Pipe shoe and saddle design basis or guide 1

[IFRs]

I'm looking for a resource that covers the design of pipe shoes and saddles. I've seen many variations in materials and especially welding details. I figure it's about time I learned how these items are designed. Any help would be appreciated!

 

[BigInch]

I've never seen anybody design a pipe shoe. They're cut to the right height and that's about it.

Large saddles (on equipment or large headers) are designed just like structural base plates for columns, etc. and sized for load distribution (ie. flange bending only) If very heavy loads, they may need to have an area giving a pressure load less than the bearing capacity of the concrete they are usually attached to. Sometimes there is a pressure limitation on a teflon friction reducing pad that goes inbetween the bearing plate and the saddle plate. Size the top one so that the teflon always remains covered before and after movement. You might have to size the saddles for longitudinal bending from the friction loads, uplift or overturning moments from wind, thermal loads, so there could be anchor bolt tension loads (flange bending) to contend with. Heavy loads might also tend to cause plate buckling, in which case (can you guess?) put stiffeners in there. Put stiffeners everywhere the bending stress is too high, then check the bending between stiffeners. If too high, put another one in there. Keep your eye on shear, especially near anchor bolt holes. "Saddles" that are really anchors can be pretty hefty, but still the same princips. Open a book on structural steel design - column base plates, Blodgett's "Welded steel structures" is one I think of if that's still around, to get the basics, then find a similar chapter in a vessel design handbook for specific applications.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[JohnBreen]

This is a frequently asked question – you should learn to use the “search” function.

Generally speaking, pipe shoes and saddles are basically commodity items. You buy them off the peg simply ordering them to suit the pipe diameter (and insulation, et al). You only need to download catalogs to get all the information you need. For example, go here for information:

http://www.anvilintl.com/db/catalogs/Pipe_Hanger_Design-07.pdf

The prevailing industry standards for the hardware are:

MSS-SP-58
MSS-SP-69
MSS-SP-77
MSS-SP-89
MSS-SP-90

For power piping design rules look at ASME B31.1 paragraphs 120 and 121, et al.
For process piping design rules look at ASME B31.3 paragraphs 321, et al.

OK, that is “generally speaking”. If you want to look at design of base elbows, trunnions, dummy legs and other welded appurtenances, look for the old Kellogg book of Standards (not to be confused with the Kellogg book “design of Piping Systems”) in the section that covers these components.

Also, when the outside diameter to ratio wall thickness exceeds about 120 you need to look at other design guides. This was discussed in another thread:

thread292-209303

This (large D/t ratio) becomes an issue as methodologies other that beam theory design are needed to determine the maximum stresses at the support point of attachment to the pipe.

Regards, John

 

[europipe]

Here's another alternative

 

[BigInch]

Great. But it appears to me (as usual) that the pipe is holding down the balloons.

I've seen a similar thing out in Pinter Reservoir in Wyoming, except the pipe was "holding down" a gas scrubbing vessel. They had to unbolt the scrubber from its foundation when the compressor station started subsiding 1.75 in/month vertically and moving south at 1.25 in/month, then they called ME! I didn't think about the balloons then. Good idea. Could have used quite a lot of them.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[europipe]

Many small companies try to solve things without knowledge.
And then at last they call pipers to clean up the mess, but it always costs too much then.
B", is a job as tie-in engineer in India nothing for you?(see photo)

 

[BigInch]

Have torch... will travel. I'm pretty sure that's a BP refinery.

Nice picture! I already downloaded and framed it. Title: "My Dream Job"

Actually I put it on the wall to remind me NOT to apply for any jobs in India, no matter how much fun they look like they're gonna' be, but tell me the rate, P.D. & completion bonus anyway and I'll let ya know. All I need to have fun with that is my trusty cutting torch... 100ºF... 100% Humidity ... and mosquitos living in the pipes!

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[DSB123]

JFRs,
Your question is a good one. You will find that generally the following applies (from experience over several years)

1. The need for a pipe shoe is identified and in some cases the loads applied to the pipe shoes is calculated.
2. The piping Designer or pipe support Designer selects a "standard" pipe shoe from the Companies pipe support standard and that's it unless the load capacity of the shoe is tabulated which can be checked against the applied loads.
3. Where a "Support Shoe supplier" is contacted and the loads are given then support shoes are designed for the loads and offered.

