Can anyone provide helpful criticism on this weird HSS moment connection detail? 2

[AaronMcD]

We are designing a display structure made up of HSS4x4x1/4 tubes that will require field assembly of all parts. It is a space frame, and deflection/vibration is the primary concern.

We are looking into bolted connections. The final dimension of beams and posts will be 6"x6" so we can't have bolts sticking out far.

This is the only thing I can think of, using slip critical bolts and a slotted connection. Among my concerns:

1) The slot at the end will cause some slight benting of individual HSS quarters due to eccentric shear. However, the slip critical connection - washer plates and internal plates likely provide enough stiffness and shear transfer.

2) Long individual plates will have angular deformation from the weld process. This is why I specify to leave in the erection bolt, which in turn is why I need slotted holes. I am worried that if try and compensate for angular deformation the plates may be too far off for proper connection in the field.

3) Are the welded nuts allowed, and will they provide enough torsion resistance during bolt tensioning?

4) I have already checked to be sure the first bolt will bring the plates into contact under a very small portion of it's total load. I will probably have to ensure that the bolt nearest the connection in tensioned first.

5) It's really weird and might scare whoever has to install it lol

 

[WARose]

The few times I've had to field bolt HSS to HSS and I've needed moment capacity at the connection......I've typically just full-pen (shop) welded a short HSS member to the column and had a (shop) welded plate on the other end. On the other side, you'd have the HSS beam with a plate (shop) welded to it's end.

Bolt them together and you are done.

 

[AaronMcD]

@WARose, that would be great, but like I said I only have a total of 6"x6" to make the connection. I would need 8" to get flange plates and 5/8" bolts. It's something we can ask about if it comes down to it. It would definitely be much easier than my bizarre idea. Right now we are still exploring options.

The contractor claims he has done plywood box beams, but I don't think we can get the required stiffness out of those especially considering we have some long double cantilevers.

 

[WARose]

I see. Perhaps a slightly larger (than HSS 4x4) stub welded to the column (maybe via a plate) and then the HSS 4x4 beam slides into it and a lot of through bolting?

I saw that once in the field but never saw the drawing (to get all the particulars).

Any reason (by the way) this thing can't be field welded? If it can, I would think that would eliminate all this.

 

[AaronMcD]

This or something similare may work like what you suggested, but within the 6" requirement.

 

[dik]

An interesting anecdote about Buckminster Fuller of geodesic dome fame... For Expo 67 in Montreal, he had brought up a dome structure that 'connected' together. During assembly, they found the deflections were too great and ended up welding the connections. It could no longer be easily disassembled so they ended up leaving the dome in Montreal. Careful that there is not too much movement at your connections.

Dik

 

[dik]

AaronMcD: HSS tubes develop a substantial part of their strength and stiffness from welding through the corners. Most HSS texts caution about not welding the corners.

Dik

 

[AaronMcD]

It's in a museum. The designer seemed to think the museum would not allow it. Also all that field welding and removal of erection aids might kinda suck.

 

[AaronMcD]

@dik:
"Careful that there is not too much movement at your connections."
I definitely don't want flexible connections. But it doesn't seem like the slots or corner cuts will have much effect.

"HSS tubes develop a substantial part of their strength and stiffness from welding through the corners."
But if I check the top and bottom for axial load and the sides for shear, the strength should be good. I'd only need to cut the corners for a couple inches to get a wrench in there. Where is extra deformation coming from?

Also, @WARose, I think a larger (or smaller) stub with through bolting would be too flexible because there would need to be a gap in order to fit it up

 

[JoshPlumSE]

Any chance you could use a shop weld the beam to the column for a portion of it, then have something akin to your connection as a beam splice closer to the point of inflection?

It's just not ideal to be messing so much with the cross section at the point of max moment.

 

[AaronMcD]

@JoshPlum, I could do the connection almost anywhere I like as long as the parts still fit through a man door. The midspan has lowest moment in some members under gravity and many in an earthquake. The end or near the ends have least moment in some members/cases. The double cantilevers lowest moment is close to the ends. In all cases the moment demand is ridiculously low.

 

[CANPRO]

How long and heavy is the beam? I think that erection bolt is doing more harm than good. If you can get the erection bolt in, you may as well have just put the two permanent bolts in and kept on bolting it up. If the erection bolt wasn't there, you wouldn't have to slot the HSS as you have noted.

