2 bolt end plate connection design

[intrados]

I'm just seeking clarification on the correct approach to designing a bolted end plate for rafter to column. I have access to the Australian Steel Institutes design guide 12 (Rigid connections) which details the Yield line analysis for open section connections where bolting is available on each side of flange. See picture of design page attached. My connection consists of a 250UB rafter connecting to top of 200x100x4.0RHS. Rafter cantilevers from top of column. Connection is via 16 end/base plates and 4/M16. As I have a hollow section for one member in connection, I am limited to only having bolts outside of RHS flanges. My first approach is to just omit any references to bolts that are inside flanges from the recommended design model calculation but I fear this is a very naive approach and would like to understand the design a little better. Is there a connection design guide available for the simpler connection detail that I have?

The design moment is near enough to 10kNm. There is certainly the provision for a stiffener plate to be installed centrally to RHS column in the same line as rafter web (rafter sits centrally over column). Although this would give me an increased sense of certainty about connection capacity, I would rather not include stiffening if not needed.

Thanks

 

[Agent666]

It might help to provide a sketch of your intended connection as I'm a bit confused as to what it might look like from the description. Is it a 250UB running over the top of the SHS with 2 bolts through the 250UB flanges either side of the RHS? Is it a canopy or something else, the 10kNm moment is very low, I suspect you need to design for the minimum design actions being (I think) 30% of phiMsx in AS4100, assuming you are of course in Australia based on the ASI reference.

Remember you need to design both the end plate onto the RHS and the UB flange and different critical yieldline mechanisms might apply to each side of the connection. Obviously the picture you posted is only one side of the connection. You can certainly design it from first principles, but simply neglecting some terms from ASI might not be valid. A useful free resource for this is the UK SCI guidance document P398. Its based on Eurocode obviously being form the UK, also refer to the latest AS/NZS2327 composite standard which borrows heavily from Eurocode 4 for the connections design portion of the standard. This new standard has a lot of guidance on this particular method in a codified procedure applicable to the NZ and AU way of doing things. You can pretty much design anything from a semi first principles approach using these documents by reviewing various yieldlines & taking the most critical one to work out an equivalent T-Stub capacity for the 3 modes of failure noted. All the additional checks normally required are also noted in quite an easy to follow process.

 

[ajh1]

AISC Design Guide 16 does have a 2 bolt condition (flush end plate, bolts inside the flange) along with a 4 bolt condition (extended end plate, 2 bolts outside of flange, 2 bolts inside of flange). You might be able to work from those if what you have is a 2 bolt condition with the bolts outside of the flange.

 

[intrados]

You’re right Agent666. A further sketch would have helped tremendously. I’ve been unexpectedly off line since shortly after making original post and hence have only recently viewed your reply. Thanks for input. I’ve attached the connection detail this time. It’s simple but overlooked in the standard designs as part of the AISC design guide 16 (and equally the Australian version which I believe is a direct copy of AISC). I’ve since familiarised myself with the yield line analysis approach and now feel comfortable to safely and economically design the connection. You’re right too about the minimum design action requirements. It’s 0.5x member capacity for rigid connections in AS4100.

Thanks too ajh1. I wish I was aware of the AISC design guides before I bought hard copies of the Australian Steel Institute design guides which is where the picture came from in my original post