Piperack Design 1

[Rabbit12]

For you guys that design process piperacks and pipe supports.

Do you combine a thermal friction force with wind load? I'm not talking about a thermal anchor just the friction force generated from expansion or contraction of piping prior to pipe slip on a support.

 

[BigInch]

Yes but.... when I did pipe racks we always took only 10% of the weight of all pipe and contents as the lateral thermal load on beams supporting pipe. If the support beam was concrete the thermal load was dead load factored, wind was live. Generally speaking, what we didn't do was take wind blowing in a parallel direction with pipe axes inside the rack, so the wind and thermal load seldom added together in a long pipe rack design. Pipe thermal load is parallel with pipe axis. We used 10% of total pipe load as a lateral thermal laod because some thermal friction went in the left direction, some friction went in the right direction, thereby canceling most all of it to zero. We just considered that 10% remained uncanceled. Also in general, we always took wind only as a load applied perpendicular to the pipe's run through the rack.

Find what you like to do, earn a living at it, and then make your lifestyle fit your income. — Chuck Yeager

 

[TLHS]

As per PIP STC01015, you don't need to account for friction forces simultaneous to wind or seismic. The exact wording is "Friction loads shall be considered temporary and shall not be combined with wind or earthquake loads. However, anchor and guide loads (excluding their friction component) shall be combined with wind or earthquake loads)"

 

[Rabbit12]

Thanks guys. I was mainly thinking of a T-support.

If you look at the new ASCE 7-10 it is more ambiguous than 7-05 when it comes to self-straining loads (See ASCE 7-10 Section 2.3.5). I recently came across a client spec that says to combine friction forces with full wind or seismic. I'm not sure I completely agree with that but I also don't know that I agree that a thermal friction force is temporary. The only way for the force to go away is for the pipe to return to it's original length or for the pipe to slip on the support.

It becomes critical on a t-support because if you consider them acting concurrently, you have biaxial bending of the t-support with a wind force perpendicular to the pipe. That can have an affect on anchor design.

 

[wadavis]

We generally consider thermal stresses to be self-relieving: as the wind load increases displacement, the thermal friction load on the rack will decrease/reverse. So we do not considered them combined, with exceptions for stability cases.

Structural, Alberta

 

[SlideRuleEra]

Agree with wadavis. At, or near design speed the structure will be buffeted more than enough by wind for pipes to slip on the supports.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[Illbay]

1. It's "pipe rack." Two words There is no officially accepted portmanteau "paperack." (Used to have a coworker whose first language was NOT english; he pronounced it "PIPER RACK" as a result.

2. I have no idea why this notion still persists that industrial structures in general and pipe racks in particular somehow are a completely separate breed of structure that will not submit to the general statistical analysis that has gone into the development of the ASCE 7 load combinations. I think it has to do with "back in the day" when there were so few applicable standards so structural engineers working for EPC companies - not having a whole lot to do, since actual structural design was forbidden them - created their own standards, and of course these got passed down to the younger guys like the stone tablets of Moses.

But these are MODERN times, guys and gals. A LOT of time and study has gone into the load combos, and I do believe you can use them with confidence!

This is what ASCE 7-10 2.3.5 says:

2.3.5 Load Combinations Including Self-Straining Loads

Where applicable, the structural effects of load T
shall be considered in combination with other loads.
The load factor on load T shall be established considering
the uncertainty associated with the likely
magnitude of the load, the probability that the
maximum effect of T will occur simultaneously with
other applied loadings, and the potential adverse
consequences if the effect of T is greater than
assumed. The load factor on T shall not have a value
less than 1.0.

So what this says, to me, is "be an engineer." That means that if you believe the "self-straining load" will "occur simultaneously with other loads," then include it with an appropriate load factor (I use the load factor associated with Live Load because in my mind, the level of uncertainty is about the same).

Now, think about the typical plant operation, and that the wind event represented by W in your load combination is an EXTREME event. For instance, if you're on an offshore platform, this is a m mother of a hurricane on the way, and the shut down and skedaddle. Same for onshore; no one is going to be operating that plant during a high-wind event.

So the answer is, if you have these events occurring, do not include the Self-Straining Load component in the combination.

Very simple.

"No one is completely useless. He can always serve as a bad example." --My Dad ca. 1975

 

[Rabbit12]

Illbay,

1) Thanks for the grammar lesson. Maybe check your spelling as I'm not sure what a paperack is.
2) Where you come from do process plants shut down prior to an earthquake? How about a thunderstorm with 100+ MPH winds?


Wadavis/SlideRuleEra,

That makes sense and I could see the pipe sliding and thus relieving the friction force with wind (or seismic) induced deflections. I was looking for discussion, not some absolute "this is what you should use" type of thing. I kind of thought the requirement to include them both simultaneously was overkill but was curious to what others would use. I think some percentage, maybe 50% of the wind or seismic load could be justified and reasonable.

 

[DayRooster]

I agree with TLHS about the use of PIP standards in addition to ASCE. I also laughed when I saw that ASCE changed their stance on thermal loads from a firm statement in 7-05 to pretty much "good luck" in 7-10. In football terms, it is like they are punting on first down after getting sacked for a safety last possession...Generally, for your situation (a basic T) I would design per the load cases outlined in PIP without adding the pipe thermal friction to wind. This is because generally any good pipe stress design that you would be working with would be placing telfon slide plates, expansion joints, rollers, hangers, or something to allow this expansion without major friction. They generally don't want to mess up there pipe. If you have the case where a pipe is U-bolted to a support so its riding steel on steel, no pipe expansion joints and you have large thermal forces then you may want to consider a load case with thermal and wind together. Maybe with modified load combinations based on your judgment (I generally don't if I think wind and thermal could occur together). Keep in mind the real "fun" with thermal loads on pipe racks is when you have to worry about the steel itself expanding/contracting due to climate in long outdoor racks...add structural expansion joints and keep an eye on the connections.

