Design bracket for failure at a certain load 10

[ajdb]

Hello all,

As the subject line says, I am curious about designing a bracket that will fail at a specific load.

At the moment my question is where should I start. When searching around on google I haven't been able to find much on the subject.

 

[oldestguy]

I'd make use of a material that has a specific value of elasticity coef. and use that to trigger a collapse at a certain strain.

 

[E720]

This reminds me of the shear pins used in aircraft carriers. The aircraft have cables that are attached to these shear pins. Once the aircraft are tugging on those cables enough then the shear pins shear in half and the aircraft are able to take off at high speeds. That is probably super dumbed down but I believe that is the general idea.

If I were to try and design a component that failed at a certain load I would try and make it fail in shear if possible because it wouldn't involve a long yielding process. I would make sure that everything else was much stronger and that I had a "structural fuse" that could break. I would probably make the fuse out of a more brittle material than regular structural carbon steel.

 

[TimSchrader2]

Hello. Yes, shear pins is the normal way.

 

[dhengr]

Ajdb:
You should start by defining your problem with enough specificity so that an experienced engineer had a chance at understanding what you were actually trying to do. It never ceases to amaze me that engineering questions, these days, here on E-Tips, are asked in such a general, un-engineering way, as to sound like we were dealing with a bunch of third graders. What kind of bracket? What loads, types of loading conditions, etc. etc? Dimensions, sizes, materials and back support structure to be used? What constitutes failure, in some detail? Most of the good engineers here on E-Tips don’t want to play 20 questions for the next two weeks, you’ve got to give a few meaningful hints.

 

[WARose]

Some type of shear is a good place to start. Shear isn't a very forgiving situation. 2-3 times the (calculated) ultimate capacity is a good upper limit I would think. (Assuming we are talking about steel here.)

EDIT: Just had to give that a star dhengr.

 

[Agent666]

My solution ensures a fast failure (seems to have been flagged as being important without any context).

Therefore if you want to ensure a fast failure, put some strain gauges on the bracket, once you reach the desired failure strain, detonate the explosives strapped to the bracket. Problem solved?

 

[ajdb]

Thanks everyone for your thoughts.

dhengr, I asked generally because I don't have a specific example yet. I tried searching on google for any papers or books on the subject, but haven't been able to find anything. That is why I was generally asking where should I start? If there was a fully defined project I would have specified more details.

Agent666, I didn't mean to flag anything as important? Sorry if I made a mistake, however I like your solution .

 

[SwinnyGG]

That is why I was generally asking where should I start?

You should start by clearly defining the full set of requirements your final part will need to meet.

 

[TehMightyEngineer]

Even a simple sketch would be a better start. Is the bracket in flexure? Shear? Torsion? All of the above? You didn't even tell us what material your bracket is made of. (Metal? Plastic? Composite?)

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[ajdb]

I don't have a project to tell you about. Are there no books or papers that talk about a component (can be anything) that they designed to fail at a specific load? Are there no examples?

Like E720 mentioned shear pins, and I was able to look that up and find data about those specifically.

What other components have been designed to fail at a specific load? I am not asking for a specific bracket or material, I am asking generally. Is there a general term for components that are designed to fail at X force or Y torsion or Z combination of shear and bending?

 

[Celt83]

there are break-away anchors:

http://www.heckmannbuildingprods.com/firewall_anchors.html

- example of some for firewall applications

I think on the Department of transportation side things like sign posts and rails are design to break or go completely plastic to help minimize injury when hit by vehicles. Same thing with vehicles themselves think bumpers and crumple zone items design to "fail" to dissipate crash energy.

Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

 

[WARose]

Is there a general term for components that are designed to fail at X force or Y torsion or Z combination of shear and bending?

Not that I am aware of. If you'll recall from basic mechanics: there is variability in material strength, load re-distribution as a widget approaches failure, and so on. Nothing is going to fail at exactly anything. The best you can do (if you actually want something to fail) is to design something so that if it sees a certain load....it won't make it. (See my post earlier in this thread.)

The shear approach is a good one (as I said). To use a real world example: I use to work at a company that did machine design.....and there was a part we designed to fail so that a gear box didn't see certain loads. Ultimately what was used was a small bolt that would fail in shear.

 

[ajdb]

The shear approach is a good one (as I said). To use a real world example: I use to work at a company that did machine design.....and there was a part we designed to fail so that a gear box didn't see certain loads. Ultimately what was used was a small bolt that would fail in shear.

For your project, did you guys do testing afterwards to verify? If so, how close were the tests compared to the calculations? And did you guys use FEM to dimension the bolt or were hand calculations enough?

there are break-away anchors:

http://www.heckmannbuildingprods.com/firewall_anch...

- example of some for firewall applications

Interesting, thank you for the information.

 

[oldestguy]

An old one is set up a hanging with the victim on a horse and the noose around his neck. One shot fro9m a pistol will get the neck broken quickly. Actually if done right the jaw bone punches into the brain.

 

[WARose]

For your project, did you guys do testing afterwards to verify?

Nope. It got tested on the job.....and we heard no complaints. What is testing going to tell you anyway? You are going to see the established variables I was discussing before. We didn't use unknown material.

And did you guys use FEM to dimension the bolt or were hand calculations enough?

Hand calcs.

 

[charliealphabravo]

If you have no specifics then you will want to start by researching some well known designed-to-fail applications.

1. Explosive bolts used in aeronautics, aerospace, and military applications.
2. Anchor bolts on highway signs and luminaries.
3. Fall arrest dampers.
4. Automobile crush zones.
5. Drive shaft shear pins
6. Pyrotechnics and blasting

 

[ajdb]

What is testing going to tell you anyway?

Would it not be possible to test until failure to verify that it will fail around the desired loading?

If you have no specifics then you will want to start by researching some well known designed-to-fail applications.

Thank you, will look those up.

 

[BUGGAR]

Problem is, materials are designed to minimums, except for "seismic" steel which is also constrained by a maximum strength as well as a minimum strength.
Except for those propeller shear keys in the old outboards. If you can figure out how they accurately made these fail when you were in the middle of the bay, you have your answer.

 

[racookpe1978]

Would it not be possible to test until failure to verify that it will fail around the desired loading?

1. But, you THEN NEED to build 8-10 of your brackets, and REPEAT the destruction test (pull-to-failure test) so you can determine (at a minimum!) the variables of your failure.
+/- 5%?
+/-25% ?
+/-50% ???? What will be your acceptable variation in the actual failure from the design (intended) failure?


3. At a minimum, to even determine a single standard deviation you need at least 33 sample destruction tests though, assuming your results are normally distributed.

2. To "improve" the failure point, use a shear pin, but machine a precise groove at your intended failure point. Do NOT let the failure occur at a random location on the pin between it two mounting points, but at the specific dia you have machined across the pin so the dia is weakest at a known position between the two holders. Specify PRECISELY your material and grade of the pin construction. Specify precisely the tolerances for the pin, the groove, the mounting holes and mounting methods.

You also have NOT said if your "load" is an impact (moving) or is a static (gradually increasing) load - as if a snow fall or rising water level in the wall of a tank. If I drop 15 lb sharp-pointed drill bit on a chair from 6 foot, I can break the same chair seat that will hold my 160 pound body if I stand on the same chair carefully and slowly.