Stiffener Retrofit

[JAX91]

I am designing a retrofit for a steel pipe rack built in the late sixties or early seventies. The pipe rack was designed (or likely built without being designed) in a stacked fashion. The bottom level beam sits on top of the columns. The next level column sits on the beam. The beam-on-column, column-on-beam repeats for three levels. I have attached a picture for reference. At the fourth and top level, the beam finally frames into the side of the column. The beams are either W6x15.5 or W8x17, and the columns are W6x15.5. All of the connections are fully welded with fillet welds all-around. There are no stiffeners at any connections, and you can see each level is leaning one way or the other because the top flange of the beams are bending. I obviously need to add stiffeners to the beams to prevent them from rotating. I am currently considering three versions of stiffeners.

1. I will first check the beam flange as if it were the column side of a moment connection. If stiffeners are required, I will add a pair of stiffeners under each column flange.
2. If stiffeners are not required for the moment connection, I will check the beam web for stiffener requirements for the vertical load. If stiffeners are required, I will place a single pair of stiffeners centered under the columns.
3. If stiffeners are not required for the moment connection or for the vertical load, I still need rotational restraint. I think a cap plate welded to the end of the beam would be sufficient. I would only be able to get weld on the inside of the beam flanges and the beam web because the beams has a column on the outer sides of the flanges. For additional stiffness, I could weld the cap plate to the outer faces of the column flanges.

As is the case with any design, I would like to go with the most cost-effective design while still making it structurally sound. Option 3 would be the most cost effective, followed by Option 2, and then Option 1. Does anybody see issues with any of the three options? With Option 1, do you think I could put a pair of stiffeners under the inner column flange and use a cap plate on the beam to act as a stiffener for the outer column flange?

 

[BAretired]

Where columns are discontinuous, they should be made continuous by adding a stiffener each side under each flange, same thickness as column flange and fillet welded both sides, cut at 45 degrees to miss the radius.

BA

 

[TLHS]

If this thing's loaded during construction, I'd think about what your welding is doing to it during the retrofit process. You're heating up the weakest part of the structure, so you'll likely have some temporary strength loss. You may be fine, but depending on how close to capacity you are under gravity loads it may be an issue.

 

[vagarwalstr]

Structure should be made plumb before stiffners are put in. Also, if structure was never designed, will it be stiff enough for wind or seismic loads, even if columns are made continuous using option 1. Column and beam strengthening may be required.

 

[JAX91]

Thank you all for your responses. I will be installing the stiffeners under each flange per BA's recommendations. As I mentioned in my original post, the top beam frames into the side of the columns. I will also need to add stiffeners at this connection because of flange local bending. I think the rack will be okay with the additional heating from the welding. It is only 12'-0" wide and has no more than 40 psf dead load per level, so it has a lot of capacity for gravity loads. I'm actually surprised that a four level rack with W6x15.5 columns is stable, but once I add the stiffeners, it passes strength and stability checks for both wind and seismic as well.