Dyno ??

[More Power]

Horsepower & torque are measured (computed) by a dyno as the truck spins the rollers. This is rear wheel power.

What assumptions can we make about flywheel power for trucks with different gearing and tire sizes?

Would flywheel power be different for two trucks dynoing the same numbers, but having different differential gearing and tire size?

Jim

[rjschoolcraft]

The dynamometer measures torque directly. It knows how much reaction torque it is generating at any particular point in time. The power is computed by taking the measured torque and multiplying by speed (then dividing by the appropriate constant to get the units to work out right).

As long as the dyno controller knows the relationship between engine speed and rear wheel speed, there should be negligible difference. There may be less than a tenth of a percent difference in loss from one gear ratio and tire arrangement to another, but I doubt that it is measureable.

The driveline losses should be fairly consistent within brands. Manual transmissions will have less loss than automatics. There will be differences in loss between brands of automatics as well. The Allison will probably have higher loss than the other brands.

[fourtenposi]

Wouldn't gear ratio have to come into play somewhere on a chassis dyno? You will be able to accelerate the dyno drum faster with 410 gears as opposed to 2.73 gears. Do you have to input the gear ratio somewhere?

[rjschoolcraft]

As I said, as long as the dyno computer knows the relationship between engine speed and wheel speed (read final drive ratio) everything is ok. There is also a calibration for vehicle speed (takes care of tire size). If any of these factors are incorrectly handled, then yes, there will be an error.

[Jim Brzozowski]

MP, somewhere in the back of my mind I seem to remember some SAE paper that showed something like a 10%loss due to driveline and accessory drive systems, but I just can't remember where the article is, maybe one of the performance magazine editors may have access to such info. on their computer of old reports on automotive technology.

[Jim Brzozowski]

Just found this formula in my Marks Engineering Handbook for calculating HP.'
2*pie*n*T/k
with n+ rpm
T=torque
k=33,000ft-lbf/hp-mim
So taking Kennedy" measured 1382 ft. lbs of torque
and multiplying it out6.283185x3200x1382/33000=
802HP. So I'm saying he was making about 800 horspower while the chassis dyno was saying if I remember correctly 725.3 HP so thats close to the 10%loss I was refering to.

[Kennedy]

In our case, we typically see 240-250 RWHP from a stock Dmax rated at 300 HP. Assuming the 300HP is accutrate and consistent, 20% loss seems std.

[rjschoolcraft]

Originally posted by SoTxPollock:
So taking Kennedy" measured 1382 ft. lbs of torque
and multiplying it out6.283185x3200x1382/33000=
802HP. So I'm saying he was making about 800 horspower while the chassis dyno was saying if I remember correctly 725.3 HP so thats close to the 10%loss I was refering to. That measured torque has the driveline loss in it...because it was measured at the wheels.

The actual measured torque is actually considerably higher (probably 5155 lb-ft in Kennedy's case). The dynamometer output is "corrected" for final drive ratio to transpose the numbers to engine output. What is being reported is torque and power delivered to the ground at engine crankshaft speed.

The reason that your numbers don't "match up" is that the torque peak will occur at a low rpm, say 1800 or so. The power peak will occur at high rpm, say 3200 or so. There is no magic in computing power...it is torque times speed with a units correction factor.

[ 09-19-2005, 09:20 AM: Message edited by: ronniejoe ]

[More Power]

SoTxPollock, I recall from my reading reports about shaft drive motorcycles that they lose about 7% through the combination of gears, compared to a chain drive. Automotive applications lose up to about 25% IIRC (through an anutomatic trans, u-joints, differential, etc.), some a little more some a little less.

Let's suppose a 3.73 geared truck produced 500 horsepower on the dyno on a particular day, and a 3.42 geared truck also produced 500 horsepower on that same day. Now, lets suppose that engine rpm wasn't measured/recorded during the dyno runs. What can we theorize about engine flywheel horsepower for these two trucks, assuming the same drivetrain loss? My guess is that the 3.42 geared truck has more power at the flywheel (about 9+% more).

Jim

[rjschoolcraft]

Originally posted by More Power:
My guess is that the 3.42 geared truck has more power at the flywheel (about 9+% more).

Jim Nope.

If you think about what is really happening here, the measurement occurs in the dynamometer below floor level. The torque is directly measured from the amount of electrical power required to resist, or absorb, the force (yes, force, because that is what is applied to the OD of the drum through traction from the tires) of the drive wheels of the vehicle. The true absorbed power is this torque multiplied by the rotational speed of the drum (divided by a units correction factor). That is the "power" the dynamometer knows about. As I said in the previous post, the wheel torque being generated to produce the tractive effort (force applied to the drum) is well over 5000 lb-ft. The numbers that are reported, though, are corrected to crank shaft speed. The only way to do that is to know the relationship between drum speed and engine speed. In this case, we must consider overall wheel size and final drive ratio. If the run were made in any other gear but direct, that ratio must be considered as well.

In summary, the two cases you mentioned have the same power, because it was measured and reported. A difference in performance will be felt because the tractive effort of the 3.42 ratio will be less than the 3.73 ratio.

In fact, after witnessing what Kennedy did to ensure engine speed measurement on my truck when he tested it last year, I'm quite sure that you cannot report meaningful numbers without measuring engine speed. If the computer knows the drum speed (which it does) and it knows engine speed, it can "automatically" correct for tire size and drive ratio without having them input directly.

[Kennedy]

I've queried Superflow to see what they have to say.

Note from me:

The Superflow will measure wheel power without tach signal, but will not display or calculate Q without it. Mustang somehow takes a wild guess, but it's so typically far off it's not even funny. Now I believe the Superflow "test group" may play into this as well. The test group has specifications entered like gear ratio etc.

