Old Design flaw, what to do. 4

[NinerStruct]

I recently started work on a site-adapt project for a public building that has already been used on another site. The structural engineer that did the previous design has since retired and I was tasked with the structural design of the new building. I rechecked the calcs from the previous engineer's design and did my own design when I couldn't find the original.

When redoing the design for a few columns, it appears that they were under-sized in the original design. The worst column is an HSS4x4x1/4 located in an area with a roof height of 19.5 ft, and a maximum unfactored design load of 62.5 kip. So my design says that it should be an HSS6x6x1/4" if there's no bracing provided.

The original building is over 10 years old now, and there hasn't been any failure. I'm guessing that's because a)it hasn't seen the full design load yet and b) the columns are buried in light gage walls that are likely providing some lateral bracing. Now, I don't imagine that the previous engineer had purposefully counted on non-structural walls to brace the column, but I can't find the original calcs for these. Either way, it's not something that I'm comfortable doing, so I'm changing them in the new design. Which leaves me with the question:

What should I do about the existing building with 4x4 columns already in service? Should I be notifying owner about this? Should we be coming up with a fix for these columns?

 

[Ron]

Agree with Mike and hokie66....wouldn't use a demising wall for bracing. The ceiling and in-plane wall will not provide adequate bracing. Stick to your guns and be as obnoxious as necessary to get your way.

 

[NinerStruct]

I spoke with the architect and convinced him that we need to change the columns in new building because I wasn't comfortable using the demising walls as bracing. I really wasn't too concerned with getting push-back on that aspect and didn’t receive any. So in regards to the new building, I believe we are fine.

My main concern is what to do about the existing building. I’ve been working to come up with a fix to have in-hand before speaking with the owner, but it's proving difficult. The building is presently in use and with the columns in the walls, I’ve been trying to find a solution to minimize the demolition and limit it to above ceiling work if possible.

The required braced length is 14’-6” for 4 columns, and 12’-6” for the remaining two. So, my first thought was to use angle braces off each face of the column, at the required elevations. The angles would go up to the beams/joists and be bolted to gusset tab connections at each location.

Unfortunately, in most cases there is large ductwork at the sides of the columns at the elevations that I need the bracing, leaving me with access to only two sides of each column. Is it feasible to provide T/C angle brace bolted to a gusset in two orthogonal directions and consider the column braced at that elevation? Or would I be inadvertently adding moment into the columns and doing more harm than good?

 

[hokie66]

If you can brace them in two directions, that should accomplish your objective. You don't have to brace in four directions. I would prefer using CHS or SHS sections for the T/C braces. Whether of not you can get adequate bracing by connections to joists would have to be confirmed.

 

[NinerStruct]

So the column would have a reduced braced length, but even if I use tubes for the braces, won't the brace up to the beam take some of the gravity load from the beam as the beam deflects under snow/live load? This would be an angled axial load that would then induce a horizontal force at the column. If I had a brace on the opposite side, I can see the horizontal forces balancing each other out, but with only one brace in each direction, I may have achieved the reduced the braced length, but do I now have to check the interaction equation? That will likely reduce my axial capacity rather than help me.

 

[hokie66]

Maybe, but the main thing is to reduce the unbraced length of the slender column so that it doesn't buckle. I wouldn't worry about the bending, but maybe that is just me. Will be interested to hear other opinions.

 

[CANPRO]

if you're concerned about transmitting vertical load, can you connect with a bolt and a vertical slot?

 

[hokie66]

NinerStruct,
As you have solved the ethical part of this, perhaps you should formulate a new thread in the Structural Other Topics forum.

 

[NinerStruct]

Thanks again for the help on this issue. I've started another post in forum507 to address the actual fix.