jdgengineer
Structural
We have another project with a multilevel basement that I was hoping to get some ideas on. We are in the process of our schematic design stage. I know I started a thread awhile back with similar questions, but wanted to revisit it as our layout is a little different and I felt we were somewhat reinventing the wheel for something that I imagine should be fairly common.
See attached for general layout of the proposed lower basement, mezzanine basement level, and ground floor level. Our initial thought is a thick concrete matslab for the basement foundation (geotech recommendations) and concrete topped pandeck for the mezzanine and ground floor level. The 2nd floor and roof level would be mainly wood framed with steel beams as necessary for the longer spans.
It is standard practice in our area to design single story retaining walls as cantilevered walls from the basement matslab (i.e. not pinned at the floor) and require that backfilling be completed prior to framing the floor. However, with the depth of the basement I don't think that will work well for us and instead will need to brace at the floor levels. In looking at the layout of the mezzanine level, I think there are two many openings to consider this an effective bracing level. Therefore, we are planning on designing the wall to clear span from the lowest level to the ground floor (~25'). To make matters worse the local building requirements for this project do not allow the basement wall to be drained and therefore we have to design for undrained conditions. Our geotechnical report recommends 85 PCF + 10H uniform for the non-yielding condition.
Running fairly quick numbers, I obtain a service level reaction of approximately 11k / ft at the ground floor level assuming I have a pinned-pinned connection for the wall. With a fixed-pinned I have a service level reaction of approximately 7 k / ft. Our wall is looking to be about 30" thick to resist shear and bending moment at the base.
1) How are walls like this commonly braced at the top for such a high reaction per foot? Part of me wants to make the mezzanine and ground floor level cast in place concrete in lieu of pandeck but I know the contractor really wouldn't like that.
2) On our past multi-level basement we were able to strut pretty effectively across the basement with the metal deck and attempted to design the topping slab (6" thick) for the compression force. In our case here, I think the geometry of the basement and openings due to lightwells and stairs makes this more difficult. Therefore, if we went with pandeck we were thinking of adding steel beams at say 10' o.c. to try and brace the wall. An initial thought was to cut in bearing seats in the wall for the steel beams and weld on end plates which could then be grouted tight to the wall to transfer all the lateral reactions in bearing instead of bolt shear.
3) With such a thick wall, how effective to you think fixity at the base would be? Our matslab would likely be 24-36" thick.
Any thoughts on a schematic structural solution?
See attached for general layout of the proposed lower basement, mezzanine basement level, and ground floor level. Our initial thought is a thick concrete matslab for the basement foundation (geotech recommendations) and concrete topped pandeck for the mezzanine and ground floor level. The 2nd floor and roof level would be mainly wood framed with steel beams as necessary for the longer spans.
It is standard practice in our area to design single story retaining walls as cantilevered walls from the basement matslab (i.e. not pinned at the floor) and require that backfilling be completed prior to framing the floor. However, with the depth of the basement I don't think that will work well for us and instead will need to brace at the floor levels. In looking at the layout of the mezzanine level, I think there are two many openings to consider this an effective bracing level. Therefore, we are planning on designing the wall to clear span from the lowest level to the ground floor (~25'). To make matters worse the local building requirements for this project do not allow the basement wall to be drained and therefore we have to design for undrained conditions. Our geotechnical report recommends 85 PCF + 10H uniform for the non-yielding condition.
Running fairly quick numbers, I obtain a service level reaction of approximately 11k / ft at the ground floor level assuming I have a pinned-pinned connection for the wall. With a fixed-pinned I have a service level reaction of approximately 7 k / ft. Our wall is looking to be about 30" thick to resist shear and bending moment at the base.
1) How are walls like this commonly braced at the top for such a high reaction per foot? Part of me wants to make the mezzanine and ground floor level cast in place concrete in lieu of pandeck but I know the contractor really wouldn't like that.
2) On our past multi-level basement we were able to strut pretty effectively across the basement with the metal deck and attempted to design the topping slab (6" thick) for the compression force. In our case here, I think the geometry of the basement and openings due to lightwells and stairs makes this more difficult. Therefore, if we went with pandeck we were thinking of adding steel beams at say 10' o.c. to try and brace the wall. An initial thought was to cut in bearing seats in the wall for the steel beams and weld on end plates which could then be grouted tight to the wall to transfer all the lateral reactions in bearing instead of bolt shear.
3) With such a thick wall, how effective to you think fixity at the base would be? Our matslab would likely be 24-36" thick.
Any thoughts on a schematic structural solution?