The problem is the "Pipe Support Supplier" only wants to provide shoes and the whole support/pipe interaction is not checked. The pipe support supplier usually never checks the stresses imposed in the pipe wall locally around the pipe shoe. The pipe support designer does not check this either and it is missed. Everyone assumes that someone else has done the pipe wall check. I know this as I have been working on a Project where this occurred. The pipe shoes were selected (but the load capacity of the "design" was not known) and applied to thin walled piping where high loads were to be transferred between pipe and steel but no-one had considered the "local" stresses in the thin walled piping.

From my experience there are not many pipe support designers who actually perform calculations on the "design" of pipe shoes - drafting what looks to be OK is the norm.

I may be a bit cynical but its my experiences only!!

/

 

[europipe]

DSB123, do you have any reference (sheet) that applies to the relation of loads, thin walled pipe and supporting?
Like IFRs I also have many problems with supports.
Greetings.

 

[IFRs]

I am also interested in the rational for cutting holes in the pipe shoes - for shoes that have a vertical plate under the pipe, many standard shoes have windows cut in them. I wonder why? Heat transfer? Welding? Asthetics? etc.

 

[DSB123]

europipe,
You need to do the relevant checks unless you have a set of "designs" for particular shoes which identify maximum loads. On the last job I was on the Sub-Contractor chose pipe shoes from his standard (which basically was for Carbon Steel) and then specified the shoes in SS for use on the Project. When challanged them about the validity of the selection and the load capability but they were not able to prove that they were adequate. They had not checked the pipe stresses induced into the pipe wall and indeed did not know how to do the calcs. We eventually had to use FEA to get the shoes to pass the Code limits.

My experience is that the pipe wall stress checks are never done which is a problem if the pipe wall is thin (Sch10s).

 

[vesselfab]

the little slots you see in standard shoes is for

insulation banding passage.

 

[JohnBreen]

IFRs

I generally agree with vesselfab regarding the "slots" in the vertical legs of the "T" supports.

However, you also hit upon another reason for providing "breaks" in the attachment (to the pipe) weld - relief from thermal stresses. Some times the "T" support or (in the case of very hot vertical risers) "batwings" that are welded to the pipe act a "heat sinks" and cause a thermal expansion differential at the weld line. Get a copy of Pipe Fabrication Institute engineering standard ES-26 to look at some recommendations for welded attachments to hot pipe.

Regards, John

 

[europipe]

DSB123, I know of the importance of stress issues,
my concern is about the most economical construction of supports.
I'm working at a site with no supporting standards whatsoever.
The standard for bridges is:
A small bridge: 1 mtr. wide
A middle bridge: 1.5 mtr. wide
and a very large bridge: 2 mtr. wide
and made of pipe, better on one column then 2.
Pipes are mounted on the side, with hangers, shoes on different elevations.
Can You imagine?

Greetings.

Greetings

 

[JohnBreen]

DBS123 and Europipe

There is a fairly rigorous method in ASME III (susections NB, NC and ND) for evaluating the stresses in the pipe wall due to loadings transferred by welded appurtenances. I will try to find the reference as time allows.

I know there is an ASME Section III (Nuclear) Code Case (Case N-318-5, "Procedure for Evaluation of the Design of Rectangular Cross Section Attachments on Class 2 or Class 3 Piping") but I think even that one is limited to D / t less than or equal to 100.

For 40 years or so olde piping engineers used the M. K. Kellogg method for "Design of Pipe Attachments" (31 pages extracted from the M.W. Kellogg Company book of Standards - covering trunnions, base elbows, dummy legs and similar appurtenances) but lately that method has fallen into disrespect after several FEA studies have show it to (in many cases) be non-conservative. At this point I feel the only accurate method is FEA and personally I use FE/Pipe. By the way, Paulin Research Group is planning to so a series of on-line seminars very soon addressing this and many other piping design issues. Go to their web site to learn more about the seminars.

Regards, John.

 

[JohnBreen]

Further to the seminars mentioned above:

Paulin Research Group Webinars Starting Next Week

Paulin Research Group will offer several webinars that discuss various subjects of interest to piping and pressure vessel designers and engineers. Presentation times are typically scheduled for thirty minutes or less and attendees are encouraged to submit questions during and after the webinars.

See

http://www.prg-software.com/Webinar.aspx

for more information.

