Analyzing stress in thin plates

[jmccalip]

Let's say I have a thin plate in flexure. I have seen engineers at me job use regular Sigma=M/S, as well as plate theory.

We have several books on plate theory, but the process is rather complicated, and I can't always use the pre-derived formulas since our stuff is always oddly loaded/complex. How accurate is use just M/S for thin plates?

 

[JedClampett]

It's OK until the deflection gets large (and that's not that large) and the plate starts to go into membrane action.
So it depends on how thin is thin and how large your loads are. For a 1/4 inch plate spanning 8 inches, you're probably good. For a 20 gage plate spanning 30 inches, you got issues.

 

[JStephen]

It depends on what you mean by "M/S". If it's a rectangular plate supported like a beam, that should be reasonably accurate. If it's some random shape, that may or may not be accurate. In some cases, you're finding the average moment across a line, but if you have both positive and negative moments acting across that line, you can get a total moment of zero.

As to the acceptability- if it's small one-off items made of carbon steel, it may be cheaper to increase the thickness than to do a more sophisticated analysis.