Understanding Rigid Frames (Residential)

[sithlord382]

Hi,
Just posting to gain a better understanding of the applicability and analysis of rigid frames...
I am a project manager/ developer and I am typically coming across residential properties with proposed single storey wrap around or rear extensions, where the design involves demolishing the original rear wall of the house to create an 'open plan' ground floor. I recently came across a structural design where the engineer had designed a box frame, ie. high level beam, columns on either side and ground beam (encased in concrete, to take the weight of the structure above. When I asked why this was used instead of a goal post frame (the columns being bolted into pad foundations) he said it was because the box frame had also considered lateral stability. He kindly went on and explained to me that lateral stability is the consideration of a horizontal wind load acting on the structure and he had then justified the frame by treating the high level beam as simply supported, taking only vertical loads with pinned connections on either side, The columns were analysed to be fixed (at the column-ground beam connection) acting as cantilevers with only the lateral load being applied at the end of the column. With the high level beam being simply supported with pinned connections, the moment created by the vertical loads was only checked in the deflection of the specified beam as the moment was too high and impossible to create a moment connection. The column to ground beam junction is treated as fixed which is where the frame gets it rigidity.

Now with goal post frames, can the same principal as above not be applied to them? Can you not have a pinned support at high level and a fixed connection at low level where the column is fixed into the pad foundation?

How do you know when to use a box frame or a goal post frame? Does a goal post frame consider lateral stability?
From a project manager/ developers perspective I would always prefer to use a goal post frame purely as it is easier to construct as opposed to a box frame.

Any comments would be very much appreciated.

Thanks
S

 

[jayrod12]

I'm not a huge fan of cantilever columns, which is what you appear to be describing as a box frame. mainly due to the uncertainty of teh connection at the base. For a goal post frame (or moment-frame as we would normally term it)the moment is developed at the high beam to column connection, and you can analyze the base connection for the columns as pinned which is much easier to provide.

Without running a number I can't confirm, but I feel you would also be able to end up with a smaller beam size for the high level beam when having a moment connection at the ends, as opposed to the simple span beam in that engineer's proposal. But a lot of that depends on the loading etc. At a minimum I would expect that the beam size would be the similar, but the load path would be a more confident load path.

As least that's my opinion.

 

[1503-44]

Generally you will need either fixed column to beam connections at upper levels or fixed column bases. Columns fixed at the bottom requires transfer of moment into the foundation or grade beam. The problem with that is the foundation or certainly a grade beam may not be designed for carrying that moment. If the column base is pinned, then it is just vertical load and some horizontal shear you have to transfer to ground. That is usually much easier to deal with than moment.

Contractors usually like fixed bases, because then they don't have to make big moment connections at beam to column at upper levels. Beams or columns can often be smaller, since there is less moment at the connections. But they usually just assume that their fixed columns are OK to dump all wind load created moments into the foundation, which is not usually the case. They also like to use small bolts into the concrete to do it, which won't really transfer much moment into the foundation anyway.

 

[KootK]

It is my understanding that using a low beam to provide some base rotation restraint to the columns is a commonly employed strategy on the British Isles. And I support it as more reliable way of gaining that restraint relative to trying to get it done on a strip/pad footing etc, particularly in a residential structure. Shown below are what I understand the possible permutations of this to be if one wishes to avoid attempting point rotational restraint at the bottoms of the columns. I'd expect the first two options to be approximately equal in stiffness with the third option being the best. Other factors do come into play as mentioned by my colleagues above.