Column Design in Post-tensioned garage

[mrengineer]

I am doing design of 2-story post-tensioned garage, using ADAPT-PT for the beams. I am using RISA-2D for frame analysis to obtain lateral load moments for ADAPT and for column loads. I am coming up with huge column moments in the RISA runs, and it is apparent that I need to take advantage of the secondary effects of post-tensioning to reduce the column moments. My question is how do I do that exactly? Do I take the end moments from section 9.7 of the ADAPT output (what they refer to as "balanced moments") and then input those into the RISA frames as beam end moments? Any advice on how to approach the column design and reduce the moments would be greatly appreciated. I am currently using 24"x24" columns, and hate to have to go back to the architect at this point and tell them I need bigger columns.

 

[haynewp]

Are you running fixed top or fixed top and base?

 

[mrengineer]

Haynewp:
I am not quite sure what you mean by "fixed top" if you are referring to my RISA runs. In the RISA runs, I am pinning the supports. However, predictably, the worst case columns as far as moment occur at top level, exterior column anyway.

 

[JAE]

I would think that you would get high moments with a moment frame/PT beam system since most PT garages have pretty long bays (60' is typical) which means fewer columns and lots of lateral resistance going through the columns in bending.

My first reaction would be to look at shearwalls instead of a moment frame. That way, the moments in the columns are sourced only from gravity loads.

Stair towers in the corners and interior wall lines (between the 60' bays) are the usual places for the walls.

Another thought would be to reduce your spacing of your exterior columns aroudn the perimeter to add columns and reduce the total moment in each singular column.

 

[MikeT14]

mrengineer,

We are using essentially the procedure you described, taking the PT moments from ADAPT and adding/subtracting them from the appropriate column to get the "design" moments for the columns. We are using RAM and STAAD for the modeling and designing the PT beams and slabs in ADAPT-PT.

As for modeling the fixity of connections, thier fixed everywhere except at the foundations in both the STAAD/RAM models and ADAPT.

 

[dik]

You also might want to look at a live loading of 15 or 20 psf and use this for elastic end moments with the strand at the top of the kern, also have used long rectangular columns to minimise the end moments... 14-16x32... even with that you can be approaching 4% steel in columns.