So I was thinking last night and doing some research on dynos and it got me to thinking. There is alot of confusion about what "rwhp" truly means, because it is not easy to directly calculate. All types of dynos require some sort of correction factor which is why all dynos are not created equal.

Ill explain the basics of dynos before I go into my "idea" if you will, so you can better understand it. There are 2 types of dynos, inerta and loading. Inertia dynos (ex: dynojet) work on the principle of physics that takes a certain amount of torque to rotate a mass of a given moment of inertia (basically rotational mass) to calculate HP via RPM. Loading dynos (ex" Mustang) work on the concept that you can acurately measure the amount of TQ given the amount of resistance required to control the power.

The problem with dynos is setting a standard. There is alot of crying about mustang dynos because they do not put high numbers like dynojets and people want to brag. The reason for this is due primarily to drivetrain weight.

Dynojets tend to be the standard for this reason, but there are inherent flaws in an inertia dyno. For example, if you have a dyno with a very low moment of inertia (not alot of mass to rotate) then drivetrain mass (actually drivetrain MOI) will make a SIGNIFICANT factor in the numbers produced. This is because you must consider the moment of inertia of the entire dyno/drivetrain system as they are being accelerated together and if the the MOI of the drivetrain is larger than the MOI of the dyno, then it can make a VERY big difference (this is why there are correction factors). If you made the inertia VERY high, then the drivetrain "mass" becomes insignificant compared to the "mass" of the dyno wheel and it calculates HP without consideration to drivetrain mass and simply calculates the amount of TQ actually produced in the motor (it would even disregard motor rotating mass to an extent) and only adjusts for friction.

Now where do you set the standard? Is RWHP simply chemical HP produced in the cylinders minus engine and drivetrain friction or should rotational mass be considered? These are the problems that Dyno producers have. If you could set up a system that would use an Inertia setup with a variable Moment of Inertia and run multiple pulls, you could actually calculate the rotational moment of inertia of the drivetrain as well as the Chemical HP minus friction... using that, somebody could derive a scale to "calculate" a standard for RWHP. This would allow you to get 2 numbers from a dyno... HP and MOI of drivetrain to better compare cars.

Anybody follow my train of thought here, or am I just delusional? I think it could be a very good idea and change the performance world for standards and comparison. It would set a level playing field and quit all the guessing with regard to dynos.

 

I've seen and so have a bunch of engine guys that RPM/sec rates on a dyno aren't high enough to show a difference in TQ numbers. Theoritically it SHOULD but it's hard to see. This is on engine dynos with motors with heavy cranks and parts vs. motors with lightweight parts usually sweeping at 300-600rpm/sec. On the track it's a different story.

I personally think a Mustang or Eddy Current style dyno is far and above better than a Dynojet. The fact that they don't match up in power numbers could be fixed in the settings on the software. I know that the Superflow Eddy Current dynos have a output that correlates with dynojet numbers.

I'm sure the old man will be all over this one and the physics.

Bret

 

Sorry to copy and paste.. i posted this on a local forum and people didnt know what I was talking about... so here goes:

........It is actually a very good idea to calculate HP without guessing. Let me give this another shot.

An inertia dyno uses a wheel with a set "mass" (really moment of inertia) and it measures how fast the car can accelerate it. The dyno program has correction factors in it to calculate HP because the drivetrain also has weight.

You have to take the dyno and drivetrain as a whole system in order to calculate the TQ at a given RPM to get HP. Because you dont know what the mass of the drivetrain is, they have to guess.

If you can make 2 (or more) similar dyno runs on a dyno wheel with different masses, you are given 2 separate equations with 2 unknowns. The 2 unknowns are moment of inertia of the drivetrain as well as the HP produced at the wheels.

Similar to F=MA, for rotational masses, Tq= I x (x=alpha, rotational acceleration. I = moment of inertia). Now as a system:

(IW1+ID)*x=Tq
(IW2+ID)*x=Tq

The known variables in these equations are IW1, IW2, and x (IW1 & IW2 are the moments of inertia of 2 separate dyno wheels used for comparison, x is acceleration calculated by the dyno). The unknowns are Tq and ID (moment of inertia of drivetrain) and can be solved with those 2 equations.

The only problem would be different gear ratios affecting moment of inertia of the drivetrain, which im not 100% sure about. Maybe someone can help out here.

Any thoughts? Anyone? Bueler?.........

----------------------

Another thing to add would be the problem with consistent dyno runs. If the inertias were drastically different, then it would cause acceleration to vary to a point where you get into tuning and cylinder head flow differneces with a motor making different power depending on how fast the motor is accelerating (ex power produced under heavy load like a hill vs power produced during hard acceleration like a launch in 1st). It might be beneficial to find a guess where the inertias of the system were a standard difference and you could run it with different tranny ratios so the acceleration would be the same time based on gearing... kind of an odd theory.

Sorry for the long drawn out post, just trying to be thorough.

-Stu

 

I largely agree with whats been written.
.
For either Mustang or Dynojet, is the raw
'drum timing' available?

From either, do we know {all} what calculations and
corrections are done?

For a typical drum, how much is the MOI?
for a typical car, how much MOI?
Does either make a correction for 'tire pinch'?
Suppose some 300hp car, with stick shift, was tested.
What gear would be used, how long would the 'pull' take.
{how fast is the car 'going' at 6000 rpm?
I assume there are different models, be as specific
as you can.
What do these companies say about the 'unknown
drivelime MOI' issue?
Most of a cars MOI,,,, is in the wheels-tires ?}
Can a Mustangdyno dissapate 300 HP? if not, how much.

