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 cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Beam connection to corner of square tube column 2
- Thread starter IFRs
- Start date
IFRs,
I've used a single plate welded to the column, like a single plate shear tab on a W column. I don't know if its aisc recommended or not. Also, I've used this on smaller beams, but I bet you could make that detail so it works.
regards,
chichuck
I've used a single plate welded to the column, like a single plate shear tab on a W column. I don't know if its aisc recommended or not. Also, I've used this on smaller beams, but I bet you could make that detail so it works.
regards,
chichuck
civilperson
Structural
A 3/8" plate inserted in a slot of the HHS and then welded to the walls could extend out for bolts or weld to web of W12.
-
2
- #4
- Thread starter
- #6
msquared48
Structural
If you are going to have a slotted connection, I would slot the opposing corners of the tube and extend the plate through the center of the column, welding the plate to the tube at both corners of the column.
I would consider a bearing seat for erection purposes only, but not the top flange connector, as the connections at the top and bottom could impart unwanted moments to the column - unless you are trying to make a moment connection.
Mike McCann
MMC Engineering
I would consider a bearing seat for erection purposes only, but not the top flange connector, as the connections at the top and bottom could impart unwanted moments to the column - unless you are trying to make a moment connection.
Mike McCann
MMC Engineering
- Thread starter
- #8
More information: There are multiple beams coming into this column. Two are 90 degrees apart and the third comes into one of the free corners not at 45 degrees but about 10-15 degrees off from 45. I don't think I can slot through. Maybe I can burn a square hole in 2 thick plates, slide them onto the column, weld them in place and bolt the beams into them. I'll try to make a picture...
msquared48
Structural
Generally, I would take the heaviest loaded connection and slot thru.
Mike McCann
MMC Engineering
Mike McCann
MMC Engineering
This is a Q/A from AISC concerning through plates:
[blue]What is supposed to happen when you need through plates in each direction for a tube or pipe connection? It seems to me that you wouldn’t get a lot of benefit out of just running the plate with the lesser load back into the other plate and butting against it.
Answer(s)
09/01/2000
It’s an amusing detail to say the least. Unless you’re at the end of the HSS, I suppose you need a really short welder to crawl down into the HSS!
Seriously, though, the AISC HSS Connections Manual covers both single-plate (shear tab) connections, which are made with a plate welded directly to the face of HSS (i.e., without a slot), and their more expensive cousin, through-plate connections, which are made with a plate that penetrates the face of the HSS. It is expressed in that publication that the conventional single-plate connection should be used whenever possible. The design procedure covers all limit states, including those to ensure that the column face is adequate without the need to extend the plate through the wall and make it a through-plate connection.
Through plates are expensive because they are difficult to fabricate, so avoid them if you can. And you almost always can. Avoid them especially when opposite through plates would intersect each other inside the HSS. If the plate isn’t continuous to the other HSS wall, I can’t see what good it does to extend it inside. It may actually do harm since you would have to hack up the wall of the HSS to make that kind of connection.
Charles J. Carter, SE, PE
American Institute of Steel Construction
Chicago, IL[/blue]
[blue]What is supposed to happen when you need through plates in each direction for a tube or pipe connection? It seems to me that you wouldn’t get a lot of benefit out of just running the plate with the lesser load back into the other plate and butting against it.
Answer(s)
09/01/2000
It’s an amusing detail to say the least. Unless you’re at the end of the HSS, I suppose you need a really short welder to crawl down into the HSS!
Seriously, though, the AISC HSS Connections Manual covers both single-plate (shear tab) connections, which are made with a plate welded directly to the face of HSS (i.e., without a slot), and their more expensive cousin, through-plate connections, which are made with a plate that penetrates the face of the HSS. It is expressed in that publication that the conventional single-plate connection should be used whenever possible. The design procedure covers all limit states, including those to ensure that the column face is adequate without the need to extend the plate through the wall and make it a through-plate connection.
Through plates are expensive because they are difficult to fabricate, so avoid them if you can. And you almost always can. Avoid them especially when opposite through plates would intersect each other inside the HSS. If the plate isn’t continuous to the other HSS wall, I can’t see what good it does to extend it inside. It may actually do harm since you would have to hack up the wall of the HSS to make that kind of connection.
