How to do engineering calculations?

[sparc]

So why does the same height W-shape take more weight than the same height S-shape?

 

[AKwelder]

Beams deflect first, then fail. The main reason you see H shapes is to avoid deflection.

A good old welder ounce showed us a trick. He showed us that if you could stop a piece of paper from deflecting it would support a huge amount of wieght. Just think about making a header, two 2 by 10's. Now put a layer of tin between them, huge strength increase

My point is, it's the deflection that leads to failure. Looking at the original posters picture, I would add a couple mid field 1-1/2 angle braces welded to the top and bottom rails. This would spread the load, and stop the expanded metal from deflecting and pulling on the perimeter.

 

[RedNeckRacin]

Beams deflect first, then fail. The main reason you see H shapes is to avoid deflection.

A good old welder ounce showed us a trick. He showed us that if you could stop a piece of paper from deflecting it would support a huge amount of wieght. Just think about making a header, two 2 by 10's. Now put a layer of tin between them, huge strength increase

My point is, it's the deflection that leads to failure. Looking at the original posters picture, I would add a couple mid field 1-1/2 angle braces welded to the top and bottom rails. This would spread the load, and stop the expanded metal from deflecting and pulling on the perimeter.

I'm reaching back in my memory banks but IIRC Deflection is not a failure mode unless the beam is unrestrained and slides off of the supports. Deflection is a controlling characteristic because a beam sagging under a load is very unnerving to the average person even though its safely carrying a load. Many of the design projects that I did were controlled by deflection if the connection points held through their failure analysis.


Torsional buckling, shear, crushing and a few others are failure modes.

 

[s219]

Deflection is OK as long as it's within the elastic limits of the material and acceptable for the design/situation. Beams deflect and come back to normal all the time.

Now, buckling, which I think is the real point of AK's post, is a problem. That's when the beam/member goes off in another direction from the applied load, and all **** breaks loose. Column buckling is a prime example, which is why columns are often a lot larger radially than they need to be for a pure compression load. They need the extra beef to resist buckling from asymmetric loads.

 

[AKwelder]

I agree, deflection is not failure, sorry to get that mixed up.

I do hope that after all this the OP can find a simple fix to meet his needs

 

[Cord]

Quote Originally Posted by Shield Arc View Post
So why does the same height W-shape take more weight than the same height S-shape?

Because it places more material from the center line.

 

[Egon]

It's all about section modulus.

 

[Baby Grand]

I like one of the earlier responses, by s219, that encouraged you to look at other, successful designs and to copy them, making improvements where needed.
To quote Tom Lehrer:
...let no one else's work evade your eyes. Remember why the good Lord made your eyes, so don't shade your eyes, but plagiarize, plagiarize, plagiarize! Only be sure always to call it, please, "research".
This is fine so long as you are doing this for your own use and not for profit.
If for profit, you will need to copy much more carefully.

Or you can execute your own design.
How to do engineering calculations?
Rule #1: Assume that you are wrong and review all of your assumptions.
Rule #2: When you have convinced yourself that you are no longer wrong, have someone else (who's judgment you deeply respect) review your work.
Rule #3: After you have fixed the dumb mistakes that your checker has pointed out, it's time to build a prototype and do some testing in the sandbox, before you commit time and materiel to a finished product. The prototype should be unpainted and easy to alter once you determine that you were wrong, again.
Rule #4: Test the prototype. Murder cycle - test it to failure and see how much margin (safety factor) you have designed into it. Ask yourself if this is enough margin to trust your best friends' life on.
Rule #5: Make appropriate changes, weld, then paint. Enjoy.

Engineering calculations are only as good as the assumptions that guide their use.
A gun is only as dangerous as the person holding it.
Either be humble in design or prepare to be humbled by your design.

This is not for everybody, but if you can live with the occasional ego crushing failure and take precautions to isolate the risk to a tolerable level, it can be a very satisfying thing when your product begins to work as you had intended.

 

[RedNeckRacin]

I like one of the earlier responses, by s219, that encouraged you to look at other, successful designs and to copy them, making improvements where needed.
To quote Tom Lehrer:
...let no one else's work evade your eyes. Remember why the good Lord made your eyes, so don't shade your eyes, but plagiarize, plagiarize, plagiarize! Only be sure always to call it, please, "research".
This is fine so long as you are doing this for your own use and not for profit.
If for profit, you will need to copy much more carefully.

Or you can execute your own design.
How to do engineering calculations?
Rule #1: Assume that you are wrong and review all of your assumptions.
Rule #2: When you have convinced yourself that you are no longer wrong, have someone else (who's judgment you deeply respect) review your work.
Rule #3: After you have fixed the dumb mistakes that your checker has pointed out, it's time to build a prototype and do some testing in the sandbox, before you commit time and materiel to a finished product. The prototype should be unpainted and easy to alter once you determine that you were wrong, again.
Rule #4: Test the prototype. Murder cycle - test it to failure and see how much margin (safety factor) you have designed into it. Ask yourself if this is enough margin to trust your best friends' life on.
Rule #5: Make appropriate changes, weld, then paint. Enjoy.

Engineering calculations are only as good as the assumptions that guide their use.
A gun is only as dangerous as the person holding it.
Either be humble in design or prepare to be humbled by your design.

This is not for everybody, but if you can live with the occasional ego crushing failure and take precautions to isolate the risk to a tolerable level, it can be a very satisfying thing when your product begins to work as you had intended.

That is some of the best engineering advice I think I have ever seen given on this forum. BabyGrand thank you for that excellent post. I have only one thing to add from my engineering degree.
K.I.S.S.
Keep It Simple Stupid The fanceir your make something, the more ways it is possible for it to fail.

 

[Baby Grand]

That is some of the best engineering advice I think I have ever seen given on this forum. BabyGrand thank you for that excellent post. I have only one thing to add from my engineering degree.
K.I.S.S.
Keep It Simple Stupid The fanceir your make something, the more ways it is possible for it to fail.

Thanks, Red!
And good you remembered to mention K.I.S.S. (I knew I was forgetting something) - that makes things so much easier.