How thin can this hollow shaft be and still be structurally sound? 2

[gkelson]

I'm designing a shaft for a turbine and I want to hollow it to save on material, but I'm not sure how hollow I can make it. There will be stress concentrations at some grooves for snap rings along the shaft. It's modeled in Pro/E, but I am having lots of trouble with the Mechanica application. I could approximate stress using a fixed end 3-point bending analysis, but for some reason I can't find a good formula for max bending moment.

The shaft is supported in bearing housings about 8 inches apart and in the middle holds a weight of about 41 lbs. The shaft will be Al 6061 and I need to know how little material I can use for it to work. The shaft diameter is 1.378 in. with 1.29 in diameter grooves for snap rings. Any help?

 

[MikeHalloran]

Do you sometimes wonder if, given your education and training, whether you should be doing a particular task?

I do.


Mike Halloran
Pembroke Pines, FL, USA

 

[desertfox]

Hi gkelson

I think a sketch of your shaft might help us, my initial reaction would be to model your shaft as a simply supported beam,supported by the bearings, but you mention a 3 point bending analysis with a fixed end, can you expand on that, ie is it a cantilever beam with end supports?

desertfox

 

[hydtools]

How will you create the hollow shaft? If you machine it from a solid bar, no material was saved. You reduced weight but did not save material.
A sketch would be helpful.

Ted

 

[MJCronin]

I second "Mike Halloran"....

Any chance that the original poster has taken a fundamental course in machine design ? Or is his background in Pro/E ?

The primary loading is in torsion and shear stresses will govern.

There will be impact as well as bending loads

Aluminum, IMHO, may not be the best material choice

 

[kacarrol]

I took this stuff in University, I'm just trying to remember which text books cover this..I seem to think that Shigley's Mechanical Engineering Design covers shafts and covers stress risers at grooves? It probably builds on the concepts learned in Hibbeler's Engineering Mechanics.

Or it could be fatigue and fracture analysis that covered shafts with grooves, that text would have been Dowling, Mechanical Behaviour of Materials (4ed I think is the latest).

Sorry I couldn't be a little more specific. I don't have access to my school notes at work and I went more into piping than machine design.

Good luck!
K

 

[rb1957]

turbine shaft brings "whirl" to mind ... the stiffness of the shaft to prevent whirl fultter.

i 3rd mike's suggestion. there is Way too much in this to be able to sensibly (and safely) help.

"6061" ... wouldn't have been my first choice, but then maybe you've got your reasons (hopefully more than "taht's what i can get my hand on quickly").

you can't find a good solution for maximum bending moment of a three pt beam ? ... pls see above, 2nd para.

 

[Cockroach]

This is a very easy problem GKelson, your simply load the shaft and do the shear, bending, rotation and deflection diagrams according to your known supports. This will give most of the information you seek. The most I can offer is a verbal direction for possible solution. Your post offers no diagram or loading details, how can you possibly expect an answer?

Stress concentrations are typically risers of three. Depending on the geometry of the disruption, you may be able to look at that case specifically from the literature. Otherwise you need to approximate it. FEA would be the best option.

Finally you would account for the effects of dynamic motion in addition to the static loads. For example, you will have a torque do to twist and some resistance load at the opposite end. There may be other resistances on the shaft. These loads would apply shear to the system.

I would then look at your normal and shear loads and apply Mohr's Circle to obtain the principle stresses in the shaft. Compare that to known material properties to obtain the factor of safety. You would then consider the wall thickness as a function of this factor of safety, look at worst case loading and see if the stresses are acceptable.

You're in for the mathematics. Hope you like equations. Could be a very cute problem, but pretty much first year statics. Start at a classical textbook problem that closely resembles yours and then apply that "recall" to your problem.

But for your particular problem, as most in the real world, there is no recipe book dictating a closed form solution. Wish it was that easy, I'm sure that is shared by many in this forum.

Good luck with it, have fun.

Regards,
Cockroach