Strength of laminated steel plates vs. solid section

[Sparky4598]

I am working on a guard design that needs a tubular steel section about 24" inside diameter and 36" long. I am wanting to make the tube section 1-1/2" thick, but no standard pipe or tube suppliers that I can find appears to be close to that.

My next thought was to make it out of laminated steel rolled plates. Say 2pc @ 3/4" thick or 3pc @ 1/2" thick so it is easier to fabricate. My concern is impact strength though. Should I expect the laminated configuration to have similar strength to a solid section or would it be drastically (say more than 30%) different?

Is this something that would likely be considered extremely difficult to model with FEA?

Any other thoughts or suggestions as to fabrication options?

The guard is a safety device to protect workers in the vicinity should the shaft coupling inside catastrophically fail.

Thanks for your time!

 

[dvd]

You should investigate forgings - https://www.scotforge.com/Forged-Products/Hollows

 

[Doug Hunter]

That is not a simple problem. The right FEA tool may be able to handle it. You'll need to clarify if you want to resist bending and yielding, or if you want to absorb energy within a certain amount of yielding. Multiple laminated plates would absorb energy differently because they could slip relative to each other with some frictional energy dissipation.

 

[ctopher]

Just a guess. Maybe machine donut plates with lap joints, weld them together to make your pipe?
A long time ago we did this once for a special test fixture to handle high pressure.

 

[Geoff13]

While your dimensions are very close to 24" Sch 100 pipe, I'm not sure who would have a 3' length laying around.

If you have fabricators that can roll 0.75", have you asked if they can roll 1.5"? It's much harder to roll, but the short length will make it easier. I'd find a fabricator with the equipment to handle 1.5" as opposed to the layered approach. Rolling plates thicker than this is regularly done, but your small diameter may limit who can do it.

If you must use multiple layers, you will have to plan for "large" gaps between the layers. The 2 or 3 sections won't be perfectly round nor will they have perfect OD / circumference. I'd be nervous, from a fabrication point of view, specifying a 1/8" gap all around and would likely go with 1/4" gap all around. Your fabricator can help select gaps that suit their capabilities. This gap size will affect the analysis.

There are ways of building layered pressure vessels without these gaps (never done such a thing myself) but I expect this would be way outside your budget for a safety shield.

 

[EdStainless]

There are fabricators out there who can roll plate into a shell, and they roll multiple layers.
Then they weld both the ID and OD seams.
For a single short piece a forging will likely be the better option.

 

[MintJulep]

Casting.

But, does it **really** need to be a tube?

Would a square, or a hexagon, or an octagon from flat plate work just as well?

 

[btrueblood]

These guys might do it for you -

Steel Plate Rolling Services | Heavy Plate Rolling

Chicago Metal Rolled Products offers full service steel plate rolling & heavy plate rolling for all applications. Contact us for your custom plate rolling project.

www.cmrp.com

2" plate x 10" wide into a 19" diameter cylinder is one listed size near yours.

 

[GregLocock]

You may find that a Kevlar tube or a kevlar wrap over a thinner pipe is a better solution, and certainly easier to make.

Strengthwise 3 laminated plates will have the same bursting pressure as a solid tube, but for local impacts it might even be better than solid, if the plates can slide relative to each other. The trick is to absorb energy in the collision, rather than trying to contain it while staying in the linear elastic part of the stress strain curve.

 

[Sparky4598]

Thank you all for the great responses!

So to answer a few things, the primary concern is containing the energy. If the guard is destroyed in the event of a failure that is fine as long as the guard contains any debris. Some yielding would probably be preferred to absorb some energy.

My initial thought was to have the inner laminated plate be AR400 quenched and tempered plate to avoid a piece of debris possibly causing too much localized damage to the guard, but now thinking about it that would be nearly impossible to punch through 1-1/2" steel and it would also greatly hinder any energy being absorbed by yielding.

Going along with this, it may be beneficial to have some gap between the laminated shells as this would allow some energy absorption before being reinforced by the plate behind it.

I have checked with several fabrication shops around and several can roll 1-1/2" plate but not as small as a 24" ID. That's what has led me to the laminated approach.

 

[Sparky4598]

These guys might do it for you -

Steel Plate Rolling Services | Heavy Plate Rolling

Chicago Metal Rolled Products offers full service steel plate rolling & heavy plate rolling for all applications. Contact us for your custom plate rolling project.

www.cmrp.com

2" plate x 10" wide into a 19" diameter cylinder is one listed size near yours.

Awesome, thank you! I will check that out and see if they can do it. Though from the sounds of some other comments, I may be better off using the laminated approach anyways for better energy absorption.

 

[Sparky4598]

You may find that a Kevlar tube or a kevlar wrap over a thinner pipe is a better solution, and certainly easier to make.

Strengthwise 3 laminated plates will have the same bursting pressure as a solid tube, but for local impacts it might even be better than solid, if the plates can slide relative to each other. The trick is to absorb energy in the collision, rather than trying to contain it while staying in the linear elastic part of the stress strain curve.

I had considered something similar like carbon fiber wrap, but expected it would be too expensive for a 1 off part.

 

[dvd]

What is the rotational speed? How did you decide how much energy needs to be absorbed? What kind of coupling is this?

Attached are a couple of references I found when investigating containment of high-energy rotating parts.

 

[Sparky4598]

What is the rotational speed? How did you decide how much energy needs to be absorbed? What kind of coupling is this?

Attached are a couple of references I found when investigating containment of high-energy rotating parts.

Wow that's some great info, thank you very much!!

It is a 400-lb part steel part fiberglass coupling spinning at 1500rpm.

Not much has been done yet as far as calculations. It's mostly a way overkill to be safe SWAG for the moment as to the 1-1/2" thickness. After calculations are done it may be thinner, but I try to think ahead and figure out if something can even be built reasonably before I put a ton of work into calculations for nothing.

 

[rb1957]

I was going to suggest kevlar but someone is already there ! easy to wrap a 24" dia mold with as many plies as you think necessary.

FEA may get you into the ballpark, but I think you'll need an impact test (or validate your FEA with panel tests ... easier to do).

At least it'll be much lighter ! (pi*24*36*1.5*0.3 = 1300 lbs ??) Of course weight is your friend in this type of application.

 

[dgeesaman]

There are shops that will roll/weld about any reasonable plate thickness into about any reasonable diameter. There's a big one in Texas that specializes in that and one of my vendors uses them regularly for even very small quantities.

 

[EdStainless]

We had something similar (small parts but much higher speed).
It was a fairly thick fiberglass internal layer, a rather thin high strength steel mid-layer, and a substantial Kevlar/epoxy overwrap.
The idea what use the inner layer to spread impact.
The steel would absorb a lot of energy.
And the Kevlar would contain the shrapnel.
We never 'tested' it.