Load carrying expansion joints

[theOrange]

Are there any expansion joints (for example between two beams) which can carry a shear force?

 

[OldEngr63]

The short answer is "yes," but it would help to know what your geometry is. How about a picture?

 

[theOrange]

The short answer is "yes," but it would help to know what your geometry is. How about a picture?

I hope the image attached is clear. I am looking for an expansion joint which allows the top beam to expand horizontally (x-direction, see image), while also being able to carry the load on the top beam.

I've found expansion joints by just searching on google, however they are all for pipes or bridges. In either case they are not applicable to my situation.

 

[OldEngr63]

How about if you simply top one of the columns with a flat plate and then put one or more rollers between the beam and the plate?

 

[billy_joule]

Calculate the expansion first. You may find over sizing bolt holes by a few mm may allow enough room for expansion.
a 500 degree delta T is only 8mm expansion for a metre length of steel (0.000016 (m/m ^OC)) so over-sizing bolts holes by just 2mm is all that's needed.

Otherwise, pin joins at the top two corners, pin and (horizontal) slots at bottom two.

 

[OldEngr63]

billy_joule said, "Calculate the expansion first. You may find over sizing bolt holes by a few mm may allow enough room for expansion. a 500 degree delta T is only 8mm expansion for a metre length of steel (0.000016 (m/m OC)) so over-sizing bolts holes by just 2mm is all that's needed."

How does that work, Billy? Will you leave the bolts loose in the holes so that they can slip when the frame is heated? Looks to me like if you bolts are snugged up, friction will eliminate your expansion capability.

 

[billy_joule]

Yes, Don't over tighten the bolts. Just like a pin and slot join:

 

[OldEngr63]

Billy, this looks pretty loosy-goosy to me for a structure.

 

[billy_joule]

That's the first picture I found to demonstrate what a pin and slot join is, obviously the slot length will be not much more than it's width in the OP's case.
Designing poorly defined structures with unknown loads on an internet forum will net loosy-goosy results.
It's always good to start with a simple design and go from there.
A humble M10 8.8 at each corner will take over 10,000kg in pure shear with a SF of 3, in most cases the material being bolted will fail before then.
Anything that's lifting those sorts of loads shouldn't be designed on a forum..

 

[OldEngr63]

"... over 10,000kg in pure shear ..."
You obviously measure shear in different units than I do!

 

[billy_joule]

I'll rephrase for your understanding:
The maximum static load for four M10 8.8 bolts in pure shear stress with a safety factor of three is over 10 metric tonnes in Earth gravity field.

 

[OldEngr63]

Are you implying that the carrying capacity of these bolts would be something else if removed from the Earth's gravity?

 

[billy_joule]

Of course. The bolts could support a much larger mass on the moon where gravity is weaker.The force will be the same.

 

[theOrange]

I'm not trying to get a hole design out of your guys ;). I was just trying to figure out what kind of joints there are that could handle shear loads and allow expansion. Thanks for the info, I'll do some designing and calculating with pin joints.

 

[OldEngr63]

If the thermal loading is cyclic (heat the structure up, cool the structure, heat the structures, ...) then there will be working in the joint (that's why an expansion joint is required in the first place). With the horizontal bolts that billy_joule has proposed, this will wear the sides of the bolt shanks. They are not hardened to prevent this, so wear could be severe.

There is also the problem of a tear-out in the structure, pushing the bolt through the side of the slot.

If the bolted joint must slip, as per billy_joule, then the nuts cannot be snugged up. This will require a second nut as a lock nut to assure that the first does not simply work its way off completely. If the nuts are snugged up, friction in the joint will possibly lock the joint, preventing the expansion motion, which was the whole point here.

Why use the bolts in a horizontal position when they could be vertical and transmit no shear at all? Let the horizontal beam simply rest on the top of the upright, and then there is no need to have bolts in shear. This is where a roller is helpful, because it eliminates the resulting rubbing.

Bolts are intended fundamentally to be tension elements. They are not intended to support shear. They support shear indirectly by putting the surrounding material in compression and letting the resulting friction support the shear.

 

[billy_joule]

I'm not trying to get a hole design out of your guys ;). I was just trying to figure out what kind of joints there are that could handle shear loads and allow expansion. Thanks for the info, I'll do some designing and calculating with pin joints.

I already gave you a hole design ;)
No one could give you a whole design as it seems you have no idea on the design specs!
The first step to design is understanding the problem - It appears you are trying to solve a problem without knowing what the problem is: That will never work.