Charles J. Carter, SE, PE
American Institute of Steel Construction
Chicago, IL[/blue]
Here’s another Q/A:
[blue]The HSS Connections Manual indicates that a single-plate shear connection can be used if the following HSS wall slenderness ratios are satisfied:
1. Rectangular tubes: b/t < 253/Fy0.5
2. Round tubes: D/t < 3300/Fy
Are these limitations alone enough to satisfy the requirements or must one also evaluate the wall for punching shear resistance? Do the HSS requirements also apply to ASTM A53 steel pipe? How about to large-diameter round and large-sized rectangular HSS?
Answer(s)
As noted in the HSS Connections Manual, the slenderness limits assure that local distortion caused by the shear tab will be insignificant in reducing the column strength of the HSS. However, even when the slenderness limits are met, there are HSS wall limit states that should be checked. See Table 4-2 for the applicable limit states (shear at weld and punching shear.) Table 4-3 gives the rationale behind the limit states.
The HSS Specification (Section 1.2.1) specifically lists ASTM A53 steel pipe as approved for use under the HSS Specification. As such, the HSS equations are applicable to ASTM A53 steel pipe. The HSS Specification also applies to the limits of ASTM A500—a maximum periphery of 64”. For example, a rectangular HSS 20x12 is right at the limit. Also keep in mind that as you get bigger, it does become more difficult to meet the slenderness requirements about which you asked.
Keith Mueller, Ph.D.
American Institute of Steel Construction [/blue]
And another:
[blue]
Which is more economical to specify—shear tabs or through-plates when connecting a wide-flange beam to a HSS column?
Answer(s)
Conventional shear tabs will almost surely be less expensive than through-plates. The main issue is the cost associated in making through-plate connections compared to using single-plate shear connections on both sides of a tubular column. The best way to confirm this is to ask your fabricator for a comparison quote. In general, through-plate connections are not the preferred shear connections for tubular columns because of fabrication cost. Note that through plates may become necessary if the HSS wall is slender or if the punching shear limit state requirements are not met.
Sergio Zoruba, Ph.D.
American Institute of Steel Construction, Inc[/blue]
[blue]The HSS Connections Manual indicates that a single-plate shear connection can be used if the following HSS wall slenderness ratios are satisfied:
1. Rectangular tubes: b/t < 253/Fy0.5
2. Round tubes: D/t < 3300/Fy
Are these limitations alone enough to satisfy the requirements or must one also evaluate the wall for punching shear resistance? Do the HSS requirements also apply to ASTM A53 steel pipe? How about to large-diameter round and large-sized rectangular HSS?
Answer(s)
As noted in the HSS Connections Manual, the slenderness limits assure that local distortion caused by the shear tab will be insignificant in reducing the column strength of the HSS. However, even when the slenderness limits are met, there are HSS wall limit states that should be checked. See Table 4-2 for the applicable limit states (shear at weld and punching shear.) Table 4-3 gives the rationale behind the limit states.
The HSS Specification (Section 1.2.1) specifically lists ASTM A53 steel pipe as approved for use under the HSS Specification. As such, the HSS equations are applicable to ASTM A53 steel pipe. The HSS Specification also applies to the limits of ASTM A500—a maximum periphery of 64”. For example, a rectangular HSS 20x12 is right at the limit. Also keep in mind that as you get bigger, it does become more difficult to meet the slenderness requirements about which you asked.
Keith Mueller, Ph.D.
American Institute of Steel Construction [/blue]
And another:
[blue]
Which is more economical to specify—shear tabs or through-plates when connecting a wide-flange beam to a HSS column?
Answer(s)
Conventional shear tabs will almost surely be less expensive than through-plates. The main issue is the cost associated in making through-plate connections compared to using single-plate shear connections on both sides of a tubular column. The best way to confirm this is to ask your fabricator for a comparison quote. In general, through-plate connections are not the preferred shear connections for tubular columns because of fabrication cost. Note that through plates may become necessary if the HSS wall is slender or if the punching shear limit state requirements are not met.
Sergio Zoruba, Ph.D.
American Institute of Steel Construction, Inc[/blue]
If the expected load from beam at the connection is shear only welding shear plate onto the rectangular/square column surface is the most logical and economical as advised. However the eccentricity on the column has to be considered adequately.