The erection bolt implies that this is of substantial size, and that it will be erected in place (as opposed to being assembled loose on the ground and then lifted in place). You need to be careful of erectability for these situations...if you have a similar connection on the other end of the beam, you won't be able to get the beam in place without splaying the posts out significantly - which could be difficult or impossible depending on what this looks like. Any chance we can get a look at the big picture here?

 

[ajh1]

Could you sleeve it? Weld a slightly smaller tube stub to the side of the column. Size to allow the beam tube to slide over the stub. Then through bolt it between the beam and the stub.

 

[BridgeSmith]

ajh1, you beat me to it - I was just going to suggest that.

I would offer that, if it's feasible, to make the sleeves long enough to use set screws instead of bolts to secure the connections.

 

[KootK]

How about a pair bolted L-shaped plates over the joint and bolted to both beam and column? I wouldn't do it for a larger scale thing but, for this, I think that fitup and capacity would be fine. I guess it would depend what proportion of the member flexural capacities you're seeking to develop.

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.

 

[AaronMcD]

@CANPRO
The point of that bolt is merely to hold the plates in place to make sure they don't deform from welding and fit into the HSS. I wanted to do what ajh1 & HotRod10 & WARose suggested with the square tube insert but I can't get a slip critical connection with the tolerance gap and am worried about rotation of the connection.

It isn't big. The members frame out a series of 9'-6" cubes with some corner posts missing and a few edge beams missing here and there. Some cross members to help a bit with stability. It will also support projectors so it can't be floppy. Here's an image with deflection including some doofus kid trying to dunk a basketball on the cantilevers lol.

@ajh1, @HotRod10
Obviously having one solid square to insert would be easier than 4 separate plates, but wouldn't the connection have some wiggle in it due to the gap between inner and outer tubes?

@KootK
Hmmm, that sounded pretty good at first. Tighter fit than inserted tube, but still can't fully tension the bolts. And I've got anywhere from 1 to 4 members coming together at a joint, moment in all directions. Would need some crazy 3-D shapes for some of them.

 

[BridgeSmith]

"...wouldn't the connection have some wiggle in it due to the gap between inner and outer tubes?"

Not if the set screws are big enough, spaced far enough apart, and installed on two sides perpendicular to each other. Probably would require welding on a nut to be able to get the connection locked down without stripping the threads. If the nut welded outside is a no-go, then a larger number of smaller set screws threaded directly into the outer tube could work as well. Alternately, if you used a sleeve with a thick wall, it could be tapped and threaded for bolts on 2 adjacent sides, then only the washer and bolt head would be on the outside. Welding nuts inside at each end of a lighter sleeve would be ideal, but I don't know whether it's feasible to fabricate that way. Maybe someone more familiar with fab shop processes could weigh in on that.

 

[SwinnyGG]

Use conventional flanges and go to 1/2" cap screws instead of 5/8" hex head bolts. If the fabricator is reasonably attentive to detail, you can easily fit 3 1/2" SHCS per side of a 4" tube, within a 6" flange.

If you have to you can go to ASTM A574 fasteners from someone like Unbrako, if it's that challenging to get your fasterners to calc out.

If your moment requirements really are 'ridiculously low' this should be no problem; if you need more shear, you can add dowels anywhere in the flange cross sections.

 

[CANPRO]

Aaron, now I understand what you need the bolt there for - I recommend not using that. The fabricator will have their own means and methods to make the connection fit, I would show the connection you need and not worry about the extra bolt. Kudos to you for considering the fabrication process during your design.

Is this connection just to support the cantilevered beams at the far left, or is it more typical? What did you have in mind for the beam to beam connection at the end of the cantilever? Hard to tell from the deflection, but it looks like you're counting on some fixity there.

 

[CANPRO]

I see your response to kootk indicates this is a more typical connection for the structure. I think having some sort of sleeve connection on both ends of the beam could make the erection a bit tricky. Are the anchor bolts cast in place? The structure is going to be rigid as you erect it due to the moment connections, so as you progress down the line standing steel, you're going to be trying to drop a column over anchor bolts and move it laterally into a sleeve at the same time - the longer the sleeve and longer the anchor bolts, the more difficult or impossible this will be. And you obviously can't install the beam inbetween two fixed columns if you need to sleeve both sides.

Since your moments are so low, would it be possible to make some bays a simple shear connection and rely more heavily on the remaining moment connections? This would make the erection easier. Alternatively, maybe you could make something like the attached work.

EDIT - I made that sketch late last night...looking at it in the morning I realize my section doesn't match my elevation view. Elevation shows a recessed plate, section shows it extending past the edge of the HSS. But you get the idea.