls from 3000 to 6000 RPM on the engine dyno at Performance Engineering. Eric would bring the engine up to about 3500-4000 RPM, then he let the computer control the pull. The computer took the engine to full throttle while loading the dyno brake, pulled it down to 3000 RPM and started recording data. Over a period of 10 seconds it increased to 6000 RPM, then Eric would stop the pull. The following data was recorded in 100 RPM increments: RPM, torque, BSFC, A/F ratio, water temp, oil pressure, barometric pressure, and relative humidity.

That last 1000 RPM increase was pretty intimidating! I can't imagine stopping every few hundred RPM and letting the engine stabilize long enough to get static numbers. I don't think I would want to subject my engine to that much punishment.

Rodger

[Bob In Ct]

"I believe this explains why a chassis dyno measures less torque/hp at the wheels if you measure in 3rd gear than in 4th gear. The lower the gear the quicker the engine revs up and the more energy is consumed by inertia."

Hey OldDog

I always thought the reason you used 4th gear was to eliminate a gear which did not have a 1:1 ratio, which is usually the case in 4th.

Bob

[Tinker51]

Originally Posted by RodgerH

Hey olddog,

I don't know if you have ever watched an engine dyno pull, but when you are "up close and personal" it gets the adrenalin flowing - especially if it is your engine!

A friend is an engine builder who has a dyno. When I send other friends to get engines built I tell them to not go see the dyno pulls. Very nerve wracking to see your brand new high dollar motor screaming at 6-7k rpm. Interesting to watch though and amazing how he can look at a graph which looks like a pile of pasta to me and determine exactly what the motor needs to make more power. Art and science.

[RodgerH]

Originally Posted by Tinker51

Very nerve wracking to see your brand new high dollar motor screaming at 6-7k rpm.

Excellent description - and still an understatement!
Rodger

[Anthony]

Originally Posted by Bob In Ct

"I believe this explains why a chassis dyno measures less torque/hp at the wheels if you measure in 3rd gear than in 4th gear. The lower the gear the quicker the engine revs up and the more energy is consumed by inertia."

Hey OldDog

I always thought the reason you used 4th gear was to eliminate a gear which did not have a 1:1 ratio, which is usually the case in 4th.

Bob

I would think running your tranny with 1:1 ratio will be the most efficient transfer of power(energy) through the tranny, as well as reduce the torque multiplication if instead you had used a lower gear.

[ItBites]

All the supposition from the initial poster is true.

The reasoning is correct when an engine is accelerating. A Brake dynamometer will give more accurate numbers than an Inertial dynamometer... Always!

Inertial dynamometers should be used for comparison purposes for back-to-back tests to determine results of changes, not for determining SAE-certified horsepower. To get true SAE horsepower, you need a Brake Dynamometer.

[Bob In Ct]

I was actually quoting olddog in the first part of my post.

Bob

[1blackbb66]

Sal,
Why is it you feel that Cobra side pipes eat up so much power? Inherent design flaw? I have heard this before but all things being equal, (pipe size, muffler flow, etc) they wouldn't be any worse than an under car system on a Mustang.

Ray

[Cobrabill]

An IRS suffers no more loss than a live axle.

[Power Surge]

Originally Posted by Cobrabill

An IRS suffers no more loss than a live axle.

Yes, it absolutely does. About 3% more loss over a live axle.

[Power Surge]

Originally Posted by 1blackbb66

Sal,
Why is it you feel that Cobra side pipes eat up so much power? Inherent design flaw? I have heard this before but all things being equal, (pipe size, muffler flow, etc) they wouldn't be any worse than an under car system on a Mustang.

Ray

Well, Cobra sidepipes are basically open headers with some kind of device to reduce noise. Most stock type sidepipes are baffle plate type (like a motorcycle muffler), and they totally kill performance. There are other better flowing designs, but basically most pipes people run to keep the car from sounding like an open exhaust dragster, chop power pretty good. Plus, with a regular under car exhaust the piping can help scavaging velocity, and you just don't get that with a Cobra sidepipe.

You have no idea how many people come to my dyno with expectations on rwhp, previously having their motors engine dynoed. The average IRS Cobra with a motor that makes 550hp on an engine dyno, will put down barely 400 rear wheel. That's a little over 25% loss. But a non-Cobra vehicle with a manual and IRS will only have about 15% loss.

[1blackbb66]

Sal,
I had a chance to go and talk to my dyno guy tonight and he had many thoughts on this, some in line with what you are saying. The real problem often appears to be threefold. First due to space and many times aesthetics, is that the primaries are of unequal length and how they merge into the collector area is not optimum. He says there needs to be a stinger or cone in the collector area to aid in maitaining flow velocity as well as taking advantage of pulse scavanging. Some manufactures address this but most do not. Two...no crossover pipe to balance the exhaust. Third, and probably the worst, is very restrictive mufflers in order to reduce sound levals. With such limited space and again, aesthitics, not enough room to have a good flow path through the muffler that maintains velocity and reduces sound levals.
My engine builder says the way he put my 460 together it should be making around 475 HP. It was never dynoed so I have no confirmation of this but on the chassis dyno we only got 283 to the tire. Needless to say I was quite dissapointed. I have compiled a lot of good information and I will be addressing the side pipe issue.
Thanks for the info.

Ray