1/4" dog head set screw shear calculation

[Reifleman2]

I am supporting a failure analysis on a set of 4 dog head set screws that failed on a mechanical seal. I am trying to determine the minimum point at which shear failure could have occurred based on the materials, cross section etc. Here is the write up I have performed thus far, do these calculations appear correct? It seems to be a straightforward problem, I just want to make sure my solution is correct before forwarding on to the customer.

Rough shear strength estimation for setscrew, part 26 on drawing:


Ultimate Shear Strength = .57 x Ultimate Tensile Strength, engineering forums recommend using .50 x Ultimate Tensile Strength.

316 SS Ultimate Tensile Strength from

http://www.onlinemetals.com:

89900 psi

Ultimate Shear Strength = 89900 x .50 = 44950 psi

Setscrew diameter: 1/4 (drawing dimensions: .250 - .210, .156 square tip)

Height of holding area: .089 (sleeve)

Cross section: .156 x .089 = 0.013884 in^2

Equation: (Failure Load)/(Area) = Ultimate Shear Strength

Failure Load = Ultimate Shear Strength x (Area)

Failure Load = 44950 psi x (.013884 in^2 x 4)

= 2496.3432 lbs of force


Thanks

 

[hydtools]

Are you estimating shear of the dog point?
Shear area of the dog point is pi*d^2/4 = 0.0191 in^2.

Ted

 

[Reifleman2]

Shear of the dog point, correct. Would it not simply be the largest surface area across which the shear occurred? In this case, the sleeve thickness (and therefore the height of dog point affected) is the 0.089". The diameter of the setscrew (where it sheared, the dog head) is 0.156.

Is it not a simple cross-sectional calculation, W x H (with width = diameter)? Maybe I'm looking at this all wrong.

 

[hydtools]

Simple crossection, yes. The area of the sheared surface. But not W x H.

Ted

 

[boo1]

Phi*R**2

 

[Reifleman2]

That makes more sense I suppose. Takes into account the full amount of material sheared instead of jsust the max horizontal area.

Thanks for all the help.

 

[boo1]

horiz area would be a bearing loading

 

[dvd]

Since perfection is somewhat hard to attain, I would recommend to consider whether bending also played a part in the failure. I am not sure what the dog points were seated into, but it is probably a hole with some clearance to the dog point which would allow a bending moment to be imposed.

 

[Reifleman2]

Just for anyone curious, thats exactly what we believe happened. The dog-point setscrews hold a simple retaining ring (no actual pressure drop) and seat into the seal sleeve on the shaft. We think the hard startup caused enough torque to twist the sleeve on the shaft (WOW!) and shear the dog-points. Its a bit of a long shot, but this is one of the strangest failures Ive had to troubleshoot.

Rotating equipment is never boring