I'm analyzing a connection where the HSS beam is fillet welded (all around) directly the the web of the column. My inclination is that it's a shear connection, but I would like some other opinions. I thought there was a thread on this subject before but I couldn't find it.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Is this a shear or moment connection?
- Thread starter CTW
- Start date
I agree. I thought there was a thread on this also.
But right to your question: I would do some simple calculations to determine what kind of rotation can be expected in the column. If the column is robust and braced in the immediate area on the top and bottom of the beam which would preclude rotation then it will transfer some moment. If it is free to rotate, its a pinned connection.
Regards,
Qshake
![[pipe]](/data/assets/smilies/pipe.gif "[pipe] [pipe]")
Eng-Tips Forums:Real Solutions for Real Problems Really Quick.
But right to your question: I would do some simple calculations to determine what kind of rotation can be expected in the column. If the column is robust and braced in the immediate area on the top and bottom of the beam which would preclude rotation then it will transfer some moment. If it is free to rotate, its a pinned connection.
Regards,
Qshake
Eng-Tips Forums:Real Solutions for Real Problems Really Quick.
- Thread starter
- #5
There are some easy tell tale signs such as: Does the column have stiffeners at the connection? Are there any doublers plates? Does the design drawings has any indications as to the connection type?
Having said that, you can go about this one of two ways:
1. If I were you, I would assume it is shear connection. This is conservative.
2. If you are at a point where member stresses are exceeding the allowable or you want to justify increasing load-carrying capacity, you can assume a moment connection. However, you must define the end fixity in percentages terms. If the flanges were full pen welded, then you would have a full moment connection. If it is not, you are in no man land! Your guess would be as good as anyone else. There are mathematical procedures that you may want to follow to determine a percent fixity; but is it worth it? I would say no. Plus you have to evaluate the column and see if it is adequate for the moment etc.
Be conservative like me and assume a pinned end condition.
Good luck
Having said that, you can go about this one of two ways:
1. If I were you, I would assume it is shear connection. This is conservative.
2. If you are at a point where member stresses are exceeding the allowable or you want to justify increasing load-carrying capacity, you can assume a moment connection. However, you must define the end fixity in percentages terms. If the flanges were full pen welded, then you would have a full moment connection. If it is not, you are in no man land! Your guess would be as good as anyone else. There are mathematical procedures that you may want to follow to determine a percent fixity; but is it worth it? I would say no. Plus you have to evaluate the column and see if it is adequate for the moment etc.
Be conservative like me and assume a pinned end condition.
Good luck
Lufti - I agree with you on the beam design - assuming a pinned connection. But with any type of all around weld there is some fixity and it is also conservative to assume a fixed connection for the column design as it would then have to take an end moment and this reduces column capacity. Agree?
- Thread starter
- #8
No stiffeners or doubler plates on the column and no other information to indicate original design intent, so my initial thoughts were that the connection was pinned. But then my thought process was similar to JAE's last post in that there is some fixity since it's welded all around.
A lot of work, I'm sure, for just a beam/column connection. You could just be overly conservative and
1. Design the beam as though it is fully pinned (gives you maximum moment at midspan),
2. Design the beam as though it is fully fixed (gives you maximum moment at the column end), and
3. Design the column as though it is fully fixed (gives you maxiumum moment at the end combined with axial).
By doing all three you'd cover all the bases - the true "reality" is that its somewhere in-between.
Its pretty difficult to get a connection rigidity and then model that value in most analysis programs.
1. Design the beam as though it is fully pinned (gives you maximum moment at midspan),
2. Design the beam as though it is fully fixed (gives you maximum moment at the column end), and
3. Design the column as though it is fully fixed (gives you maxiumum moment at the end combined with axial).
By doing all three you'd cover all the bases - the true "reality" is that its somewhere in-between.
Its pretty difficult to get a connection rigidity and then model that value in most analysis programs.
I think the HSS welded to the column acts as a moment connection as you would develop a couple between the top and bottom flanges of the HSS beam.
Don't forget also to check if the column flanges need stiffeners. If the column flanges are too flexible, the stresses in the welds joining the HSS flanges to the column would be very high in the vicinity of the column web, potentially leading to failure.
The steel code addresses this under the moment connections section.
Don't forget also to check if the column flanges need stiffeners. If the column flanges are too flexible, the stresses in the welds joining the HSS flanges to the column would be very high in the vicinity of the column web, potentially leading to failure.
The steel code addresses this under the moment connections section.
The joint will transfer moment to the extent of its moment capacity. Welding of the flanges to the column will introduce moment capacity to the joint.
I presume that the joint is a part of a frame.You can analyse the frame considering the respective stiffness of the members and arrive at the moment carried by the beam at the junction.
With the fillet weld on both the flanges, the moment carrying capacity of the joint can be assessed for the provided weld thickness. If the moment from the frame analysis is less than the capacity, the entire moment will be transferred to the column. Otherwise, the joint will transfer the moment up to its capacity and the excess moment will cause rotation of the joint thereby cause increase in the moment in the span of the beam.
However, you need to stiffen the column web on the other side of the column in line with the flanges of the beam to resist the forces transferred by the flanges.
The stiffness will be dependant on the distance along the column web from the welded edge of the beam flange to the root fillet of the column flange.
This area of the column web may also be subject to some severe stresses and should be checked if your beam loads change frequently. If the beam carries rapidly changing (dynamic) load then the connection detail is not good.
This area of the column web may also be subject to some severe stresses and should be checked if your beam loads change frequently. If the beam carries rapidly changing (dynamic) load then the connection detail is not good.
The Connection would be somewhere between a moment connection and a pinned connection. It is not truly one or the other. My approach would be exactly as JAE suggested and I would consider the three worst case scenarios to cover all bases. If the system can handle these extreme cases, then the design is adequate.
On a similar note, how would one model a truss where the chord and diagonal members are made out of tubes (HSS Members) and are welded together in a similar manner: all around? I always model them as pinned truss connections. Does anyone see any problems with this approach?
Thanks,
JS.
On a similar note, how would one model a truss where the chord and diagonal members are made out of tubes (HSS Members) and are welded together in a similar manner: all around? I always model them as pinned truss connections. Does anyone see any problems with this approach?
Thanks,
JS.
- Thread starter
- #18
- Status
Similar threads
- Question
- Question