What is the power/torque lost through a u-joint?

[Strider17]

Gentlemen:

I have been through 20+ of my engineering books and a web search and I cannot find anything regarding torque or
power loss through a u-joint. Also, I am trying to
calculate the radial load for a bearing adjacent to
the u-joint. Can anyone direct me to a source of info
or provide info as well as the basis?

Thanks...


The more you know, the more you
know you don't know....

 

[GregLocock]

A) Not very much. You might be able to estimate it by looking at the velocities of the needle rollers and the oil film etc.

B) Free body diagram.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[DesignerMike]

I believe it would depend on the angle of the shafts.
Per Greg: Free body diagram.

Straight shafts would be very close to zero.
Severe angle would have significant loss while stressing the joint.

Bearing load would have similar answer.

 

[JStephen]

"Severe angle would have significant loss while stressing the joint."

If I remember correctly, if you have a single U-joint connecting two shafts at an angle, then the angular velocities of one or both vary as they go through a revolution. That could throw in more force than friction, as well as vibration.

 

[tygerdawg]

U-Joint manufacturers "SHOULD" have an Applications Engineering Department (or similar function) that has access to engieering data like that. I would expect to get an answer of "our model XX u-joint system is qualified to be XX% efficient in motion transfer" or something of that nature.

TygerDawg

 

[handleman]

JStephen's comment on velocity variation is right-on. That's one reason for the more-complex CV (Constant Velocity) joints instead of plain U-joints on FWD vehicles.

 

[JJPellin]

It is also noteworthy that you cannot control these losses by improved alignment. U-joint drives actually require a certain amount of misalignment in order to function reliably. The U-joint drives on some of our diesel engine driven cooling tower pumps require a minimum offset misalignment in order to get proper lubrication to the needle bearings. If you align them too well, they will fail. Tygerdawg is correct. The U-joint manufacturer needs to provide this information.

 

[Fabrico]

A minimum of 1/2 degree is recommended by most for lubrication and sealing.

I'm not so sure about manufacturers supplying power consumption information. They have not exactly been in competition to lower the values, and they very likely don't want to be.

As mentioned, it's affected by angle, bearing friction, speed, drive arrangement, etc.

 

[mayt4u]

Hi, I believe "U" joints have a lead and lag of 30 degrees from power input and torque for a grand total of 60 degrees per rotation. The older 4wd vehicles used this type of drive and were difficult to control thus speed was recommened below 50kph. Phasing of the joints is very important for this reason also. Have a look at a rear wheel driveline in an automobile for orientation. The unblanced mass of these joints can make them unsuitable for higher rotational speeds.

The power loss can be as high as 30%-50% depending on the cofiguration.

I believe the minimum offset per joint is 5-15 degrees.

Have a look at a drawing of a rear driveline of any auto and you should find the bearing and seal info.

Good luck on the calculations, I think its just the matter of building it and testing.

Cheers

 

[GregLocock]

Can you provide a reference for any of those claims? They are completely out of line with my experience.

Recommended offset for a UJ is 4 degrees

Old 4wds like say Landrovers or Jeeps had no speed restriction

The power does not lag

The unbalanced mass of the UJ is not responsible for high rotational speeds.

Other than that, yes, great post.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JStephen]

Note that on a vehicle, it is normally set up so the drive shaft rotates much faster than the wheels- not hardly what you'd expect if there was some sort of mass-vibration problem.

 

[rmw]

I used to be involved with Heavy Duty Truck drivelines and the rule of thumb I used was 5% power loss (friction heat of the roller bearings mostly) through the u-joints. However, I could not find any of the references that I used back then (25 years ago) to substantiate that number. Also, I can't remember if it was 5% overall (drivelines could have more than 2 u-joints) or if that was for each one. I think it was for the driveline as a whole, thought.

The number was probably on the high side, too.

rmw

 

[Strider17]

Gentlemen:

Thank you all for your valuable input. For those still
interested I found a source who has an empirically
based formula.

To= Torque Out
Ti= Torque in
a= Operating angle of U-Joint


To= Ti*cos^2 a

Also suggested for reference was an SAE publication
" U-Joint and Driveshaft Design Manual".

Thanks again to all...

Terry


The more you know, the more you
know you don't know....

 

[GregLocock]

That's not empirically based, and it does not answer your question.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[mayt4u]

Hi, the base of the information is from a third year millwright course. (Alberta Apprenticeship Program)

I had on old International Travelall 4x4 locking hubs and on the dash a warning sign was posted, "Do not to exceed 30 mph (50kph) in 4 wheel drive". I did it once and almost shook my poor, dead and buried, grandfathers teeth out.

The mass vibration problem comes into play when the rpm exceeds >1500 rpm. Wheels of autos and drivelines rarely exceed this, as 1000 rpm is ~ 132kph. The phase orientation is important (2uj) to cancel the lead and lag of the individual joints. Dynamic balancing is done on drive shafts prior to installation in two planes.

Have you ever tryed to turn the chute on a snow blower and find it sticks on part of the revolution, this is part of the lead/lag problem.

Based on Strider17's equation 10 degree tir would only amount to a 3% loss, however if begin approach 45 degrees you get a 50% loss and 90 degree is a 100% loss. This can be proved using a uj on a ratchet assembly.

Theres an idea, torque a bolt to say 50lbf straight up, then apply a certain angle to a uj c/w torque wrench and see how many lbf it takes to rotate the bolt further. Might help to prove the formula.

Bearings and seals take a kicking, if you try to over build the thing, bigger in this case is not better.

Beer time

Cheers