Pipe Rack Moment Connections

[thefatmanrosco]

Our company has been designing thousands of lineal feet of pipe in the O&G industry. The typical construction consists of lateral moment frames connected by struts in longitudinal directional that are x-braced. The columns are typically W8's, but preferably W10 or W12's for constructabilty. The moment connection of choice has been end-plates. However, due to the quantity and size of pipe and the cross-members, the check of the column requires web stiffeners. Are FMC (flexible moment connections or the "old" type 2 wind type connections an option? I was hoping to eliminate the need for the costly web stiffeners. The pipe racks are located in a low seismic hazard.

 

[structSU10]

Since the stiffeners are typically costly, designing a 'clean' column that requires no stiffeners is desired. I know I have seen an excel program that optimized this ( think it is called clean column). I would think a stockier web is all you would need to help the situation, rather than using a more flexible connection. AISC says in DG 13 goes into some detail about how to economically select columns.

 

[jimstructures]

The excel spreadsheet "Clean Column" is available from the AISC website Steel Tools.

Jim H

 

[amec2004]

I would suggest you stay with the normal MC for the following reasons

1. For industrial piperack it’s normally serviceability governed, not steel member strength governed. You need the MC to obtain stiffness to control the lateral sway in the transverse direction so that the lateral sway is under pipe Stress requirement

2. You can increase the transverse beam depth to eliminate the stiffener plate or doubler plate. The material cost like higher depth beam is minimal while the labor cost of cutting and welding stiffener plate is higher

In a big project we normally provide a project spec, with tabulated MC capacity without stiffener plate under different size of beam column. The engineer will lookup the table for MC capacity without stiffener plate under a beam column size. If the capacity cannot be achieved, just change to a higher depth beam.

AISC clean column spreadsheet is very good. I have made one too with more choices.

If you turn your piperack to partial restraint it will be difficult to make accurate computer modeling as you may need a rotation spring joint to make it between pin and rigid connection, plus your transverse direction lateral sway will just slide …

anchor bolt design per ACI 318-11 crane beam design

http://www.civilbay.com

 

[JoshPlumSE]

I don't see how going with a different type of moment connection would change your stiffener requirements. You'll be transmitting the same basic moment, won't you?

I would just use a column with a thicker web or flange. That's a pretty easy solution.

In high seismic connections, I have used a "haunched" moment connection in the past. That's where I take a Tee section and weld it onto the beam at the end plate connection. Thereby increasing the moment arm and reducing the flange force significantly.... but, this isn't a pre-qualified connection, so you would still be stuck with R = 3.0.

 

[amec2004]

>>You'll be transmitting the same basic moment, won't you?

No. If the beam-column desgined as rigid, it gets moment, if it's designed as pin, it doesn't get any meoment and all the moment goes to mid-span plus lateral sway increasing a lot, if it's designed as partial restrain, it gets the moment between 0 and rigid case

anchor bolt design per ACI 318-11 crane beam design

http://www.civilbay.com

 

[amec2004]

>>I would just use a column with a thicker web or flange. That's a pretty easy solution.

Thicker column web is expansive and not effective

Thicker column flange won't help.

Increasing beam depth is more effective.

anchor bolt design per ACI 318-11 crane beam design

http://www.civilbay.com

 

[TLHS]

Both thicker column webs and thicker column flanges can help significantly.

Beam depth is not necessarily the right solution depending on the load condition. If you have a few big pipes rather than a number of small pipes the weak axis forces from pipe friction can get quite significant, which points you towards using a short fat section. Deepening the beam becomes inefficient quickly because a similar weight deep beam has less width. To maintain weak axis capacity you have to add a large amount of beam weight. If you're in a situation with torsional forces it gets even worse, but you can often work around that.

To throw in a different solution, I'm a fan of knee braces when you can fit them in! There's generally enough space in an industrial situation.

 

[structSU10]

The limit states that control the need for stiffeners are directly related to the web slenderness - the thickness to depth ratio. You can have a thin web that is short that performs the same or better than a thick web that is long. Look at equation J10-4 and J10-5 in AISC 360-05 - tw (web thickness) is squared, and in the numerator in another part of the equation, thus a larger value increases capacity. depth of section is in the denomenator, thus it tends to decrease the capacity.

 

[amec2004]

>>Both thicker column webs and thicker column flanges can help significantly.

I would say increasing beam depth is more effective.

Please see the comparison at

http://www.civilbay.com/download/Increase-Beam-Depth-More-Effective.pdf

anchor bolt design per ACI 318-11 crane beam design

http://www.civilbay.com

 

[TLHS]

As I said in the rest of my post, vertical loads are not the only consideration in a pipe rack beam.

In a rack servicing a large number of small diameter process pipes and cable trays you're going to see something trending closer to pure vertical loads as you normally can assume that not all the pipes will be trying to expand in the same direction at the exact same time. At the other end of the spectrum, you could be carrying a single large pipe where you end up assuming horizontal loads of 30 or 40 percent of the weight due to friction against pipe expansion. There are significant downsides to having a deep, thin beam when you're looking at a case with top flange weak axis bending.

I wasn't disagreeing that use of a deeper beam can be a good solution. I was pointing out that it isn't necessarily the right solution for pipe racks seeing large horizontal loads because additional beam depth is generally a negative in those cases and it causes other detailing and sizing problems.

 

[amec2004]

Below is my comments

>>As I said in the rest of my post, vertical loads are not the only consideration in a pipe rack beam.

I didn’t say the vertical load is the main concern. The piperack transverse moment frame’s moment can be caused by vertical gravity load, it can also be caused by the lateral wind or seismic load. It’s the combined of both gravity load and lateral load to get the final beam end moment.

>>In a rack servicing a large number of small diameter process pipes and cable trays you're going to see something >>trending closer to pure vertical loads

“pure vertical load” is not the case, lateral load cause beam end moment as well.

>>end up assuming horizontal loads of 30 or 40 percent of the weight due to friction against pipe expansion.
>>There are significant downsides to having a deep, thin beam when you're looking at a case with top flange weak >>axis bending.

That’s true but what we are talking about is the moment connection of TRANSVERSE moment frame, that friction load goes to LONGITUDINAL direction and it will go to strut beams and vertical braces.
Please note the deeper transverse beam is never intended to resist lateral load like pipe friction or anchor load. When there is big pipes and causing big pipe friction or anchor load, we use horizontal braces to handle that.


>> I wasn't disagreeing that use of a deeper beam can be a good solution. I was pointing out that it isn't necessarily >>the right solution for pipe racks seeing large horizontal loads because additional beam depth is generally a
>> negative in those cases and it causes other detailing and sizing problems.

Use of deeper beam is aimed to avoid the use of stiffener plate or doubler plate in the transverse moment frame’s moment connection. The use of deeper beam will not cause any negative to effect to the lateral load in the longitudinal direction. For large lateral load caused by pipe friction or anchor load, we never expect the open wide flange section can handle the torsion plus weak axis bending, in this case we use horizontal brace to brace out the lateral load to strut beam, to vertical brace, to foundation.

anchor bolt design per ACI 318-11 crane beam design

http://www.civilbay.com

 

[humanengr]

You mentioned check of the column required "web stiffeners". Are these stiffeners actually continuity
plates required due to the column flange limit states? If so, you can try using 1/2 depth continuity plates - reduce
your welding by half.
If you are not referring to continuity plates and actually web stiffeners - do you really need web doubler plate, or
are you providing a diagonal stiffener in the web panel zone?