[Kennedy]

Let's see how this goes. Bret is genrally reluctant to get into discussions that can get heated but here's a go:



Horsepower & torque are measured (computed) by a dyno as the truck spins the rollers. This is rear wheel power.

[(BEW)] Horsepower is never measured directly. It cannot be. It is a calculated value. Mr. Watts equation is thus: HP = Tq times RPM divided by 5252 (his constant)


What assumptions can we make about flywheel power for trucks with different gearing and tire sizes?

[(BEW)] None.


Would flywheel power be different for two trucks dynoing the same numbers, but having different differential gearing and tire size?

[(BEW)] Since flywheel power is not available unless it is measured at the flywheel, this question is moot.




RJ's post

The dynamometer measures torque directly. It knows how much reaction torque it is generating at any particular point in time. The power is computed by taking the measured torque and multiplying by speed (then dividing by the appropriate constant to get the units to work out right).

[(BEW)] On a chassis dyno, torque is not measured directly, especially not flywheel torque. Assuming the discussion is concerning the inertia power measurement, here

[rjschoolcraft]

Did not intend to suggest that flywheel torque is measured directly. It cannot be unless the engine is mounted in a test stand coupled directly to a dynamometer.

I know nothing of how the Superflow or Mustang dynamometers work specifically, but I do know how the large GE eddy current dynamometers worked that we used at Allison (and how the water brakes worked that we used). The large GE dynes (could be used as a motor, too, 500 HP each) were mounted in a cradle with either large rolling element bearings between the cradle and motor housing or the housing floated in a sealed oil bath. Reaction torque was measured by placing a load cell under a reaction arm attached to the motor housing. The load cell measured a force, which when multiplied by the radius from the motor centerline gives reacton torque.

John, your last post actually confirms much of what I was trying to say. BEW's comment about speculation, though, is just that - speculation. In my professional career, I've had opportunity to do a lot of comparison work between Caterpillar, Allison, TurboHydramatic, Ford and Dodge transmissions. I know for a fact that some are more efficient than others (read less loss). Allison's are much more efficient than Cat's. While at Cat, I did some work for them to show them how much so and what they needed to do to get better. The TurboHydramatic is more efficient than either the Dodge or the Ford (talking specifically truck transmissions here). It's not anecdotal and it's not speculation.

The Allison has more loss than the TurboHydramatic 4L80E. This I know for a fact. It is physically larger than the other brands and has more rotating internal mass than the others...one reason that it has more loss. It has more friction surfaces with oil shearing between them on the clutches that are not engaged. That means more loss. An allison 3560 has more loss than an Allison 1000. Simple physics.

Edited after figuring out what [(BEW)] meant.

[ 09-20-2005, 05:20 AM: Message edited by: ronniejoe ]

[madmatt]

I'don't know about JK's superflow but the last one I ran required axle ratio, tire size, lug type, number of drive axles, wheelbase, peak HP and torque rpm ranges and also required a tach signal as that determined what the dyno pulled the engine down to. we just entered the info, got in Direct drive, set the cruise and pressed a button and let the dyno do the rest. How Does your's run John? I can't remember the model number on that one, but it would handle up to multi-axle class 8s to 2000hp and 8000ft lbs and was just an absolute monster.

[madmatt]

Also, I haven't seen it mentioned but if it has been please disregard but I feel it should be noted that there are two completely different types of dynos. Load type and resistence type. It is my opinion and experience the the load type is the most accurate, even more so on diesel engines.

[Kennedy]

Bret has only part of the story here based on the first two posts that I fwd. He is reluctant to patricipate due to the typical outcome of many discussions in chat rooms/message boards. I reassured him that this was not your average board, but doubt he will participate.

[Rollingon]

Just a thought based on my passing experience with dyno testing. Ronniejoe certainly knows much more than I do. BEW clearly didn't want to say anything that might be questioned and I didn't find much of value in his comments.

Chassis dynos measure torque at the roller drum (or something geared to it) as has been mentioned. Rearwheel horsepower is simply computed by multiplying this measured torque by the drum (or other) RPM and dividing by the appropriate constant.

A torque can then be determined by dividing this HP by engine RPM, and multiplying by the same constant. This is engine torque, as would be measured at the flywheel, diminished by the drivetrain and roller losses. This, I believe, is what's commonly quoted as rearwheel torque. Knowledge of gearing and effective tire diameter is not necessary for this process, although they may be useful for other purposes.

I hope I said this right!

[Kennedy]

I just wanted to be clear that Bret did not view the thread as he has not registered here so I kinda gave him a quick note and he didn't quite get the jist of the entire thread. Bret is one of the lead Technical Support Engineers at Superflow. I've dealt with most of the guys at SF and they REALLY know their stuff when it comes to dyno testing and their products. Had this been an issue with my dyno there would have been more "said" in regards to the questions. What I will look at next is putting MP in contact with Bret so that he may have his questions answered.

[rjschoolcraft]

Rollingon put it pretty succinctly. As I said in one of my earlier posts, the dynamometer knows about power on the drum and Rollingon restated that (probably more clearly than I did). Kennedy's comment about the SF being able to report wheel power with no engine speed measurement (but not torque) supports this understanding.

MP's questions have been answered here, but if more understanding of the dynamometer's workings are required, then Kennedy's thought of putting MP and SF in direct contact is a good idea.

The bottom line is, you don't "lose" 9+% between a 3.73 and a 3.42. In fact, as I said previously, the difference in loss between those two configurations is probably not even measureable. And that statement is not speculation or anecdotal. It is based on nearly 20 years experience in the power transmission business.