In case the connection attracts bending moments and axial force, with/without the column is having the similar actions on the other sides you may need to think about how the load paths are to be. Sometimes you may be forced to cut off the connected area of the column and replace with properly designed column piece integrated with connections. When this is required and concrete filling is essential, you need pay attention to any detail.
Load path at the connections which includes the plate surfaces of columns are very important, whoever selects the connection and make decisions on the entire system this has to be the focus at front.
Hope it helps
Ibrahim Demir
In case the connection attracts bending moments and axial force, with/without the column is having the similar actions on the other sides you may need to think about how the load paths are to be. Sometimes you may be forced to cut off the connected area of the column and replace with properly designed column piece integrated with connections. When this is required and concrete filling is essential, you need pay attention to any detail.
Load path at the connections which includes the plate surfaces of columns are very important, whoever selects the connection and make decisions on the entire system this has to be the focus at front.
Hope it helps
Ibrahim Demir
- Thread starter
- #15
I do appreciate all the comments and suggestions. What we have is a 6 x 6 x 1/4" column with (3) W12 x 16 beams framing into it. Looking down from the top, the column sides are north-south-east-west, beams come in from the north and west and the third beam comes in from the south-west directly into the corner at 45 degrees. The top of the beams are flush with the top of the column. We welded double clips for the north and west beams to grab the W16s in double shear. We cut slots into the south-west and north-east corners of the column and welded a 1/2" thick plate through the column to connect the beam in single shear. Does this seem adequate and sufficient?
- Thread starter
- #16
One correction/amplification to the above post. In the same column, one set of beams joins in the middle of the column and another set joins flush with the top of the column. The image I uploaded shows the lower connection. the top connection is the same but the column ends at the tops of the beams.
Top of column seems to be critical due to the local load path and from the flexible double shear connections. The expected flexibility comes from the wall flexibility under the rotation of the beams and this cause some deflections on the wall. However the continuity ends at the top of the column in your application with the same elevation of the beams, you may not be able to control deflection of the wall with open section. This may cause fracture starting at the top in the middle of the wall.
You need extend the column leave some distance ( at around width of the wall ) to get this flexibility required and close the top of the column with a plate. Or you need to design the beam depth not to introduce rotations ( or introduce for very small rotation) at end connections. This option might be expensive due to the large depth calculated for the beams.
The flexible end connections were given for a limited rotation at connections, and this rotation needs to be checked for each application normally. However, I often see that they are mostly ignored by engineers unfortunately.
While the connection rotates it pulls the top of the connection and push the bottom into the column and the wall change the shape, deflects within the yield criteria (yield line analysis), and restrained by the perpandicular walls. Therefore, you need to control stresses within the hook law or allow some yielding. All these are the design issues you need to go through.
Hope it helps.
Ibrahim Demir
You need extend the column leave some distance ( at around width of the wall ) to get this flexibility required and close the top of the column with a plate. Or you need to design the beam depth not to introduce rotations ( or introduce for very small rotation) at end connections. This option might be expensive due to the large depth calculated for the beams.
The flexible end connections were given for a limited rotation at connections, and this rotation needs to be checked for each application normally. However, I often see that they are mostly ignored by engineers unfortunately.
While the connection rotates it pulls the top of the connection and push the bottom into the column and the wall change the shape, deflects within the yield criteria (yield line analysis), and restrained by the perpandicular walls. Therefore, you need to control stresses within the hook law or allow some yielding. All these are the design issues you need to go through.
Hope it helps.
Ibrahim Demir
It is typically assumed by AISC that a simple shear connection will develop no more than 20% of the fixed end moment. In reality, no connection is either perfectly "simple" or perfectly fixed. Your detail does not restrain the flanges in any significant way, and I would consider this to be a simple connection. However, I do agree that connection eccentricities need to be considered.
When worrying about distortion of the tube wall, as long as the tube wall is not a slender element and you are not detailing for moment connections you should be ok (see page 10-158 and 1-93 of 13th AISC Manual). In any case, I doubt that a 6x6x1/4 tube has enough stiffness to attract significant moment, and I assume that your beams have been designed as simple spans.
