Structural Skid Design

One engineer's skid is another engineer's frame. We need a more complete description of your structure, preferably, a sketch.

Michael.
Timing has a lot to do with the outcome of a rain dance.

I have attached a sketch. Both ends are 100% full pen welded and the beams are coped together.

Since both are fixed ends, is this particular beam statically indertminant?

What is the best method for determining deflections.

Thanks

I would consider the beam hinged at each end, hence statically determinate. Although the ends are fully welded, the exterior beams cannot prevent rotation.

If the skid rests on flat ground, the beam cannot deflect as it is the same depth as the exterior beams, so the statics are not entirely clear.

BA

The skid will be supported with 4 point supports at the four corners.

The length of the skid will not be supported by the ground.

Thanks

If the structure is supported only at the four corners, then the exterior beams spanning 26' will be more critical than the cross beams. You could consider shifting the four supports away from the corner to reduce deflection of the edge beams.

In any case, the cross beams are statically determinate with a hinge at each end.

BA

Thanks a lot! I will assumed hinged.

Just one more question for my curriosity.....Why cant the exterior beam prevent rotation if they are welded and the ends of this beam are tied down?


You have been a big help!!!

Neglecting the stiffness of the four columns, the edge beams will rotate as dictated by the end slope of the cross beams. If all cross beams are identically loaded, the slope will be the same at every connection so the edge beam rotates as a rigid body.

If the cross beams have different loads their end slopes will be different and the edge beam will be required to twist to accommodate them all. So when the beams are not simultaneously loaded, the torsional stiffness of the edge beam comes into play, making each cross beam indeterminate to a slight degree.

For your purposes, it is conservative to consider all cross beams as equally and simultaneously loaded.

BA

engpes:
WOW, paddington had no idea what he was asking for; I’d call this at least an engineer’s frame and a half, and a very complicated sketch at that.

Who determined that any of the welds have to be 100% full pen. welds? You could save enough in coping, joint preparation, fit-up, positioning and welding consumables and time, to make this whole thing out of 24 or 36W beams; if the detailing and welding were actually designed to meet load conditions and real structural needs. Contrary to popular opinion, in some quarters, it is not true that all the strength will drain out of the ends of beams if they are not full pen. welded.

I agree with BA, these are basically simple beams, partially fixed at their ends, but only to the extent that the beams they frame into will offer a little, very little, torsional resistence to cause a small end moment. So, design and size them as simple beams. There is probably no advantage, and maybe a disadvantage, in all the main beams being the same size and depth. And, you might, well consider moving the support columns to more ideal locations, or add more columns if you can.

So far you have produced a CAD sketch of a frame which pays particular attention to the locations of various columns and drums and their mounting holes. Now you have to explain to us, if you want our help; or yourself if you might stand under this frame or be responsible for it standing up, what your real design criteria is. Maybe you have some inches out of plum ± criteria for the contactor and flash drum, however tall they are, and thus you are interested in the deflection at their base. But, you may better control that by where you place supports, that’s what BA was getting at, not by some final point defections from a column 10' away, over several routes, through several different beams. Maybe your framing system isn’t that good for what you need to accomplish. Maybe it should be four 26' long beams; two under the surge drum/stripper col. and two under the contactor/flash drum, etc. Then you’ll still probably need a computer frame program to take a meaningful shot at the deflections at specific framing joints. I don’t think wondering if one of these beams is fixed-fixed and then applying Roark’s Formulas will really solve your problem. You need a deeper understanding of the whole problem.

I suggest that you consider the frame pinned at the lifting lugs. Unless your loads are extreme, I would not consider that CJP welds are necessary.

Ron,

The location of the lifting lugs have nothing to do with it. And whether or not the loads are extreme, CJP welds are not necessary.

BA