strengfl
Structural
- Apr 1, 2010
- 2
I have a single story building with light gauge roof trusses (58'-8" span; 3/12 pitch; 4'-0" o.c.) bearing on a 12'-8" CMU wall. The FBC allowable deflection for the span would be 2.9" to 3.9" for LL or DL+LL respectively. I modeled a test light gauge truss and came up with a max deflection of 2.1". In my opinion, when you have a longer span member that will have a fair amount of deflection you must consider the effect of the bearing point rotation, due to deflection in the span, on the bearing wall. I see that I have two options on how to look at this situation.
The first option is to allow the bearing point to rotate which would result in the truss reaction having to be applied to the edge of the bearing wall (i.e. point of rotation). This would force us to use a load eccentricity of half of the wall thickness and add the resulting moment to the wind load moment on the wall.
The second option would be to consider the rotation restrained by the truss anchor. The rotation would be restrained by a moment couple produced from the tension force in the truss tie-down and the compression force at the edge of the wall. In this condition we would have to add this moment to the top of the wall as we would actually have a semi-ridged connection. This problem with this analysis is that you would have to know the rotational stiffness of the truss bearing node. Since we don't have any truss design information at the time the engineer of record designs the building, not to mention I don't think that the delegate truss engineer would be able to provide this information without substantial addition effort on his part, I don't see that we could ever properly consider this option in the design of the wall.
In my opinion the actual condition would be somewhere between these two cases. Some small rotation due to the deflection caused by the truss weight, and possibly the weight of the deck depending on the sequence of construction, and then restrained rotation due to the rest of the DL and LL applied after the connection is made.
In my case, if I were to apply the DL+LL reaction at the inside face of the 8" CMU wall due to the rotation, I get a moment of about half of the moment due to wind load on the face of the wall so this is a substantial contributor to the wall stresses.
I could use a bearing plate that was half of the wall thickness and locate it so that one face of the plate aligned with the outside face of the wall while the other face of the plate aligns with the centerline of the wall. This would put the center of rotation in line with the center of the wall and eliminate any eccentricity of the reaction.
Unfortunately I think that I still cannot properly design the connection at this time due to the fact the there are at least three different types of truss chord profiles that are commonly produced in my area, quite possibly more, and some with such unique shapes that the anchors used to attach them are proprietary to the manufacturer. My best bet may be to provide my required reaction in all directions for the main truss types that I will have on my project (along with the FBC load cases to consider, unity equation, etc.) and have the truss manufacture submit his suggested anchorage for my review and approval once the truss manufacture is chosen by the contractor. On this project I will have enough control during the submittal process that the final order of the trusses won't be made without my review and approval. This is not so easy with all projects.
I have not been able to find any literature on this specific topic nor do I think the delegate truss engineer's software will consider the reduced bearing area based on the end node rotation as it relates to increased bearing stress on his bottom chord.
Any thoughts would be appreciated.
The first option is to allow the bearing point to rotate which would result in the truss reaction having to be applied to the edge of the bearing wall (i.e. point of rotation). This would force us to use a load eccentricity of half of the wall thickness and add the resulting moment to the wind load moment on the wall.
The second option would be to consider the rotation restrained by the truss anchor. The rotation would be restrained by a moment couple produced from the tension force in the truss tie-down and the compression force at the edge of the wall. In this condition we would have to add this moment to the top of the wall as we would actually have a semi-ridged connection. This problem with this analysis is that you would have to know the rotational stiffness of the truss bearing node. Since we don't have any truss design information at the time the engineer of record designs the building, not to mention I don't think that the delegate truss engineer would be able to provide this information without substantial addition effort on his part, I don't see that we could ever properly consider this option in the design of the wall.
In my opinion the actual condition would be somewhere between these two cases. Some small rotation due to the deflection caused by the truss weight, and possibly the weight of the deck depending on the sequence of construction, and then restrained rotation due to the rest of the DL and LL applied after the connection is made.
In my case, if I were to apply the DL+LL reaction at the inside face of the 8" CMU wall due to the rotation, I get a moment of about half of the moment due to wind load on the face of the wall so this is a substantial contributor to the wall stresses.
I could use a bearing plate that was half of the wall thickness and locate it so that one face of the plate aligned with the outside face of the wall while the other face of the plate aligns with the centerline of the wall. This would put the center of rotation in line with the center of the wall and eliminate any eccentricity of the reaction.
Unfortunately I think that I still cannot properly design the connection at this time due to the fact the there are at least three different types of truss chord profiles that are commonly produced in my area, quite possibly more, and some with such unique shapes that the anchors used to attach them are proprietary to the manufacturer. My best bet may be to provide my required reaction in all directions for the main truss types that I will have on my project (along with the FBC load cases to consider, unity equation, etc.) and have the truss manufacture submit his suggested anchorage for my review and approval once the truss manufacture is chosen by the contractor. On this project I will have enough control during the submittal process that the final order of the trusses won't be made without my review and approval. This is not so easy with all projects.
I have not been able to find any literature on this specific topic nor do I think the delegate truss engineer's software will consider the reduced bearing area based on the end node rotation as it relates to increased bearing stress on his bottom chord.
Any thoughts would be appreciated.
