Shear stress from Ultimate or Yield? 2

[awol]

I have two somewhat conflicting engineering references defining the shear stress of steel. One reference shows shear stress =.40 x tensile yield stress and another shows shear stress = .50 x tensile ultimate stress. One is of course much higher than the other. Are they both correct given certain criteria? One is more conservative but I'm after the most common approach.
Application is the shear of a key between a shaft and a hub.

Thanks,

Alan

 

[SnowCrash]

Seems to me neither of these is a 'definition' of shear stress, just a rule of thumb. They also imply that ultimate stress might be 125% of yield. The actual values would be determined by testing.

Crash "your mileage may vary' Johnson

 

[SAP]

As a reference for Key design, I suggest you have a look at AGMA 6001-D97. This lists the allowable compressive stress as Sca = 0.7 x Sy (ie 70% of yield). The allowable shear stress is then given as Ssa = 0.5 x Sca (ie 50% of allowable compressive stress).

SAP