Dyno SAE correction math

[rrrocketman]

So I'm looking at the so-called gold standard for dyno correction factors which is the 2004 update to the SAE J1349. I'm reading about how it works and it talks about assuming a mechanical efficiency of 85% or in other words a frictional loss of 15%. Problem is it's kind of vague. What friction? Friction in the engine? Friction in the dyno drum? Where's the friction?

If it's talking about the accessories and other crap dragging on the engine, then I see no reason why this would need to be "corrected". When you dyno your car, you want the power that gets laid down on the road, inefficiencies and all. To correct for frictional losses is to artificially increase your results, which doesn't translate into real world performance.

Having said all this, perhaps there's another type of friction being considered here or perhaps I'm just misunderstanding some detail that is well understood. I was just hoping someone could clear this up for me, since my most superficial interpretation has me convinced that this standard is used for lab testing an engine or something.

Thanks.

[rrrocketman]

Anybody know about this?

[CamaroDreams76]

I always read that correction factors are for weather.

SAE J1349 standard day is 77ºF, 29.23inHg (inches of Mercury) pressure, 0% Relative Humidity.

Motorsports Standard Atmosphere (MSA) is 60ºF, 29.92 and 0% humidity.

STD standard day is 68ºF, 29.92 and 0% humidity.

I'm sure somewhere along the lines this has to do with friction and thermal energy loss. I believe the friction you want to know about the the friction inside a motor. Its standardizing that the operating efficiency of a motor is 85% and 15% is lost due to friction/heat.

[lt4camaro]

Originally Posted by rrrocketman

So I'm looking at the so-called gold standard for dyno correction factors which is the 2004 update to the SAE J1349. I'm reading about how it works and it talks about assuming a mechanical efficiency of 85% or in other words a frictional loss of 15%. Problem is it's kind of vague. What friction? Friction in the engine? Friction in the dyno drum? Where's the friction?

If it's talking about the accessories and other crap dragging on the engine, then I see no reason why this would need to be "corrected". When you dyno your car, you want the power that gets laid down on the road, inefficiencies and all. To correct for frictional losses is to artificially increase your results, which doesn't translate into real world performance.

Having said all this, perhaps there's another type of friction being considered here or perhaps I'm just misunderstanding some detail that is well understood. I was just hoping someone could clear this up for me, since my most superficial interpretation has me convinced that this standard is used for lab testing an engine or something.

Thanks.

Its really drive train loss. Engine power is measured from the flywheel. So you have power loss thru the transmission, thru the drive shaft, thru the rear end, thru the axles, tires and wheels. so a good rule of thumb is a 15% loss of engine power to get it to the ground. You have 460hp rated engine in a car that puts down 400 wheel horsepower on a chassis dyno which shows how much power it takes to turn the complete drivetrain. 400 + 15% =460

[joshnmn]

http://https://www.sae.org/standards...rtified-power/

[rrrocketman]

Originally Posted by lt4camaro

Its really drive train loss. Engine power is measured from the flywheel. So you have power loss thru the transmission, thru the drive shaft, thru the rear end, thru the axles, tires and wheels. so a good rule of thumb is a 15% loss of engine power to get it to the ground. You have 460hp rated engine in a car that puts down 400 wheel horsepower on a chassis dyno which shows how much power it takes to turn the complete drivetrain. 400 + 15% =460

I could get on board with that IF you were trying to calculate brake power of the engine. BUT, when you get a dyno you want the power to the wheels. You're measuring dyno drum power and correlating that to the power the car puts down on the concrete because this is what gets you your 0-60 or your 1/4 mile or whatever. The engine power before getting attenuated through the drivetrain is largely irrelevant to the customer.

Joshnmn: link broken.

[joshnmn]

https://www.sae.org/standards/develo...rtified-power/

[joshnmn]

That one works.

It says right in the article that the standard is to certify factory engine numbers. So... For instance, GM can claim the LS7 make 505hp, even though it was never on an engine Dyno.

[ADVMax]

Actually hundreds of hours on a dyno. Dynos are the main test tool for manufacturers doing R&D on engines old and new. I have to think that somewhere in the several pages of data they are usually monitoring HP has to be on the list? Manufacturers are looking for a whole other list of stuff when doing different levels of engine testing. Wheel testing is leaning toward emissions and mileage for them as well as longevity. We work with all the major manufacturers on the foundry side but you meet people. I’ve had opportunity to watch factory engineers spray engine innards all over a dyno cell and celebrate for the data they got. And getting another engine to play with is a non issue when it’s factory. You can’t imagine the fun shit automotive Engineers get to do.

