bolted connection equations 4

[dharhay]

I use the bolted connection equations from textbooks to determine either load on part or load on bolt. These equations have as inputs: bolt torque load, part load, stiffness of the part, stiffness of the bolt. My question is do many of you use this or are familiar with these equations.

Often I need to show co workers these equations and I am told these cant be correct. I am not shown a more sound way to find bolt load or the part load. I have seen other companies use this method. Maybe its not that well known.

One reference for these equations: Shigley & Mitchell, McGraw-Hill, 4th Edition, page 369. Thanks for your input.

Dave Harhay
Fives Bronx

 

[rb1957]

your approach seems generally correct.

i'd be interested in why others think "these can't be correct" ?

another day in paradise, or is paradise one day closer ?

 

[dharhay]

Thanks for your reply. The single comment that I get most is that the bolt sees no additional load over that of its preload until the part load exceeds the bolt preload. I show the equations that I mentioned above and say that this could be valid under certain circumstances but its not a generality. People then walk away.

The good thing is that the bolting equations are very conservative and using the above equations I have had no failures and neither have the others that I know of. I feel confident because of the use of fundamental equations.

Dave Harhay
Sypris Technologies
Tube Turns Products

 

[rb1957]

"the bolt sees no additional load over that of its preload until the part load exceeds the bolt preload" ... this is a very common simplification of the real world.

draw a bolt load diagram. Simple story is bolt load equals preload untill the joint gaps, and then bolt load = external load. This implies an instanteous reduction of joint compression.

The real story is, as you can picture, the external load gradually reduces the joint compression untill the joint separates. This means two things, counter to simple story ...
1) the bolt load increases above preload as the external load approaches the preload, and
2) the joint separates at a load exceeding the preload.

Neither of these things is very large, so the simple story is a reasonably correct view of the world.

another day in paradise, or is paradise one day closer ?

 

[dhengr]

Dharhay:
I think what rb1957 really meant to say is that the difference btwn. these two load conditions ‘is not very large.’ And, I agree with his post. Otherwise, the loads on a 1.5" high strength bolt can be quite large.

 

[desertfox]

Hi dharhay

After a bolt is preloaded it's does see additional load when the external force is applied, how much it see's is dependant on the stiffness of the bolt and the clamped parts.
The joint will only separate when the bolt preload is exceeded.

The calculations you quote are used daily by millions of engineers.

 

[dharhay]

Yesterday, three engineers had no idea of the bolted equations, provided none in lieu of them and told me that what I say cant happen. Today I get to talk to R&D mgr and hear what he says. Thanks desertfox for adding your familiarity with those equations.

Dave Harhay
Sypris Technologies
Tube Turns Products

 

[Screwman1]

Absolutely the tension on the bolt will increase as the tensile service load increases. the amount of increase will depend on the relative stiffnesses of the bolt and joint. A stiff joint and elastic bolt will see much less load increase than the opposite combination. This is why we tend to say that as a rule of thumb you want your joint stiffness to be 4-5 times the bolt stiffness. The joint diagrams will show how the proportions change.

 

[rb1957]

how it looks ...

another day in paradise, or is paradise one day closer ?

 

[dharhay]

I am truly amazed that so many engineers don't see this. Thanks again for adding your input.

Dave Harhay
Sypris Technologies
Tube Turns Products

 

[ricklts]

DHarhay,

The very best and most definitive author on bolted joints has to be John Bickford. Get one of his books, read it, and you will have a much deeper understanding of the complexities of such a simple mechanism.

Rick

 

[EHudson]

I call it the 'flea' equation: Force x Length / Area x Elastic modulus = deflection. This holds true for both the bolt and the joint in compression.

first iteration: Get the proof load of your fastener material say 120ksi for a grade 8 put in the grip length of the bolt and figure the area of the bolt shank and use about 30E6 for steel modulus and calculate a deflection at the proof load. Start your load diagram by making the vertical scale your load values (lbf.)and horizontal the deflection in .001". Use 0,0 and the point you calculated and draw a straight line. Extend the line past the proof load about 20%.

As a rough approximation you can divide the clamped area by the number of bolts to get an area to plug into the formula.

Figure out how much the clamped element will be compressed by the proof load of your bolt. The slope of the compressed joint will be a negative slope on your diagram. Start at the proof load point of your bolt and go across to as much as you calculated for the compression of the joint and down to the zero load line and make your second point. The slope of the joint compression should be much steeper than the slope of the bolt stretch. Let's hope it is three times or more as steep. Use appropriate material properties.

Now for fluctuating loads place a vertical line that scales to the fluctuating load between the bolt slope and the joint slope lines. The amount of load above the proof load point is the amount of additional load carried by the bolt and the amount of the load line below the proof load point is the amount of force removed from the joint during loading. Your diagram should look something like the attached sketch.

In truth the bolt stretch is not a straight line and there can be thermal deflections and gasket compression to account for but you have a start.

 

[rob768]

this site has some very usefull information:

http://www.boltscience.com/pages/basics1.htm