Universal Joint setup and vibration

[Petri]

When setting up universal ( Cardan ) joints on a large high speed shaft is there a preferred orientation of the yokes on either end of the shaft. Should they be in-phase or 90 deg out of phase. Does it make a difference?
I have a problem in that I am experiencing a horrible 1X shaft speed knock in a drive train with two of these shafts in a drive system. The vibration is always there during low speed and light load but when the drive speeds up and load increases the vibration is not evident. Any suggestions ?

 

[GregLocock]

Yes, the orientation is crucial. For the simplest case, where the input and the output shafts are parallel, then the joints should be in phase. For the other simple case, where the input and output shafts are at the same angle to the intermediate shaft, and all lie in the same plane, then the joints should be 90 degrees out of phase.

Any other case needs some calculation to determine the optimum relationship. It's a moderately straightforward bit of 3d geometry.

Phasing of the cardans cannot cause a 1 per rev vibration. It causes even harmonics, starting with 2nd order. You are describing a very typical take-off shudder problem. As you increase the load your axle or engine moves, and changes the relative angles in the system, which causes better cancellation. Alignment errors as small as 0.3 degrees can cause this to happen (typically a diff can move by 5 degrees in total, due to the applied torque). The problem is speed dependent because the UJ forces are exciting a resonance in the driveline, typically the centre bearing of the driveshaft, or even a bending frequency of the driveshaft itself.

Cheers

Greg Locock

 

[Petri]

Yes this makes sense to me. I should have mentioned also that the dive train consists of several sets of hypoid gears. Having said that, your comment about alignment is crucial. A summary of the drive train is;
DC Motor - gear coupling - hypoid gear shaft continues through to next gearbox via another gear coupling. Inside the next gearbox is another hypoid gear and the common shaft again continues onto the next gearbox which is a planetary type with speed adjustment via the planetary arrangement. Between the 2nd hypoid gearbox and the planetary is one of the connecting shafts in question with the uni-joints. The shafts are supposed to be parallel and the angle of shaft is about 3 deg.
The output of the planetary is connected to a second connecting shaft ( again uni-jointed ) to a third hypoid gear box and the third box is connected to a forth hypoid via a gear coupling.
I am thinking now that the combination of clearances within all this drive system is accumulating to a larger that desired value. Combine this with the shaft angles possibly varying the system is showing a backlash effect amplified by the motor responding at load load conditions.
I had originally been thinking that the cardan shafts were showing an inertial vibration. They are quite heavy and about 5 years has passed since last replaced. The previous uni-joints failed due to lubrication after only 5 years. A learned comment in “The simplified handbook of vibration analysis – A Crawford”, page 186 described the condition I am experiencing exactly. The author suggested having the shaft yokes at 90 deg rather than in line.
Thanks for your comment.

 

[Petri]

My last posting requires a correction. my statement about putting the yokes at 90 deg is incorrect. I had assumed incorrectly.
( when you ASSUME , you make a ASS out of U and ME ).
I have looked into the phasing of yokes a bit more and found that subtle changes on the yoke phasing can be made to adjust for shaft misalignments.
Do not place yokes at 90 deg - not recommended !
Thank you.

 

[shabba]

I ASSUME you have sorted this problem by now, but will put in my 2penneth anyway.

The *only* time you should have to adjust the phasing of your cardan joints is if the input and output shafts do not lie in the same plane. If this is the case, then the degree of phasing is the same as the angle between the output shaft plane and the input shaft plane. In any case, the input and output shafts should always be the same angle to the intermediate shaft.

However, as Greg pointed out, if you are designing a powertrain where there is movement between the input and output elements. You should set up your shafts for full load conditions (ie take into account torque reactions).

Hope you did sort it...

 

[Petri]

Shabba
Thank you.
Yes the vibration problem was solved. The root cause was due to a faulty feedback resolver on the main dive motor. The F/B signal was causing an unstable control condition at low speed and light load.

 

[thirumalai]

I want to know the (formula) torque transmitting capacity of a universal joint. If possible th steps involved also. Thank you in advance.

 

[mloew]

Additional information was discussed in the thread Compound U-Joint Angles (thread78-53988).

Best regards,

Matthew Ian Loew
"Luck is the residue of design."
Branch Rickey

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