Thin Walled Tube Torsional Buckling

[Gumpmaster]

I'm investigating the buckling of a thin walled long tube that apparently buckled under torsion. The tube also had an external pressure on it, but that pressure was not large enough to buckle to tube. I can figure out the torsional buckling of the tube and the compression buckling of the tube separately, but does anyone have an idea of how I could figure out their combined effect?

The tube is fully braced for Euler buckling. The tube is very deep underground so there isn't any way to look at the buckled shape.

 

[azcats]

AISC Spec combines compression and torsion (along with shear and bending) in Equation H3-6. Although I think you may be looking for something a bit different than that.

 

[KootK]

I'm not clear on the failure mode here. As I see it, the options are:

1) Overall member torsional buckling due to axial load (unlikely for a tube).

2) Overall member in-plane buckling due to axial load (fully braced per your post above).

3) Local side wall buckling due to axial and/or torsionally induced shear.

Are we talking about #1 here or #3? #3 seems much more likely to me.

Is this a round tube or a rectangular one?

Is there actually an applied torsion force on the tube?




I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Gumpmaster]

The compression is not axial. It's from an exterior pressure. I don't think that AISC covers this.

The torsional buckling is from a torsionally induced shear. AISC covers this, but I think it's fairly simplistic. The subject is fairly extensively researched.

I don't think that AISC cover the combinded loading. I have resources that cover both individually (although both are pretty complex), but I haven't seen one that covers the combined effects.

I'm not sure if an interaction equation such as that in the AISC spec would adequately predict the buckling load. It just adds torsional shear to direct shear & direct shear does't really have anything to do with torsional buckling. It could be a place to start, but I wouldn't have a lot of confidence in the result.

 

[KootK]

I'll assume round tube rather that rectangular. That being the case, and considering no axial load:

1) We must be talking about local wall buckling rather than overall member buckling.

2) I would not expect hydrostatic-ish pressure to contribute much to local wall buckling.

3) I'd recommend looking at the torsional shear buckling mode in isolation, using elastic formulae from sources like Timoshenko, Roark, Stability Design Criteria for Metal Structures, etc.

What is this thing and what causes the torsion? Utility pipe? Helical pile shaft?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Gumpmaster]

It's a type of well casing. Round thin walled pipe. They're also applying torque to it. Yes, it's some sort of local buckling. We can't really see the buckled shape.

Exterior pressure can certainly cause local buckling. It's pretty well documented. I have quite a bit of pressure, but not enough to get me there. A pipe I looked at a while back had a buckling pressure of only 75psi. It's greatly dependent on outside diameter and wall thickness. I would guess that the two have some sort of interaction. Certainly the exterior pressure doesn't help.

 

[JStephen]

"Very deep underground" you sure enough can get overall buckling from hydrostatic pressure.

Short answer: I doubt there's information available on the interaction, as that's an uncommon combination. A lot of the external pressure buckling criteria was either developed with submarines or pressure vessels in mind, and they don't have significant torsion. And drive shafts and the like that have significant torsion seldom have enough external pressure to be significant.

 

[KootK]

Buckling under external hydro-static pressure is possible but it tends to be a higher mode thing. Compared to the expected buckling mode associated with torsional shear, I suspect that the capacity would be much greater for buckling due to external pressure.

Can we evaluate the buckling capacity ratios for torsion and external pressure independently and compare the results? If either is less that 0.20, I'd be inclined to shelve the interaction concern.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WARose]

It's been my experience that (raw, overall) torsional buckling almost never controls the design for most sections (especially closed ones). The rare occasion I've checked a section for it, it only controlled for very short sections.....and it would yield before it ever hit that stress. (I can post the elastic torsional buckling equation if you want.)

This buckling is most likely more of a shell/local type buckling. You say you have exterior pressure in combination with this. Therefore you may want to check it via a pressure vessel handbook. They generally have means to combine overall & local stresses in thin walled cylinders.

 

[Ussuri]

API RP2A might also be worth a look. It includes hydrostatic pressure as part of the design checks. I just treats torsion separately though.

 

[Gumpmaster]

I'll take a look at those. I'm about 30 percent utilized in compression via the ASME BPVC. From the limited amount we can see, only 1 part of the wall caved in, so it would be more of a local type buckling. Thanks for the input.

 

[WARose]

Another good reference (that I didn't think of before) is 'Guide to Stability Design Criteria for Metal Structures'. I've got an old version [1976], but it has a equation for torsional buckling of such shells......and suggests some interaction equations with this and bending & external/internal pressures.

 

[Gumpmaster]

The Guide to Stability Design Criteria for Metal Structures was the best resource. It has a great summary. It didn't have the particular interaction that I was looking for, but it had several similar interactions. It was helpful. Thanks for the suggestion.