Based on my experience and plenty of forum member input, I believe your post to be the most accurate estimate available.

 

Wow, moma told me my GED would not get me far.

 

I don't know for certain why, but I can guess, RW Dynos that use the big drums are notorious for having a lot of variability from one machine to the other. It may be the installation or the calibration or whatever but they are usually consistently high compared to a true wheel dyno which would be one that does not work on the rate of change of momentum of a drum but the power actually generated by the drive wheel. Among other advantages you can tune at constant rpms and part as well as full throttle.

I have not done any recent comparisons but on a previous car I did a number of comparisons on different dynos with only one being a electricity generating dyno. The Dynajets were all over the map; probably +/-10% variation with the same setups but the dyno with generators was the lowest of all. The owner of the shop warned me that if I was looking for something to brag with then this was not the dyno to rent. He specialized in track cars only by the way.

Anyway my point is that the drive train and auxiliary losses are probably higher than what you might think from the typical dyno numbers.

 

You are correct that dyno's vary, and which dyno is the most accurate is anybodys guess. I have not seen dynojets vary 10%. On 450 rwhp your talking plus or minus 45 hp. On a nation wide sampling, on dynojets, you will see maybe 10 or so hp difference. An ls2, h/c, LTH, cai and a tune will run 450. Some 435-440 some 455-460, but thats about it. Adding headers in California or New Jersey is going to give 25 hp on average, cai 10, heads 20-35, cams 35-45 etc. Mustang dyno's usually run lower and dynomatics run higher.

 

At 450 you are about 100-110 over stock. Headers 25, cai 10, heads 25, cam 35, tune 5-10.

 

No. Our cars can be driven down the road with the engine only producing about 20 horsepower. The drivetrain isn't consuming a constant 60 horsepower. It is consuming a percentage of the engine's power.

 

Bob, you are right the variation I observed was based on much smaller HP numbers and that probably inflated the variance I saw. Bob

 

Ok, let's look at it from your perspective. Let's say we have a 20 HP engine and we have the car on a dyno where aero and weight have no impact. All the engine has to do is spin up the dyno roller in 4th gear to ~150 MPH...it might take some time but it's absolutely possible since the only thing you're doing is adding kinetic energy to the roller. Using your 12% number from above, only 2.4 HP is required to drive the drivetrain...in your words, that would be impossible. At 6000 RPM in 4th gear, there's way more than 2.4 HP lost to oil windage alone in the transmission and rearend.

Trying to use a 1500 RPM/60 MPH part throttle cruise situation to discredit the use of a constant during a full throttle 6000 RPM/150 MPH dyno run is comparing apples to oranges. I'm not saying the use of a constant is absolutely correct, but it'll get you a lot closer than a percentage unless all a person is interested in is inflated HP numbers. In that case, we should all just use 25%.

 

So for a given rpm, it is fairly constant? If so, is there a relevant range of HP that would apply (e.g. 400 RWHP +/- 20%) before we would want to update the constant say if we add a S/C or bigger cubes?

 

Now who's talking apples to oranges? There's no way you're gonna run the drivetrain up to 6,000 rpm in a high gear, spinning a dyno with only 20 hp. On the other hand, a low rpm/low throttle very low hp scenario does show that a constant hp loss through the drivetrain is incorrect.

 

Like I said in my post, I gave the BASIC formula...if the guy knew what he was talking about, he wouldn't need an explanation or the basic formula for that matter. Not only do I not have a PhD in physics, I'm no where close.

Actually, it might surprise you to know that kinetic friction is not the primary source of HP loss through the drivetrain on a chassis dyno.

 

I'm not the one making senseless comparisons, I used your thought process in reverse to show a percentage is incorrect also. Oh yeah, I already said a 20 HP engine won't spin the drivetrain at 6000 RPM in 4th gear. It will spin the dyno rollers up to an RPM equivalent to 6000 RPM in 4th gear but not through a C6 drivetrain which means there's more than 20 HP lost in the drivetrain which means your percentage method fails.
A low RPM/low throttle scenario does not show a constant HP loss through the drivetrain is incorrect...there is a big difference in the oil windage losses in the transmission/rearend between 150 MPH and 60 MPH. There are inertia losses involved in accelerating to 150 MPH on a chassis dyno while there is zero inertia loss going a steady state 60 MPH down the road. That's the apples to oranges comparison I'm talking about. There are many, many, many things you aren't considering.

 

The difference between RWHP and FWHP will be a constant + a linear variable + an exponential variable like my formula indicates. The difference in the exponential variable will be very small when comparing before and after mods and can be ignored leaving the constant + a variable of 3-4% of the increase. Of course this is based on peak HP occurring at the same RPM as I indicated in my post above. If it peaks at a higher RPM, then you have more fixed HP loss due to increased oil windage in the transmission/differential etc. It also depends on what type of chassis dyno you're on. And that's just the beginning of this RWHP to FWHP conversion problem. The 15% number always gets thrown around but all I have to do is show one example that doesn't fit that "theory" to disprove it and there are many more examples of cars that don't have a 15% loss than those that do. You'll see front wheel drive cars with only 9-10% loss because they have a helical cut gears for the final drive (what we refer to as ring and pinion in rear wheel drive cars). At the gear tooth interface, a good quality involute helical gear will have pure rolling at the pitch line giving HP loss of only 1% as opposed to the hypoid cut gears in our cars that have a combination of rolling and sliding action which results in a large component of sliding friction. That's why you see the heavy 75W90/140 gear oil in our differentials while they can use 5W30 engine oil or transmission oil in their transaxles. The point I'm trying to make is that there is no way to convert RWHP to FWHP as there are way too many unknown variables. The best you can hope for is an educated guess and that involves the constant + variable of 3-4% then try to compensate for peak HP coming at a higher RPM and whether it was on a pure inertia dyno or an absorbtion dyno. I have a Mechanical Engineering degree with a minor in Internal Combustion Engines from Georgia Tech where I was the teaching assistant for my senior year. I ran the engine lab where we had every type of engine dyno known to man including an ancient "prony brake" dyno. I've studied this RWHP to FWHP problem for several years and the more I get into it the farther away the solution gets as I keep coming up with more and more variables and there are tons of places where HP loss occurs no matter how small it may be...even the losses due to seals count. When I started coming up with variables changing in two and three dimensions, I decided it was a futile effort and came up with the constant + variable of 3-4% as a very rough estimate that I would slide 5-10 HP based on new peak HP RPM and the type of chassis dyno. The bottom line is you'll never know unless you take the engine out and dyno it...anything else is just a guess. I refer to my method as an educated guess with emphasis on guess.


