Breakaway Bolt Design 3

[Rsk2mc]

I am currently working on a breakaway Bolt Design. Ideally the first head of the bolt will break off at 45 ft-lbs. How can I calculate when the head will break off, assuming I know the neck down diameter that will break.

 

[Screwman1]

You can get an fairly close estimate by taking the 70% of the UTS of the shear area to give you the shear value, but you will still have to confirm by testing. I have done many of these over the years and depending on the alloy and HT, you may need to adjust up or down to get the shear torque that you are looking for.

 

[fegenbush]

Screwman1 is correct that you need to start with a calculated value, verify with testing, and modify design as necessary. What I would like to add is that for the intended weak area, you should have tight controls on surface finish and geometry. As an example, if you intend to have a sharp notch formed by cutting, you should take into account tool wear at the sharp point of the tool. As it wears and/or rounds, the cross sectional area will change as will the stress riser effect.

 

[Rsk2mc]

I have used the equation for torsional shafts tau = T*R/J and calculated a value that is almost spot on to the tested value. I have tried 3 different bolts and solved for T in the equation and this seems to give me a very close approximated value.

 

[IFRs]

You may need to test each batch of bolts to make sure they are not varying. Some specifications guarantee minimum values only.

 

[chicopee]

Three bolts tested is hardly statistically valid. You'll need to test a lot more bolts. Batch testing as mentioned above is one way, however the number of bolts per batch to be tested will have to be determined by someone knowledgeable in the statistical method.

 

[Dougt115]

Batch size can be determined based on your confidence level that the bolt will break within a certain range. 80 to 90% confident that your part will break between two load limits. The smaller the range or the larger the confidence usually result in the larger the test sample size.

If your tests came back with a range of +-2 lbs and your requirement is +-50 lbs you may be good. But if your requirement range is +-5 lbs you will need more samples.

 

[Compositepro]

If there is a great deal of test scatter relative to the desired tolerance, more testing will not make the tolerance any better.

 

[desertfox]

Hi

I have done these kind of fixings in the past, the formula you've quoted is for the torsional stresses in the elastic region, the formula I would use is the one that gives the plaits stress for the whole cross section ie:- 4/3 * T.
Then I would do some practical tests and adjust my bolt groove to allow for strain hardening etc.
It's worked well for me in the past.

 

[rb1957]

out of curiousity, why do you want the bolt head to torque off ? normally it would be a tension (axial load) failure, no?

Quando Omni Flunkus Moritati

 

[Rsk2mc]

i have tested 12 bolts in the lab. Two different sizes, the M10 broke at 32.5 ft-lbs, using the above formula i calculated them to break at 31. The m12 bolts broke at 30 and i calculated 33. I think the formula is pretty good estimation to design the bolt then verify with testing.

rb1957

is so the customer can just tighten them till the head breaks off, makes it dummy proof.

 

[Screwman1]

The biggest variable in the break off torque is the heat treat variation. You need to try to hold the hardness in mid range or you can get some fairly wide variation in values. When we do these we make the torque a minimum and then allow plenty of headroom for clamp load variation to prevent possible tensile failures when they are installed in the field and 'Bubba' decides to lube the bolts. The other way to control things a little bit better is to use a pre-applied high lubricity coating to keep the friction low and consistent.

 

[bigTomHanks]

Not sure if you are aware but this type of fastener is used all the time in aerospace manufacturing. You may be able to buy what you need from a current manufacturer.

http://www.lisi-aerospace.com/products/fasteners/externally-threaded/pin/Pages/hi-lite.aspx

 

[Rsk2mc]

Thanks for the link, unfortunately the bolt we are using is a custom fastener for a project.

 

[tbuelna]

Shear of a drive feature on a bolt head is probably one of the least accurate methods of controlling preload. Using a similar feature on a nut would be better. And using something like a PLI washer under the bolt head would be better yet.