Stress in hub due to press/shrink fit or keyless locking assembly

[ewans]

Hello all,
I'm looking at the stresses generated in a hub due to press fitting on a shaft or due to an internal keyless locking assembly (such as from Ringfeder, B-Loc, Bikon etc).

I'm wondering if exceeding the yield strength of the hub material at the contact interface (i.e. hub ID/shaft OD) should be avoided at all costs, or if it is acceptable (and perhaps routine) due to the surrounding hub material which remains below the yield stress.

Thanks in advance for your input.

Regards,
Ewan

 

[GregLocock]

Depends on whether you want the part to be a service item, and whether your hub material can be used reliably post yield.

FWIW the hub unit in a particular front drive car is a press fit in the cast iron spindle, and we are careful NOT to exceed yield.







Cheers

Greg Locock


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

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

 

[RolMec]

You mention BIKON, there's this catalogue about shaft - hub connections (hydropress), maybe it's online too.
They remain at least 25% below yield stress at shafts, the surf. pressure on shaft being more critical. The diameter of the hub being set to min. / standard values, they remain approx. 40% below yield.
But this system (shaft-hub pressure fit) has to be carefully engineered. There's a lot more to it than only a stress limit at the fit. Check also for bending that maybe has to be transmitted over the hub.
Regards
Roland

 

[geesamand]

If you yield locally, you lose contact pressure and that directly affects the rating of the coupling. I've not studied this but it would seem that as you tension the coupling and yield the hub, you'd not be able to achieve proper fastener torque in the locking hub.

Do you require the full torque / thrust rating of the locking assembly? If not, consider reducing the assembly torque of the locking assembly fasteners per the manufacturer's recommendations. Another option in some sizes is an alternate design such as the B-Loc Trantorque couplings.

David

 

[rb1957]

i think you can exceed yield stress by having a large interference (you'd probably have to heat the hub and freeze the axle) and you'd have to be very careful with the design. i guess you'd only do this in order to transfer a lot of torque. if either member is cast, then i'd suggest not exceeding yield.

 

[ewans]

Thanks to all for your replies... appreciated.

I presumed that you would want to keep all hub material, including directly around the contact interface, below the yield stress. The problem I'm having is reconciling the hub pressure and minimum hub diameter values from the Ringfeder/B-Loc catalogs with the stress results I'm getting from FEA and various press-fit calculators.

When I run a quick FEA using the hub pressure and minimum hub diameter values from the catalogs, the resultant Von Mises stress in the hub is above the material yield strength at the shaft/hub interface. The FEA results are consistent with results I get from my own, and several online calculators for press fit stresses. As expected, as you move radially outward from the interface, the stress drops to below yield. This result made me question whether or not it's okay to have local yielding at the interface.

Regards,
Ewan

 

[rb1957]

you were running a material non-linear FE ?

 

[ewans]

Hi rb1957,
I'm an FEA novice, but I'm pretty sure it was a linear FE analysis (I'm using basic Pro/Mechanica). I applied a uniform pressure load to the hub ID (using the value from the mfg catalog for hub pressure), and constrained rotation and axial displacement of the hub ID.

I got good correlation between the Mechanica results and the results from a couple of press fit stress calculators. Using the calculators, I adjusted the radial interference to achieve the same interface pressure as listed in the catalog. The Von Mises stresses using both techniques were approximately the same, and both were above the 45ksi yield strength of the suggested hub material.


Regards,
Ewan

 

[hydtools]

Reaching yield will only limit the press fit pressure to that which causes the material yield. It will not result in a loss, only a limit. When you take the joint apart you will have less interference than before the original assembly because of the yielding.

Ted