concrete column loads - worst case 1

[KootK]

I'm working on a six story concrete building (beams & columns with pan joists). I'm having difficulty coming up with a rational and efficent method of determining the worst case design loads for my columns(satifying ACI 8.8). Here's what I'm thinking of doing:

1. Full L + D all the way down.
2. Partial L + D patterns all the way down (same patterns on all floors).
3. Partial L + D alternating patterns as I go down (different patterns on alternating floors).
4. Same as (3) but switching which floors get which patterns.

This is already a substantial undertaking. In addition, I'm wondering if I need the consider the following:

1. Same as above but with no live loads applied above the floor under consideration. Perhaps this could produce the loading with the greatest moment/axial ratio (ACI 8.8.1)?

2. I must, of course, do a lateral analysis. Am I required to peform each of the combinations above in conjunction with each of my lateral load cases? I'm examining wind from four directions. Each direction comes with 4 load cases for different combinations of L+D+W. So far then, I've got 4x3=12 load cases to consider for lateral + full live and dead load. Now...do I have to consider each individual load pattern in combination with the lateral loads??? That would yield 12x4 = 48 separate load cases!! I think that there must be a better way.

3. How do I apply live load reduction factors to the column moments? In a moment frame structure it is diffcult to determine the corresponding loaded area that produces each moment. Especially when there are different magnitudes of floor loading applied to various areas of the building.

Can anyone recommend a simpler way for me approach my column design. I am using RISA 3D for the structural analysis. Thanks in advance for your help!

 

[JAE]

AdamP - you are faced with the challenge to any engineer who designs concrete per ACI and tries to meet the full intent of 318.

I believe that your high number of load cases are technically required. In fact, if you are using the moment magnification procedure in ACI, you have to develop deltas for each load combination separately.

Patterned loads:
I think your concept of alternating different patterns to get at a maximum moment is correct. In 3D you have alternating floor sequences as well as checkerboard or adjacent loads on each level to contend with. I would probably try to work through the different possible patterns and focus on isolating those patterns that would, by observation, yield the highest moments.

Live Load Reduction:
In RISA - your combinations should include your standard load factors of 1.7 or 1.3 for live loads. You can calculate the live load reduction percent allowed for each level of columns and sequentially apply this reduction onto your load factors and run the program for each level of column. So lets say the top floor gets a live load reduction of 25%, the next floor down gets 50% and thereafter the columns get 60% (the max.)
You would first analyze for the top floor by applying a factor on ALL model live loads of: 1.7 x 0.75 = 1.275 and run the frame - get your Pu and Mu values for each combination and each column and then design using PCACOL or by hand or by spreadsheet.

Then, you would apply a live load factor of 1.7 x 0.5 = 0.85 to ALL live loads in the model combinations and re-run, getting Pu and Mu for the next floor....etc.

Tedious? Yes, but that's ACI. It also shows why there aren't that many programs out there that successfully do concrete design for column/frames. I believe PCA has one but I'm not familiar with it and not sure that it correctly applies delta factors, live load reduction factors and alternating loads. RISA is apparently heading toward a concrete package later this year - will be interesting to see if they meet ACI in these terms.

 

[KootK]

Thank you very much JAE. That's not the answer I was hoping for but I suppose that I'd better just buck up and begin the dirty work. I spoke to RISA about their concrete module last week. It's supposed to come out in september now. We'll see. For any other RISA users out there, here's a tip that I learned on this project: The "TOP OF MEMBER" feature does not work in conjuction with the use of diaphragms. Yields screwy results. Don't get me wrong though, I still love the program.