Spline failure on hydraulic winch 5

[jlm033]

I'm trying to determine what went wrong with a custom hydraulic winch we had in service.

The winch drum is connected to the frame by an 80x74 DIN 5482, 18NiCrMo5 annealed, case hardened shaft on the dead end, supported by a spherical roller bearing. There is an internally splined bushing pressed into the bearing to accept the splined shaft. Shaft and bushing are held from linear movement by external snap rings on each and a cover plate on the end of the bearing hub. The driven end is supported by the output shaft of a right angle gearbox, also 80x74 DIN 5482. The drum has an internally splined wheel hub (39NiCrMo3/42CrMo4,tempered) welded into each end to accept the splined shafts.

The winch was in service for about 5 years with about 500 hours of run time. It's rated for 12,000 lbs but we measured 18,000 lbs of line pull, so we know it's operating beyond design. However, it still seems like the drum should've been supported some other way than on the two splined shafts. The shaft on the driven end is not lubricated, and the internal splines stripped out of the wheel hub. The shaft on the dead end is greased, presumably only for the bearing. The splines in the floating end of the drum are badly worn, and it looks like there was significant drum wobble before the driven splines stripped.

There's a link to pictures of both sides of the drum and shafts. Any insight you guys can give is greatly appreciated.

winch photos

 

[gearcutter]

An incredibly poor design, to expect the splines to both transfer torque and to also have to support all the radial load.
I've never seen a winch drum supported by a pair of splined shafts-alone.
Usually the drum is directly mounted on bearings of some kind and does not rely on the driving shaft for support.
The photos show splines that have undergone high radial loads which has eventually led to a gross misalignment situation. Continued operation under these conditions has resulted in tip/root contact along with non-uniform wear along the teeth flanks. The male spline on the 'dead side' displays a classic 'dog bone' or 'barrel' shape usually associated with misalignment.
Looks to me like a major re-design needs to be undertaken as this problem will continue to occur.
Great photos by the way.

 

[jlm033]

Thanks for confirming my suspicions. Unfortunately, we have six more of these in service that we are trying to find replacements for. We can't provide forced lubrication to try to extend the life of the driven end, but we could pull the gearbox off and grease the splines to try to 'band aid' the drive until we get replacements in. I know this is not ideal (or even recommended really), so I'm open to suggestions. I'd also like anyone's thoughts on the fact that the manufacturer welded in the wheel hub. I don't know if they did any heat treating after welding, but I suspect the welding helped accelerate the wear in the hub.

Gearcutter, I gave you a star for your post. I really appreciate all the information you gave me.

 

[Tmoose]

https://www.youtube.com/watch?v=ifBABVClpZY

 

[tbuelna]

I would agree with gearcutter's comment about not imposing radial loads on shaft spline connections. Shaft spline connections should only be used to transfer torsional loads. This winch design should have used a bearing system to support the drum that isolated the shaft spline connection from any radial loads. And the shaft spline connection should have been designed to accommodate any small misalignment at the spline tooth contacts that might occur during operation.

 

[3DDave]

If the splines aren't yet worn, try shaft bonding compound. If it fails, you'll be able to take it apart. If it doesn't you won't have to take it apart. Just keep the input gearbox lubed so it doesn't become a problem too soon.

If not, then I'd experiment with putting some oil in the drum and setting up some felt to wick the oil to the splines. You can supply a lot more oil that way than a little bit of grease can supply without also introducing the build-up of thickener that occurs as the oil in the grease is depleted.

 

[gearcutter]

Does the failure of one of these winches have the potential to cause injury or loss of life?
If so; then you need to engage the services of a qualified mechanical engineer with the necessary experience............as a priority!
They will properly evaluate the situation and then advise you accordingly.

 

[jlm033]

They aren't in a position to cause injury. 3DDave, we will look into your suggestions to buy us a little time. We are trying to get replacements as soon as possible.

Many thanks to all for the replies.

 

[RolMec]

Are there any operational requirements as e.g. easy disassembly, max. weight of single parts or else? This spline shaft design might have been done for some reason.
Then, with a line pull of 18k when 12k lbs was the rated capac., how did the wire rope look like? If the rope was rated 12k lbs as well, I would not consider this mechanism noncritical to possible injury... even if the winch was safe.
Pls. be sure to redesign the complete appliance to the required rating.
Regards

RSVP

 

[jlm033]

The wire rope is ok and operating with a 3.5:1 safety factor at the 18k load. Regardless, we are reevaluating the complete winch system (shackles, sheaves, etc.).

 

[RolMec]

Hello jlm,
you pls. be careful when re-evaluating, checking what is and what you want...
.. as design standard:
Just take e.g. FEM II for a general base & the current state of things, with 500 h you are in utilz. class T2 and if the 18k is the usual load spectrum class is L4, which runs together into a mech. class M4. Minimum practical safety factor to be 4.
Regards

RSVP

 

[moon161]

In what ways might you do the same job while reducing the load the winch sees to 12,000 lb or less?