Jump to content

Approximate drivetrain power loss


Recommended Posts

Hi everyone, I recently dynoed my dirt bike and got 50wheel horsepower. I am curious to compare the results to what the manufacturer claims since I've made several upgrades. I have noticed that manufacturers claim 50 brake horsepower [bhp]output and that is something very different from wheel hp. [whp] Since I got whp readings I am curious to know the approximate drivetrain losses in percentage to calculate how much whp would a 50 bhp bone stock machine make. That way I can compare the results! I've heard that a bike in average losses ~ 15% of power through the drivetrain but don't know how well I can trust that information. Any idea? Thanks!

Link to comment
Share on other sites

It is usually 5% to 10% with a motorcucle, 15% to to 20% for a car.

You needed to keep in mind dyno differences, what smoothing as used, tires, tire pressure and a bunch of other things that can make up to a 15% difference too.

I have seen dynos report 40Hp on a bike that I know only made 33. I have seen reports of bikes making only 35 when I knew it was making more like 43. Dyno numbers should be used as comparison for before a modification and after, using the same dyno, same operator and similar air density. Otherwise, the numbers are easily dismissed as merely bragging sheets and little more.

From manufacturers, they use a much more acurate dyno that the typical bike shop, it is typically chain driven (not tire on a drum) and the conditions strictly controlled and done in a regulated method to provide fair numbers. So comparing a Kawasaki to a Yamaha in a sales brochure to fair.

  • Like 1
Link to comment
Share on other sites

Losses will vary considerably from one bike to another because of design differences. I don't have any hard number estimates for you, but for example, the most efficient drive train you're likely to see is one with either a Gilmer belt or oiled roller chain primary, and a Gilmer belt or roller chain final drive. Assuming the rear chain is in good condition, losses can be very low in such setups, maybe as low as 5%.

Next would be spur geared primaries, then helical primaries, then you start running into other odd variations. Any sort of right angle drive, such as at the rear of a shaft drive bike, or the output at the trans of a bike with a shaft final drive and a transverse crank, will be less efficient. Also, any offset drive unit as you might see used with an inline crank to step the drive shaft over will cost something.

Another issue is that most inertia dynos are rather undependable as strict measuring devices. The same bike can produce 8 different numbers on any four different dynos. Their best role is as a tool for comparing one unit to another in a relative sense, or in comparing the effect of changes made to any one unit. Your bike put up 50 on the one you tested it on on that day, but it's just as likely to run up 48 on another one and 52 on a third.

Manufactures claims of power output are not particularly usable for anything other than filling up space on a pamphlet, in my opinion.

Link to comment
Share on other sites

Ok it is clear to me that dyno is not trustworthy but it did show the approximate power. It's not like it will show 50 whp if the bike is only capable of 30. It shows the approximate so that you can estimate it's power. I actually did a whole bunch of mods and am curious about it. Even though dyno may lie, it couldn't lie that much could it? The dirt bikes are basically all the same and I was wondering if anyone knew the approximate power loss of a dirt bike drivetrain (clutch, transmission and chain). Could it be those 10% William1 claimed? I am just hoping that the mods I've made have changed my power curve more than the dyno can lie.

Link to comment
Share on other sites

I have an idea what you are trying to do and it just won't be truly accurate. There are formulas to get you close to what you want but not exact and not real world numbers. To find out exactly what you want to know in real world numbers you will need an engine dyno. and a chassis dyno. Compare both before and after mods.

I am the type of guy that loves to hear someone going on and on about how much hp their bike, car, or whatever makes on the dyno and then they ask me (because my whatever appears to be so much more powerful then theirs) have I had mine on a dyno and how much power does it make. I usually respond with a number about 20% lower than theirs just to see the look on their face. Then I remind them it's not about what that spec sheet says, it's what you do with it that matters. As far as I am concerned a dyno is nothing more than one of many tools we use to verify weather or not one is going in the direction he intended (in search of a different power curve).

Link to comment
Share on other sites

Ok it is clear to me that dyno is not trustworthy but it did show the approximate power. It's not like it will show 50 whp if the bike is only capable of 30. It shows the approximate so that you can estimate it's power. I actually did a whole bunch of mods and am curious about it. Even though dyno may lie, it couldn't lie that much could it? The dirt bikes are basically all the same and I was wondering if anyone knew the approximate power loss of a dirt bike drivetrain (clutch, transmission and chain). Could it be those 10% William1 claimed? I am just hoping that the mods I've made have changed my power curve more than the dyno can lie.

What you have to do is first ensure the dyno operator is competent and the dyno is in acceptable condition. Then you must do runs prior to and after to confirm the change you made improved things. Otherwise you are still guessing.

Dynos can lie over 30%

Link to comment
Share on other sites

Any "formulas" you might find as to power train losses would only be estimates unless actually measured. I remember a test of rear drive axles from muscle cars in the 70's. They ranged from 16 to 25 horsepower require to turn the axles 750 RPM under the same load.

