keyway vs spline 3

[jayarrbee]

Having had a fatigue failure on a drive shaft with cracks starting at a keyway, I would like to evaluate a splined drive as an alternative. I'm struggling to find some form of comparison chart / guide for comparison of capabilities between keys and splines. Any pointers?

 

[Yorkman]

Fisher Conrtol Valves

http://www.emersonprocess.com

has a PDF file on control valves, they use a spline shaft on their rotary actuators. they may have some data that can help you.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int./JCI

 

[CoryPad]

A spline is a multiple keyway. Since you are spreading the torsional forces over multiple areas, you can decrease the induced stresses. This is shown in Machine Design texts like Shigley and Mischke.

Regards,

Cory

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

 

[ct140]

I would recommend a shrink fit without a key as we use succesfully since 5 years or so for the coupling hub on the drive shaft of reciprocating pumps (max. 1500 kW at about 50-60 rpm). Shaft is straight cylindrical, no key(s), no spline. Mounting is easy (just heat, usually about 240°C) and for removal the hub is provided with a few grooves and connections for hydraulic equipment. The hub is removed cold and without any damage to shaft or hub.

 

[GregLocock]

A key should not be designed to transmit torque, in general. They should be used for alignment. They can be used as mechanical fuses, but are rarely satisfactory when used like that.

Roughly speaking the torque capacity of a spline will be n times that of the same sized key, where n is the number of teeth. In practice since splines are often rolled, their torque capacity is even higher than that.



Cheers

Greg Locock

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

 

[Cockroach]

Take a peek at Machine Design Data Handbook, Lingaiah, ISBN 0-07-037933-5. I highly recommend this text as a strong reference book for mechanicals who choose to design.

Good luck.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[Tmoose]

Generating splines is expensive. Splines that provide accurate centering are even worse. If the components with the keyed connection are otherwise satisfactory then modest modifications to the geometry of the shaft and key seat may be all you need.

Like others said, torque transmission via key is often looking for trouble. A healthy dose of friction (diameters with interference or clamped faces or even setscrews radially loading the shft against the hub ID) is much better at day-t0-day chores.

 

[CSLufkin]

I like a tapered bore/shaft and a key, like a flywheel on a small engine(and many many other things). Very reliable, accurate, and strong

 

[Farmboy71]

You may be able to reduce the stress concentration of the milled keyslot by changing to a sled-runner design. I'm not sure of what kind of keyway you currently have, but this could be an improvement. This could also be used on the the spline desing as well.

For more detail check out Shigley-Mischke (5th Ed.) pg 368.

Farmboy

 

[eromlignod]

Are you cutting the shaft keyseat with an endmill? This is a common practice nowadays, but was once a no-no. Shaft keyseats cut on a horizontal mill (with a transverse-axis hob that cuts into the shaft like a tire into snow) don't have the stress concentrations that obround slots have.

I'm not sure what you're attaching to the shaft, but if it is a drive component have you considered QD or taper bushings?

Don
Kansas City

 

[Screwman]

Agree on switching the keyway to splines. The stress concentration factor will be much lower for a spline. And if you go with a rolled spine you will have the additional compressive stress to improve the fatigue life even more.

Trying to use a keyway in a fatigue loaded joint is the mission of the devil. They are going to break...and always at the worst possible time.

 

[MintJulep]

QD bushing.

http://www.martinsprocket.com/PDF/b-3.pdf#search="QD bushing"

 

[rob768]

Fatigue means alternating loads, and fatigue starting from a key usually relates to torsion. A spline will give some improvements, but means redesigning your connection at much higher costs. Start at the other end and see if you can lower the vibration amplitude by means of a torsional analysis and some simple engineering options (inertia disk, elastic coupling)

 

[Mandrake22]

QD and other taper lock bushings will still use a key. Providing a minimal radius at the corners of the keyway will help. This would require a clearance chamfer or radius on the key as well.
Here is an alternative to keys or shrink fits.

http://www.fennerdrives.com/keyless_bushings/trantorque_features_benefits.asp

 

[eromlignod]

QD and taper lock bushing have a key, but it is not used to transmit the torque. The bushing has a taper and acts as a collet when the flange screws are tightened.

Also, I have found from experience that Trantorq bushings are often a pain in the ass. It is very difficult to tighten them while controlling thier position on the shaft/housing. They have a tendency to change location as they are tightened, so it is difficult to use them with a drive component that must be accurately located along an axis. If position isn't critical, then they're OK.

Don
Kansas City

 

[macmil]

you may want to upgrade the shaft material; deoends what it is to start with; if you can do that it may save a lot of other changes, and work.

 

[rob768]

The effect of upgrading material is minimal if you have a fatigue problem. With you win in fatigue strength, you loose in notch sensitivity.