Gearbox Efficiency 2

[FredGarvin]

Has anyone had any experience in testing, under controled conditions, gearboxes for overall efficiency and losses? I'm stepping into this particular testing area (I have been testing for almost 15 years now) for the first time and would like some theory behind the tests (I haven't been able to find much in my literature or the internet).

I know the premise for efficiency...the "goes-outta/goes-inta" but I'd like to get some insight into parameters measured and general idea behind quantifying losses.

Thanks.

 

[GregLocock]

SAE paper 2002-01-2821 Fuel economy improvement through frictional loss reduction in light-duty truck rear axle

Author(s): Gangopadhyay, Arup, Asaro, Sam, Schroder, Michael, Jensen, Ron, Sorab, Jagadish,


Has some good details.


Cheers

Greg Locock

 

[EnglishMuffin]

I have some experience testing large machine tool gearboxes, measuring losses etc. Most of the losses end up as heat of course. The heat is generated in bearings and gear meshes, and there can also be churning losses and windage. Typically, one would instrument the major potential hot spots with temperature sensors. Non contact thermal imaging can be valuable to get an idea of where the hot spots are. In addition to measuring the overall efficiency with a dynamometer, if your gearbox has a recirculating cooling oil system you can do some checking of the heat balance for that. You (or someone) can build a thermal FEA model, and use some of the external experimental data to get an idea of what is going on inside (I've done that too), but it's not a project for the faint of heart, or those with little time available.

 

[FredGarvin]

I was thinking along the lines of doing a heat balance, but wasn't quite sure if that was usually looked at.

Thanks.

 

[EnglishMuffin]

I would think it all depends on why you are doing the testing. If it's just to see if the boxes are in spec (whatever that may be) then there might not be much point, although maybe people do it in specific cases. You don't say what industry you are from or what kind of gearboxes they are - you probably need to see what other people do with similar products, and I can't help much with that. However, if you are doing it for R & D purposes, improvemnts etc, then the sky's the limit.

 

[GregLocock]

CV joints are 99% efficient. The following describes a heat balance test.

SAE paper 1999-01-0936

Cheers

Greg Locock

 

[FredGarvin]

As some background (in a nutshell)...

This is a brand new custom box we have designed for our test cell work. It is simply a speed multiplier (or reducer)since most of our work is done at very high speeds. The unfortunate part is that I have a round of testing coming up very soon that will require that we know what we are losing thru it. Unfortunately we have no data on this GB except theoreticals.

Thanks for the responses.

 

[EnglishMuffin]

Well, if you can, I would just dyno it and see how much you are losing directly. But if you don't have a dynamometer that can handle the speed, then it gets interesting.

 

[FredGarvin]

We have a dyno, but I was thinking of what would be supporting analysis along with that. I think a complete energy balance will be in order for this particular round of testing.

Thanks!

 

[EnglishMuffin]

Well, it depends on what you mean by "complete". If you have the type of box which is equipped with a circulating oil system and some type of external cooler, you can make some measurements on that. But you will still have convection and radiation losses, which are difficult to measure directly. Of course, you can calculate what the total thermal losses are from the dyno measurements. But to get anything useful beyond that, you may have to combine your measurements with some analysis. This is just a repeat of what I said before, of course.

 

[ccw]

Hi Fred,
What about energy conservation: (steady state points over speed range of interest)
[Torque out x speed out] / [Torque in x speed in] x 100 = efficiency. [Torque in x speed in] - [Torque out x speed out] = heat, noise, etc. For 100 per cent efficient gear box the first equation = 100, the second = 0.

I am not sure why you need to compute the losses independent of the torque/speed functions, unless you need to troubleshoot problem areas, such as specific bearings and gear sets within the gear box.

 

[FredGarvin]

Hey guys. Thanks for the replies.

My main concern is that this is a brand new design that has not had a chance to have been tested and I am going to have to throw it into the mix without any real knowledge on the box itself seperate from the actual test device it will be driving. I will have to look at 2 test articles instead of 1 which is not the way I prefer to run tests if I can help it. I definitely will be having the torque and speed measurements pre and post of the box like everyone has mentioned. My main concern was whether anyone who has done this sort of thing before worried about doing things such as a heat/energy balance to characterize their application or if they did it based solely on torque and speed.

In re. to what EM has said, I agree with you that there definitely will be aspects to the set up that I simply can not capture, i.e. convective and radiation sources. This GB will be right next to a gas turbine, so it's definitely not going to be an adiabatic situation. I will though be measuring temps in a lot of places to get a good thermal picture of what is happening in the GB.

Thanks for the input guys!