You'll do good to use the numbers from this post as a basis for guessing your new FWHP/Motor HP. Take 3-4% of the amount over the RWHP listed here and add it to the constants. For HP numbers 100 more than stock, add another 5 HP if you're on a Dynojet and another 5 HP if you're at a higher RPM for peak HP. Subtract 5 HP if you're not on a Dynojet. Or just make up a number...it's a lot easier and what difference does it make anyway, we don't race dynos.

 

See the post above.

 

The problem with this thread is that anyone who has dynoed their car after mods is desperately seeing a big Ego-padding number to justify the money spent on mods, so if you give them the option of adding a fixed number, or multiplying the dyno by a percentage, they will always cling to the idea that losses are a percentage.
The ridiculousness of that notion manifests itself when you apply it to very high HP car.
Lets say we take a bone stock LS2 making 340WHP factory SAE certified crank HP is 400). I.E. 60HP or 15% loss.
Now we twin turbo the car and it makes 800WHP. How much does it make at the crank?
860HP, or 920HP?
If you believe it makes 920HP, you have to explain to me how that transmission is dissipating 90Kilowatts of heat every second and not causing the gear oil to vaporise
In reality the fixed number is not correct either because with more torque on the drivetrain losses do go up, but it is a lot closer to reality than multiplying WHP by a fixed amount. The only time the fixed number method becomes really innacurate is if you move the RPM at which peak power is made significantly; since losses will go up exponentially with speed, making peak power at, say 6500RPM will incurr a lot higher drivetrain losses than making it at 5500RPM...

 

Thank you Sam. I know I've seen you say it many times...a dyno is a tool used for mods to make sure you're going in the right direction.

 

Ah, beat me to it. Not only for mods though, but for tuning as well. As long as the numbers go up, you're doing something right. By the way, it is definitely a % not a hard number for drivetrain loss. There have been plenty of great explanations in this thread about why. To put them in layman's terms: If you have a 60 hp draivetrain loss all the time, that would mean that at 1200rpm when your motor is making about 60 hp, that the drivetrain would take it all away and your car would not move. This is obviously not the case. There are many variables in places that also make the %, the most obvious being friction and drag. The higher rpm, the faster things are moving, the more drivetrain loss. Also, the higher hp, the faster things are moving, the more drivetrain loss.

 

An inertia dynamometer measures the rpm of the drum under the rear wheels as a function of time. Knowing the rotational inertia of the drum, and a little bit of math, the operator (or his computer) can determine the horsepower applied BY THE TIRE CONTACT PATCH to accelerate the drum.

But while the drum is being accelerated, the crankshaft, conn rods, pistons, camshaft, clutch, driveshaft, transmission gears and shafts, differential gears and shafts, rear hubs, brake rotors, rear wheels and rear tires are ALSO being accelerated. This uncounted acceleration of the pieces AHEAD of the tire contact patch account for the biggest part of the difference between flywheel HP (at constant speed, by the way) and RWHP measured on an inertia dyno.

My very stock 2006 C6 MN6 dynoed at 340 RWHP, or exactly 85% of its rated flywheel HP.

When I increase the HP with cam, heads, FAST, Kooks, etc. the car will be able to spin that dyno drum up faster, and thus measure a higher RWHP. However, the crankshaft, conn rods, pistons, camshaft, clutch, driveshaft, transmission gears and shafts, differential gears and shafts, rear hubs, brake rotors, rear wheels and rear tires are ALSO spinning up faster, requiring more of the new larger flywheel power to do so. Thus, for my case, I would still GUESSTIMATE the flywheel HP at 117.6% ( 1 over 0.85 x 100% ) of the new RWHP.

With 3.90 gears, the innards (except for the brakes wheels and tires) are spinning up 14% faster than with 3.42 gears, so I intend to perturbate the % factor by 14%, i.e. RWHP will be 83% of flywheel power for a manual C6 with 3.90 gears.

Glass Slipper and PowerLabs - your comments would be most welcome.

 

My head hurts from these threads. I am just going to go and drive my vette until the headache goes away. Should be just a few seconds

 

I have some experiance with this subject on my 68'' nova 360 small block. I have had it rebuilt twice both times I ran it on an engine dyno and on a mustang chassis dyno. The first time was a mild strret build up that put down 380 engine hp, when it was put in the car it made exactly 304 wheel hp exactly 20% off and 76 hp down. Then it broke a pushrod and I rebuilt it with a hotter cam and some AFR's and race gas compression, this time it put down 512 engine hp. In the car it put down 410 wheel hp exactly 20% down and 102 hp down. The car has a powerglide they were done in the same month one year apart within 4 degrees ambient temp of each other. The drivetrain remained unchanged and the only reason why I had the engine dyno done was because all the engines that shop did were broke in on the dyno. Not trying to pick a fight just thought you guys would want to know.