 

A couple of threads have discussed this a while ago:



http://web.camaross.com/forums/showthread.php?t=236297

http://web.camaross.com/forums/showthread.php?t=176810

No offense intended disco, but you are a little confused about some things. Read the parts about what an inertia measures and what the software calculates.

 

Maybe I didnt clarify, but I beleive that I have a firm grasp on the concepts. If I have made any mistakes it is most likely due to oversimplification. If I am blatantly wrong, Im sure you could point that out. Thanks.

First off, you must assume that all the equations are either instantaneous or a time average of a very short period. The graph will be a representation of many calculated events fit to a curve.

Lets see here...

Power is work per unit time and work is torque done over a given radial distance, thus power = torque * angular velocity.

Torque is also the same as moment of inertia * angular acceleration. Thus Power = MOI * angular velocity * angular acceleration. An inertia dyno cannot directly calculate torque because it does not know gear ratios that engine TQ is multiplied by and thus must be calculated by RPM.

When I said that earlier it was simply to clarify things.

How did I do prof?

 

The inertia dyno calculates how much hp it took to accelerate the drum(s). Bearing friction and windage of the dyno itself are built into the calculations. The dyno doesn't really care what is accelerating the drums nor what gear ratios or converter slippage the vehicle is experiencing.

Gear ratio isn't needed if you measure engine rpm and know rwhp. If one measures engine rpm during those small increments (1/10 second, 1/100 second or whatever) and you know rwhp (inertia wheel hp), with engine rpm you back into engine torque (reduced by driveline losses), right? Call this "rw engine torque".

If you are calculating power from torque and angular velocity (AKA rpm), make sure you get the units correct.

We're pretty much on the same page.

 

So any ideas on a variable inertia load to single out the drivetrain weight variable? I think its a good alternative and results could easily be repeatable on separate dynos. It would also explain why some cars are dyno friendly and some are track friendly.

 

Last I checked you race cars with tires/driveshafts/transmissions so the point is moot. Once you baseline the car you can make incremental changes to the entire propulsion system and verify what, if any, difference it made. The point of a chassis dyno is to test the whole system. (well actually its to make money) An engine dyno focuses on the engine and accessory drag so you can focus on that system without putting other systems into the mix and skewing the results. What matters is how it all comes together. Bench racing is one thing and throwing a HP number around to compare with friends makes for good bench racing, but personally at least what matters is how it works on the street and at the track.

Eddy current dynos can dissipate over 800hp now. There is no better way to map a fuel system than being able to load test your combination on a mustang dyno.


Track friendly cars have had more of the "system" thought through. You know suspension and stuff so they actually use what power they have more effectively than the guy who hasn't yet.

 

I agree with what you are saying, but if you did it this way, you could calculate both the Chemical HP created in the motor minus the frictional losses through the motor and drivetrain moment of intertia. From there, if you wanted to... you could create a standard of comparison. This would also work for just about any car regardless of drivetrains.

For example, at top speed... drivetrain weight makes little difference. All you care about would be HP produced minus the frictional losses. The weight doesnt make your top speed any lower unless it causes more friction. It just takes you longer to get there.

If you could calculate both, it would take all the guessing out mods and could get an actual comparison of gains. For example, switching to an aluminum flywheel and carbon driveshaft will make you dyno higher, but how much would it affect your performance? If you knew the difference in MOI of the whole system of before and after, you could more accurately see the benefits and understand why things do and dont work.

Then when you get a printout of a dyno, you would get HP AND MOI of the drivetrain.

If nothing else, it would give very accurate and consistent dyno results. It would also better educate the general public on what "drivetrain losses" really are. Maybe I am just OCD about these things and HATE estimates. I want to KNOW!

 

I'm not bagging on your reasoning, we all want to learn more. Surely a coast down test would be able to tell you what you want to know. I would bet there is a formula for figuring coast down from a specific mph to another mph (drum rpm) to factor frictional losses.

 

Dynojet has a "coastdown" option for estimating drivetrain HP loss due to inertia.

Dynojet also now also makes a chassis dyno that incorporates the Eddy Current Load Absorption Unit.

One way to eliminate a lot of the guesswork, and theoretical calculations of questionable accuracy is to test the engine first on an engine dyno with full intake, exhaust and accessories, and then on a chassis dyno, while holding the rpm/sec rates similar in both tests.

I can tell you with a fair degree of certainty what the losses were with my setup in both T56 and TH400 configurations.

I agree..... OCD

 

Mark's point is close to what I want to say. You have to ask "what do I want to accomplish with the dyno?" I think of it as a tuning tool. A unit like the Dynojet allows me to make observations and measurements under load and see how any changes I make are reflected in the power output AT THE WHEELS. Since the car must have a drivetrain to be useful, I want to know the effect of changes there as well as motor changes. What matters the most though is reproducibility and sensitivity. These allow the effects of changes in the combo to be measured. Accuracy, in the sense of producing a number close to the "gold standard" is not an issue.

But unless you are a Supra owner, you don't race dynos. You race cars. What matters most is how the combo (including driver) performs. Sometimes it seems like a copout, but it really reflects reality when an F1 driver is interviewed after a poor race and refers to how the team just didn't have the "package" today. The package includes the engine, the chassis, the setup, the electronics, the tires, the driver, etc. All have to work right and all have to complement another.

Rich

 

Now that's funny. I don't care who you are, that's funny.

 

How about this kind of dyno. IMHO, it is the best for tuning.

http://www.dynapack.com