Does anyone know if a U-Joint could be used for vertical loading?

[MechEng1235]

I need to drive a motor rotor down into a stator using a linear actuator but I have yet to get a good design for holding the rotor. I was thinking of having a lever arm extend out from the actuator and connecting a u-joint by tapping both ends of the joint. I need the rotor to have some play to push down into the rotor which is why I was going to use a u-joint but they aren't rated for vertical loading since that isn't what they were designed for. I'm only going to be lifting 200 to 300 lbs.

 

[MintJulep]

Generically you are looking for a "gimbal". However I think a self-aligning bearing might work for you.

 

[MikeHalloran]

You can work backwards from the torque rating and the joint geometry to get a thrust load rating for what you propose. Don't expect to exercise any warranty.



Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

I assume you mean an automotive style Hookes joint? Yes, they are used for dangling shafts in some applications I've seen. Basically the part of interest is a needle roller and it is easy to work out an axial thrust equivalent to the torque rating for the joint.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Tmoose]

Like Fig 7 here ? -

http://www.web-cars.com/corvette/1963_SAE-paper.php?page=5

 

[MechEng1235]

Tmoose, the ujoint and rotor will both be hanging vertically 9 inches away from the vertically mounted actuator.

 

[MikeHalloran]

I worked on a flight simulator comprising a Stewart platform that used truck u-joints at each end of six long-stroke hydraulic actuators to suspend and move the simulated aircraft.

The u-joints had two potential problems in that application.

If the machine ran for a long time without moving much, the needle bearings would be subject to false brinelling. Since the installation was for an attack aircraft, it probably moved around enough to keep the needles lubricated.

The 'ears' on each yoke were not designed to sustain large radial loads as from an unbalanced driveshaft. The yokes attached to the cylinders were okay because the load stayed axial-ish, but the yokes attached to the moving platform and to terra firma deflected uncomfortably much from the load trying to spread them while the platform was articulated far away from its neutral position.

In your case, radial loads on the yoke ears are not a problem, and occasional unloaded large articulation of the yokes should keep the needles lubricated.



Mike Halloran
Pembroke Pines, FL, USA

 

[tbuelna]

MechEng1235-

Can you provide more details of your proposed assembly fixture? You mention that some "play" is needed in the mechanism for positioning the rotor. What type of "play" is needed? A single u-joint will only provide angular displacement about the apex of the u-joint axes. It will only be possible to tilt the rotor about the single u-joint's 2 orthogonal axes. If you need to make small adjustments to align the rotor axis parallel/concentric to the stator axis at installation, then a single u-joint does not provide the necessary DoF's.

It should be no problem for a u-joint from a 3/4 ton P/U driveshaft to handle a 200-300 lb axial load. However, make sure you take into account the angular displacement limits of the u-joint.

 

[MechEng1235]

tbuelina

"Can you provide more details of your proposed assembly fixture? You mention that some "play" is needed in the mechanism for positioning the rotor. What type of "play" is needed? A single u-joint will only provide angular displacement about the apex of the u-joint axes. It will only be possible to tilt the rotor about the single u-joint's 2 orthogonal axes. If you need to make small adjustments to align the rotor axis parallel/concentric to the stator axis at installation, then a single u-joint does not provide the necessary DoF's."

Yes, alignment of the rotor to the stator is my ultimate goal. There is a .020" gap between the two and the closest I can get to concentric, the better because the magnets will favor one side until I can lower the rotor down onto its pilot. I was going to use a double u-joint to give me the movement I need in the u-joint's 2 orthogonal axes.

 

[tbuelna]

MechEng1235-

I've seen similar fixtures for installing PM rotors into stators, but the PM rotors were way smaller. I was told that it was critical to ensure the PM rotor did not contact the stator ID during installation. The fixtures used a rigid set of linear bearings and a dummy shaft to guide the PM rotor into the stator bore.

I can only imagine how difficult it would be to install a 200lb PM rotor into a stator bore with only .020" clearance. It would seem to me that you would need to guide both ends of the rotor during the process. Hanging a 200lb PM rotor from a u-joint at one end would not seem to provide sufficient positional control to prevent the rotor from moving less than .020".

 

[MikeHalloran]

So, a truck u-joint can carry the load. You need two to support it and allow you to, by other means, guide the rotor into the stator.

If what you are designing is the means of alignment and also the support, then Terry is right. You need a rigid fixture, not a flexible one.

If the linear actuator itself is not stiff enough to constrain the rotor's trajectory, then you need a stiff linear bearing and rail, piloted into the motor shell or otherwise rigidly located relative to the stator. The traveler then need be rigidly coupled to the rotor, and flexibly coupled to your linear actuator.




Mike Halloran
Pembroke Pines, FL, USA

 

[MechEng1235]

Mike, I am definitely on board with the rigid fixture idea but now I'm not certain how I could align the rotor to the stator and have it be repeatable. With only .020" to work with, it would be hard to use a guide (thin wall tube?) and I'm worried about the tolerance stack from the housing location and the deflection in the carriage.

 

[MikeHalloran]

For a guide, a sheet of mylar or polyethylene or polypropylene wrapped into a cylinder around the rotor would work just fine.
... but then you have to remove it after assembling at least one of the end bells. It may or may not be reusable after extraction.

Maybe some perforated sheet could be stretched over the rotor and left in place, to be shredded on first start. With my luck, the remains would get balled up and do major damage within the warranty period, but maybe you can do better. Maybe a mesh tube, e.g. from Caplugs, could be made thin enough to remove or frangible enough to leave and shred, or soluble in something you could immerse the assembly in, e.g. PVA and alcohol.

If the tolerance stack from stator ID to shell is looser than the air gap, maybe a rigid fixture won't work. I hesitate to recommend something like a self-adjusting rigid fixture, e.g. a robot that uses a laser or a probe to locate the stator ID, then guides the rotor along its axis, well, it can be done, but that kind of stuff gets expensive fast.



Mike Halloran
Pembroke Pines, FL, USA

 

[Tmoose]

Hi MechEng1235,

after I posted a link to Corvette irs you said " Tmoose, the ujoint and rotor will both be hanging vertically 9 inches away from the vertically mounted actuator."

I interpreted your description to mean the force was "into" the u-joint, similar to the force shown in Figure 7, but with opposite sign. As if Fig 7 were rotated 90 degrees.
Did I miss something?

 

[MechEng1235]

Tmoose, you are correct. The force is putting the u-joint in tension and will be oriented vertically but there should not be any side force since the u-joint and rotor will be concentric.