Flywheel dynos don't "lie" exactly, they just have no way of being specifically precise, mostly because of the fact that they rely on the rear tire, which can not only slip, but it can become a variable by absorbing power and converting it to heat because of how severely the carcass has to flex when tied down on a roller. For another thing, the more power is produced, the more it will vary from one day to the next just based on weather conditions.

Then there is the whole way that they work. The principal is base on the fact that torque is what causes acceleration, and the acceleration the bike produces is graphed, and the torque required to produce that acceleration against the known load is then used to figure out how much power was produced. This method is much faster, and much easier on the bike than the old method of using a hydraulic brake to find the max torque produced at different speeds, but not quite as accurate in precise terms.

What kind of bike are we talking about in the original post?

Link to comment
Share on other sites

Another issue is that most inertia dynos are rather undependable as strict measuring devices. The same bike can produce 8 different numbers on any four different dynos. Their best role is as a tool for comparing one unit to another in a relative sense, or in comparing the effect of changes made to any one unit. Your bike put up 50 on the one you tested it on on that day, but it's just as likely to run up 48 on another one and 52 on a third.

sorry for the thread hijack but it's directly related to thought above.

assuming the dyno it repeatably accurate (in the same day)

Using 2 different bikes. one as the actual test subject and one as the control, you should be able to calculate out any change in variables from a later session using the change in the control bike as the correction factor correct?

Link to comment
Share on other sites

Using 2 different bikes. one as the actual test subject and one as the control, you should be able to calculate out any change in variables from a later session using the change in the control bike as the correction factor correct?

Yes. That still doesn't allow a comparison between two different bikes on different dynos unless the the control bike is also used on the second dyno on the day of that test.

We're probably talking about differences of 4% HP +/- in most cases, but it could be greater.

Link to comment
Share on other sites

I understand what you are all talking about very well. And I don't wish to argue with you about it. Though the readings I seek have no need of being accurate. As I said this is just a "I wonder..." quest an not something that's obligatory.

I believe the dyno was fine and the results could vary of +/- 4 to 5 hp..

manufacturer states 55 bhp @ 9k RPM. We got 50 whp @ 10k. We could as well gotten 46 or 55. But it would still make changes for us to notice it. I just need an approximate percentage of drivetrain power loss to see what came out. I dont need an accurate reading I am simply curious :eek:

Link to comment
Share on other sites

I understand what you are all talking about very well. And I don't wish to argue with you about it. Though the readings I seek have no need of being accurate. As I said this is just a "I wonder..." quest an not something that's obligatory.

I believe the dyno was fine and the results could vary of +/- 4 to 5 hp..

manufacturer states 55 bhp @ 9k RPM. We got 50 whp @ 10k. We could as well gotten 46 or 55. But it would still make changes for us to notice it. I just need an approximate percentage of drivetrain power loss to see what came out. I dont need an accurate reading I am simply curious :eek:

Come on now... be serious, this type of test is not worth doing if you don't want to be as accurate as possible. You either need to find somebody who has done this before or you be the first. It sounds to me like you have access to a chassis dyno, now all you need is access to a flywheel dyno and a stock control engine of similar design. (another dirt bike)

Link to comment
Share on other sites

Wouldn't it be easier to find someone who has dynoed a stock bike exactly like yours on a chassis dyno, rather than finding someone who has calculated the losses of your specific bike's drivetrain?
Well, to start with, you'd have a 4 to 5% margin of error (or more) just if two dynos were used. Plus all the other variables already mentioned.

What would have been the best way to tell would have been to have dyno'd his bike on the same machine pre and post-mod.

Link to comment
Share on other sites

  • 3 weeks later...

This would be far from actual testing but very interesting .When the first honda 450 was released honda claimed 54 crank hp .I measured the entire engine port volumes flow and port lengths and frictions based on bearings and piston skirt areas .The simulation software showed basicly the 54 hp .When the bike was actually dynoed it produced around 47-48 hp @ the rear wheel .I have always used a constant of 14% loss to roughly calculate crank hp based on the simulation and information from my dyno.Its all just numbers .I have also switched my dyno jet over to a diferent system that can be adjusted i chose an moment of inertia that dyno jet claimed that there drum produced and kept changing it untill the graphs produced the numbers i was acustom to seeing on my race bikes it reads hiher than some and lower than others .my dyno has become a yard stick for dirt track bikes I have seen graphs producing as much as 70 hp from a dyno jet and the bikes that produced that power would get severly beat by my 62 hp bikes .I never get up set when some one tells me there build made more power than mine and then they tell me the list of parts they put in to acheive that number .The dyno is just a comparison tool you need a base line from bikes that win and go from there.

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Reply with:

×
×
  • Create New...