If the thermal loading is cyclic (heat the structure up, cool the structure, heat the structures, ...) then there will be working in the joint (that's why an expansion joint is required in the first place). With the horizontal bolts that billy_joule has proposed, this will wear the sides of the bolt shanks. They are not hardened to prevent this, so wear could be severe.

There is also the problem of a tear-out in the structure, pushing the bolt through the side of the slot.

If the bolted joint must slip, as per billy_joule, then the nuts cannot be snugged up. This will require a second nut as a lock nut to assure that the first does not simply work its way off completely. If the nuts are snugged up, friction in the joint will possibly lock the joint, preventing the expansion motion, which was the whole point here.

Why use the bolts in a horizontal position when they could be vertical and transmit no shear at all? Let the horizontal beam simply rest on the top of the upright, and then there is no need to have bolts in shear. This is where a roller is helpful, because it eliminates the resulting rubbing.

Whether these issues are even relevant depends on the design spec.
If the joins only experiences self weight during expansion/contraction then wear will be low. If it's used once a week then wear will be slow. If the bolts cost $1 each and last 10 years then the cost of wear is low. Tear out will only occur if the rated load is exceeded.

It's just a simple cage to hold bits of Al in an oven. I've seen welders whip up more complex structures in an afternoon without even drawing a sketch - all they need to know is a few dimensions, the mass of Al and the oven temp. No need for some guy with a degree to design it!

 

[Dr.D]

"If the bolts cost $1 each and last 10 years then the cost of wear is low." This assumes, of course, that you change them out in time to avoid a catastrophic failure.

I sure wish I was a smart as billy_joule!

 

[theOrange]

I already gave you a hole design ;)
No one could give you a whole design as it seems you have no idea on the design specs!
The first step to design is understanding the problem - It appears you are trying to solve a problem without knowing what the problem is: That will never work.

You shouldn't assume so quickly about people. My question was: 1. is there an expansion joint which can carry shear; 2. If yes, what kind of expansion joint. I haven't told you anything about the exact problem or design specs, but that doesn't mean I don't know them.

 

[billy_joule]

"If the bolts cost $1 each and last 10 years then the cost of wear is low." This assumes, of course, that you change them out in time to avoid a catastrophic failure.

I sure wish I was a smart as billy_joule!

I performed preventative maintenance as an apprentice electrician - it is a bore and I assure you requires no smarts so is often passed to labourers or apprentices.
Many machines in industry are designed so cheap easily replaceable parts wear before the expensive bits. They become part of the PM schedule (along with greasing nipples, cleaning filters, replacing seals, torquing bolts etc etc) and are checked periodically (daily/yearly, whatever is appropriate).
This is how industry works - you don't just put plant in service then use it until it breaks! You perform PM so it never breaks down.
There are often diminishing returns on buying high quality long life parts. So it's often far cheaper to replace a cheap part more often than pay a premium for long life parts.

You shouldn't assume so quickly about people. My question was: 1. is there an expansion joint which can carry shear; 2. If yes, what kind of expansion joint. I haven't told you anything about the exact problem or design specs, but that doesn't mean I don't know them.

I think it was reasonable to assume this thread was about the same topic as your last as they appear remarkably similar:
https://www.physicsforums.com/threads/material-properties-for-annealing-cage.770210/#post-4873375

I apologise if that assumption was wrong.

It doesn't' take long to see that here at PF the best response is given to those who provide as much relevant info as possible. If you know more specs you shouldn't hold back.

 

[theOrange]

I think it was reasonable to assume this thread was about the same topic as your last as they appear remarkably similar:
https://www.physicsforums.com/threads/material-properties-for-annealing-cage.770210/#post-4873375

I apologise if that assumption was wrong.

It doesn't' take long to see that here at PF the best response is given to those who provide as much relevant info as possible. If you know more specs you shouldn't hold back.

Yep, I think it is pretty obvious they are about the same situation. However that still doesn't mean you need to know everything.

I have everything I need, but I can't wrap that up in a nice pdf and just give out it on the internet. Yes, I gave limited information. Either give me the best answer you can with that limited info or don't.

 

[OldEngr63]

<Mod note: removed snide comment.>

Since every thing is regularly maintained, why do we read from time to time about failures in bridges, aircraft, and various other machines? Doesn't regular maintenance make this impossible?