Square HSS beam to Sq. HSS beam pin / shear connections 4

[GD_P]

Hello Forum,

I am dealing with structure made up of square HSS. I have few queries regarding design of pin / shear connection.
Please find the attached image of 2 options of shear connection.
In option 1, how to determine the thickness te? SCI recommend to use min thickness required to distribute loads from fin plate to welds surrounding beam web in ratio of 2.5 to 1, which is very high in my opinion. For e.g., of I have 150 wide square HSS beam, this means the thickness te has to be (150/2)/2.5 = 30 mm, which is too high irrespective of the forces to be transferred to through it. Correct me if I am wrong & suggest alternative method for determinating te.
In option 2 we have extended the plate inside the sq. HSS beam up to depth d. How to evaluate this depth? Any suggestion will be very helpful.
Regards


GD_P

 

[phamENG]

Is that a beam to girder connection or a beam to column connection? If beam to column, I would suggest rotating the plates 90 degrees. In either case, take a look at AISC Design Guide 24. It should answer most of your questions.

EDIT: Sorry - didn't read the title closely enough.

 

[DrZoidberWoop]

For Option 1) If I remember correctly, the 2.5:1 dispersion ratio was for axially loaded HSS braces/members. If you have a basic shear connection, I'd base the thickness on the required weld size of the stem plate bolted to the shear tab. If you're loading the HSS cap plate in shear, it's going to have more weld to the hss, and be quite stiff at the hss walls. Plate flexure within the HSS perimeter shouldn't be an issue.

For Option 2) Make the "d" dimension greater than or equal to the depth of your HSS member. It can be larger depending on what welds you need to transfer the shear force.

 

[human909]

Don't underestimate the effect of the eccentricity of the connection. As observed by quite a few in this forum it seems to be rife amongst many engineers to ignore this. You need 'excessively' thick plates to avoid plate buckling. (Or you could use non gussets or non eccentric connections.)

Here is some research on this matter:

http://ascjournal.com/down/vol6no2/vol6no2_2.pdf

 

[BAretired]

I don't believe the end plate t_e is more critical than the wall thickness of the supporting beam. Of the two options proposed, I prefer Option 1 but the supporting beam must be designed for torsion and its wall may need reinforcing at the connection. The end plate thickness and wall reinforcing can be determined using yield line analysis.

Perhaps a better detail would be an angle at the end of the supported beam with horizontal leg lapping over the supporting beam.

BA

 

[hokie66]

DrZ and human909 are correct. The thick plate thickness is undoubtedly for a compression strut. If it is a beam as your title suggests, not so thick.

 

[BAretired]

This is the connection I was thinking of:

It's not a perfect pin but it has less eccentricity than the proposed connection and it does not overstress the wall of the supporting HSS beam in local bending.

BA

 

[GD_P]

Dear All,

Thank you very much for your response.
Yes the 2.5:1 ratio is for bracings. AISC Design Guide 24 (DG24) also have same ratio.
Thank you @ phamENG, for telling the DG24, a complete reference for hollow section connections, really good document.
Thank you Dr, human909 & hokie66 for your comments. I was also worried about the eccentricity.
BAretirec I really appreciate your efforts, thank you. I am looking forward to use your recommendation i.e., end plate welded to supported beam which will be bolted to thick tapped plate (having higher thread shear resistance than mating bolt) welded on supporting beam. This will have very little eccentricity on supporting beam.
Thank you once again for your help.


GD_P