You can bet before we got to see our Camaro hot rods some factory boys beat the lovin snot out of some test mules before they started punching them out and selling them to us.

[BehindBlueI's]

A lot of you guys are confusing the dyno correction formula with the SWAG of engine horsepower from a vehicle dyno. That's not what the SAE standard is about. It's about correcting for air density due to temperature, elevation, etc.

Dyno a motor on a 5 degree day at sea level.
Dyno the same motor on a 110 degree day at 5000' feet.

Will the numbers be the same? No, of course not. We all know colder air is denser air and that the higher one goes the thinner the air gets. It's unrealistic to expect every dyno run to be at the same barometric pressure, same temperature, etc. Results would not be standardized. So a formula based on how much oxygen is available to the motor is used to standardize tests done at different temps/pressures/etc.

[rrrocketman]

There's a couple of different streams of information going back and forth here worth discriminating between.

Joshnmn: I'm inclined to agree with your assertions but if that's the case, the formula has no business being used at the consumer level for a chassis dyno, where the customer wants to know what real power their car puts down. That 15%, which helps auto makers predict the gross power of their engines, is not what a street dyno tells you. A street dyno tells you that a 426 HP advertised camaro puts down 360 for example. But you'd get the 360 from direct measurement. Assuming the environment matches the formula mean (990mbar, 25C), you'd get 15% more than that to account for losses??? Then you're back where you started at 426 again. What's the point?

BehindBluel's: I think it should be about the weather, as you said. However if that's the case then why does the standard talk about mechanical losses at all?

Perhaps... and this is just a wild guess. Perhaps, the dyno shops are quoting SAEJ1349 but are simply pulling the environmental correction factor equation from it and nothing else. Therefore it's a bit of a cheat. They are not following the full procedure, just pulling the portion that serves their purposes... the environmental correction factor.

The equation is 1.176((990/Pd)((Tc/273)/298)^0.5) - 0.176

If you enter 990mbar and 25C into this equation you get exactly 1. This means when you dyno your car and get 360, it comes out as 360 both before and after SAE correction. Therefore, if the weather is perfect, there is no adjustment. That means 1 of 2 things. Either the formula is bullshitting about the 15% losses OR (and this is what I think), this factor goes into a larger equation that adds the 15% but dyno shops aren't using that part of the procedure.

[BehindBlueI's]

Originally Posted by rrrocketman

That 15%, which helps auto makers predict the gross power of their engines

The manufacturers aren't using a chassis dyno then predicting the power of the engine via calculation. They are using an engine dyno and directly measuring it via crankshaft/flywheel.


15% is just kind of the hotrodder's rule of thumb for quick 'n dirty guesstimates. It's not some universal scientific rule. For a fairly deep dive, see: https://www.motortrend.com/how-to/mo...in-power-loss/

[rrrocketman]

That's not what I meant though. The standard itself (or the excerpt I read at least) explicitly mentions that a 15% loss is baked into the calculation and it even goes so far as to point out that STD doesn't do this but SAE does. So I know the rule of thumb you're talking about but what I'm talking about is something in the literature saying that this is how the calculation for SAE works. Now the Cf coefficient doesn't show any evidence of this at all but because I don't have the standard to consult I can't see if it's part of a larger equation. If it is, then either the dyno shops aren't using it, which is fine or they are using it and it's distorting their power figures severely. Either way it would be good to know what math they are using because 15% is a lot.

[gtstorey]

I didn't find the full text for J1349 but din find the description of on the SAE website
"This standard is intended to provide a method to obtain repeatable measurements that accurately reflect true engine performance in customer service. Whenever there is an opportunity for interpretation of the standard, a good faith effort shall be made to obtain the engine’s typical in-service performance and avoid finding the best possible performance under the best possible conditions. Intentional biasing of engine component or assembly tolerances to optimize performance for this test is prohibited."

It looks like this spec is written to compare engine performance, not full system performance. It is probably written to get everything as close to equal so that if you compare an engine on an engine dyno to one in a car at the wheels, they are using the same base line.

The wheel HP SAE correction done on the dyno is probably used partially to correct for the same weather & altitude conditions for comparison conditions, but likely shouldn't use the 15% friction lost unless comparing to an engine that is tested on an engine dyno.