*Common Piping Errors and Ways to Avoid Them*
Presenter: Tony Paulin, P. E.
Tuesday, February 26, 2008, 9:00 AM CST

*PVP Failure vs. Calculation and a Common Problem*
Presenter: Tony Paulin, P. E.
Tuesday, February 26, 2008, 10:00 PM CST

*New Aspects 2007 ASME Section VIII Div 2*
Presenter: Tony Paulin, P. E.
Thursday, February 28, 2008, 9:00 AM CST

*Shell and Brick FEA and Fatigue Rules*
Presenter: Tony Paulin, P. E.
Thursday, February 28, 2008, 10:00 PM CST

*Fatigue Design with ASME and other Worldwide Codes*
Presenter: Chris Hinnant
Tuesday, March 4, 2008, 9:00 AM CST

*Low Cycle Fatigue Tests on Piping Components at PRG*
Presenter: Tony Paulin, P. E.
Tuesday, March 4, 2008, 10:00 PM CST

*Analytical Gasket Properties and Leakage*
Presenter: Tony Paulin, P. E.
Thursday, March 6, 2008, 9:00 AM CST

*High Temperature Guidance for Nozzle Loads*
Presenter: Tony Paulin, P. E.
Thursday, March 6, 2008, 10:00 PM CST

 

[DSB123]

JohnBreen,
Thanks for the info. I know of the ASME III approach and have actually just written a routine in VB to perform the assessment as it is in the BS EN 13480 Code.
Also I am aware of the "olde" Kellog methods and also have these coded up in VB. This is also adopted in a BS document (can't remember the number at the moment). However both the Kellog and ASME III approaches tend to be for trunnions or lugs rather than pipe shoes. My point is that in general designers apply pipe shoes to pipework without the necessary checks of the stresses induced in the pipewall. For trunnions then, yes, either ASME III or the "olde" Kellog approaches are used but for some reason stresses are not checked for pipe shoes. Generally there is no problem but on large pipe sizes or high loads stress levels can exceed the Code limits.

Many thanks for the FE/pipe info on the Seminars. I have been wanting to know more on the FE/pipe product and have tried to get past Employers to purchase the software but as a Contractor it's difficult to have enough "clout" to specify purchases. I have downloaded what I can and beleive it could be beneficial to the type of work I do but to buy it myself is not an option.

Regards

Denis

 

[DSB123]

Thought I would resurrect this thread with a few examples. Would like to hear what you think about these situations.
Consider a 20"nb pipe for which the pipe shoe is basically an inverted Tee which is welded to a 5.4mm thick reinforcing pad which extends over 120 Degrees of the pipe. The pipe shoe has two gussets (each end of the shoe which are also welded to the re-pad such that the total width is 203mm at the repad. Any concerns??

What about another example where for a 36"nb pipe it is intended to support the pipe on a pipe shoe fixed to the pipe with two bolted pipe clamps-one either end of the pipe shoe. The shoe is welded to the pipe clamps and has gussets which are 305mm total width. Any comments??

 

[BigInch]

The pipeline codes do not permit the welding of supports directly to pipe, hence the second option is preferred for pipelines.

The dimensions of the shoe and gusset plates don't mean much without the associated pipe spans or loads on the pipe/support. However the less than 1/4" thick reinforcement pad could not be counted on much more than distributing any load by more than 45º or so to each side of the intersecting gusset plate and the pad wouldn't take much in bending load to do it in. Also important could be the contact length of the T (distance between gusset plates).

Say the pipe wall will experience a uniform load along the stem of the T that will tend to bend the wall inwards (make an upside down Valentine) for the length = the distance between gussets. The gussets will act to restrict that bending to only the length inbetween them (length of the T). I'd at least take the pipe support load applied to a length of pipe equal to the length of the T, look at the pipe x-section and check a differential bit of pipe wall near the vicinity of the T's stem for combined stresses of radial pressure stress, hoop tension in the wall plus the bending stress from the point load of the stem of the T acting on the ring, plus any axial load and the bending stress in the pipe's axial direction (from the pipe spans on both sides of the support). Also add/subtract any pipe axial load due to temperature and/or pressure.

http://virtualpipeline.spaces.msn.com

"What gets us into trouble is not what we don't know, its what we know for sure" - Mark Twain

 

[DSB123]

BigInch,
Thanks for the reply. I was just illustrating the type of pipe shoe that some "Pipe Support Designers" come up with - totally inadequate. These are proposed for a Project within the company I work at. Blind leading the blind comes to mind. Luckily I am not on the Project but it just goes to show the type of pipe shoes used.