I am concerned about a couple of issues regarding your detail. With a 6" tube, it will be difficult to get weld on the double angles where they meet the tube because of the corner radius of the tube. I think this detail may be difficult to construct. In the 13th AISC Manual they tabulate values for workable flat dimensions. As long as the tube wall is not a slender element, you can use a shear tab connection, which would be easier to fabricate, see page 1-93 of the 13th AISC Manual. If you stay within the guidelines of AISC, you should be fine with single plate shear connections and not have to worry about yield line analysis.
Regarding the beam comming in to the corner of the tube, I think this would be a very difficult to slot the tube at the corner. I recommend a bent plate welded to the wall of the tube and then you can bolt the outstanding leg to the beam web since this is much easier to fabricate.
I don't know what your loads are or what the unbraced length of the column is, but a 1/4" wall thickness seems a little on the thin side. With complicated connections, I would recommend 5/16" minimum.
When worrying about distortion of the tube wall, as long as the tube wall is not a slender element and you are not detailing for moment connections you should be ok (see page 10-158 and 1-93 of 13th AISC Manual). In any case, I doubt that a 6x6x1/4 tube has enough stiffness to attract significant moment, and I assume that your beams have been designed as simple spans.
I am concerned about a couple of issues regarding your detail. With a 6" tube, it will be difficult to get weld on the double angles where they meet the tube because of the corner radius of the tube. I think this detail may be difficult to construct. In the 13th AISC Manual they tabulate values for workable flat dimensions. As long as the tube wall is not a slender element, you can use a shear tab connection, which would be easier to fabricate, see page 1-93 of the 13th AISC Manual. If you stay within the guidelines of AISC, you should be fine with single plate shear connections and not have to worry about yield line analysis.
Regarding the beam comming in to the corner of the tube, I think this would be a very difficult to slot the tube at the corner. I recommend a bent plate welded to the wall of the tube and then you can bolt the outstanding leg to the beam web since this is much easier to fabricate.
I don't know what your loads are or what the unbraced length of the column is, but a 1/4" wall thickness seems a little on the thin side. With complicated connections, I would recommend 5/16" minimum.
- Thread starter
- #19
First, I would like to thank you all for your insight, time and effort. Thank you. Thank you. Thank you.
Moving on then...My involvement was detailing the beam and column lengths and the connections. My direction was "design the connections for A325N bolts to develop half the maximum beam shear per end." I assumed from this that the connections were from simple beam analysis.
I have sent the architect (and I hope he sends to the engineer) questions regarding the overall stability of the framing and the lack of rotational stiffness in the connections, particularly at the column tops.
I was hired by the fabricator who chose the slotted detail. He's making the parts in the next few weeks so we'll see how wisely he chose.
The angle clips are 2 1/2" x 3 1/2" x 3/8" 6" long. I'm asking the fabricator to weld three sides of the angle clips - top, side and bottom. Neglecting the vertical side, there could be 5 inches of 1/4" weld for each angle but if I count on only 8" total I would expect those 8" to hold about 32kips. The maximum shear per beam end varies from 18.5 to 35.5 kips. We added beam seats for the 35.5 kip loads.
All in all I think we will be OK but I'm looking forward to the architect / engineer's review of my drawings.
Moving on then...My involvement was detailing the beam and column lengths and the connections. My direction was "design the connections for A325N bolts to develop half the maximum beam shear per end." I assumed from this that the connections were from simple beam analysis.
I have sent the architect (and I hope he sends to the engineer) questions regarding the overall stability of the framing and the lack of rotational stiffness in the connections, particularly at the column tops.
I was hired by the fabricator who chose the slotted detail. He's making the parts in the next few weeks so we'll see how wisely he chose.
The angle clips are 2 1/2" x 3 1/2" x 3/8" 6" long. I'm asking the fabricator to weld three sides of the angle clips - top, side and bottom. Neglecting the vertical side, there could be 5 inches of 1/4" weld for each angle but if I count on only 8" total I would expect those 8" to hold about 32kips. The maximum shear per beam end varies from 18.5 to 35.5 kips. We added beam seats for the 35.5 kip loads.
All in all I think we will be OK but I'm looking forward to the architect / engineer's review of my drawings.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