Edit: if you do have large enough thermal forces at the top of the T support (assuming the top beam is a W-shape) then don't forget a torsional check...

 

[BigInch]

So then thermal loads of bridges are ignored when the wind blows or the earth quakes?

DaRooster, This is because generally any good pipe stress design that you would be working with would be placing telfon slide plates, expansion joints, rollers, hangers, or something to allow this expansion without major friction.
Uh... Absolutely NOT! Teflon, rollers, or somethings are not normally used. The best it normally gets is steel pipe shoe sliding on a steel or concrete beam. Special circumstances ONLY. Structural expansion joints are not used in pipe rack design. Independent rack sections separated every 100 ft distribute the growth and contraction to amounts manageable without expansion joints in the structural steel frame. Much more important to consider that when designing long continuous conveyor belt bridges and towers.

TLHS, PIP STC01015 (like all codes and design specs) should be used only as a guide when specific and detailed knowledge and information about concurent loading conditions (or whatever else concerns you) is nonexistant.

SlideRuleEra,
wadavis,
Illbay, (I tend to agree with you, except for shut down load case)
Pipe supports or structures buffeted by wind, or otherwise shifting under the pipe is immaterial. The (maximum) value of any lateral load that can ever be transferred between a pipe simply resting (not welded anchored or guided) on its support is the weight of pipe and contents x friction factor. Thermal loads of the pipe to a support can increase only until [&fi;]W is reached, then slipping occurs so the load never reaches higher values. Likewise wind acting on pipe, transferring the load to supports below could also not reach values above [&fi;]W, otherwise the pipe would first move and perhaps even slip off the support. Likewise, wind should also be applied with thermal friction load during shutdown. Once the pipe has shifted on it's support "from wind buffeting" or whatever, to the thermal position, the thermal friction load might be less, however shutdown at that point would simply reverse the thermal friction load as the pipe cooled off to ambient temperature.

Rabbit12 (IMO you are using solid and good logic),
It is obvious to me that such concurrent wind (or seismic) and thermal loading (mentioned in the previous paragraph) can and will occur simultaneously. I think some percentage,50% Thermal loads are "dead" loads, as desing temps are used for design values and thermal effects are relatively well known, easily calculated and their application straight forward. Wind loads are live. Chose appropriate factors when applicable to your material design. In your T support case the 10% W[&fi;] would not apply, because the one or limited number of pipes would not counteract each other. Use 100% W[&fi;] for your small limited number of pipes.

Find what you like to do, earn a living at it, and then make your lifestyle fit your income. — Chuck Yeager

 

[DayRooster]

Big Inch - See Lisega's guide for pipe supports to see all of these examples. I guess general is a relative term only related to my experience. Back to what I said originally if the pipe is an issue then you need to handle it accordingly with the guidance of PIP and ASCE. Also, the structural expansion joints I was referring to are typically in my designs double columns which would be independent rack systems...just curious what type of process situations are you getting into where they just set a pipe shoe on some concrete and calling it good? Do they do anything else or are we talking simple a pipe shoe on concrete that's it?

 

[BigInch]

Just a pipe simple ole shoe sliding on a concrete, or a steel beam. Teflon is used more commonly on horizontal vessels, ie. relatively hot heat exchangers, etc. Teflon is very seldom seen on pipe, unless a really hot steam line. Steel on steel is pretty much the only thing used in oil refineries & chem plants, gas comressor stations, oil pipeline pumping stations, least until you start getting into 350F steam lines, or go to nuclear plants. There is the odd anchor, stop or guides, but not so many really. Hangers are used offshore where there is usually more structure to work with and you don't have a lot of long runs in pipe racks. Nucs are a whole-nother story, but mostly 100s of pipe whip restraints composed of 2"-3" thick plate-built up beams (South Texas Project, Matagorda, TX 1978).

Find what you like to do, earn a living at it, and then make your lifestyle fit your income. — Chuck Yeager

 

[BigInch]

We would never pay for two adjacent columns. Just discontinue the horizontal thermal bracing in one bay every 100 ft.

Find what you like to do, earn a living at it, and then make your lifestyle fit your income. — Chuck Yeager

 

[DayRooster]

BigInch - Yeah I am familiar with your mindset based on the industry you serve. The Oil & Gas industry produces some interesting threads such as this gem:

http://www.eng-tips.com/viewthread.cfm?qid=189187

Just so we are clear steel on steel is not an issue. I just recommend that one looks close at those types to understand what the pipe is doing globally. I don't even know what to say yet about the steel sitting on concrete yet...you are correct about the use of Teflon on high energy "hot" systems or sometimes very large "cold" systems. Most pipes that don't have much thought put into them are usually (from what I have experienced) small cold systems and these have almost neglible friction loads...but all that said I think your input is important especially if Rabbit is doing similar work to you. Keep in mind that process designs can cover a wide range of markets.

Edit: That picture is pretty much the same concept as a double column (column bay then another column bay with no connection beam). It's just a break in the pipe rack steel. In my designs those are usually a bit closer together (about 10 feet apart) to account for small bore pipe and not just only large bore.

http://www.structuremag.org/wp-content/uploads/2014/08/C-StructDesign-Bendapudi-Feb101.pdf

Also, here is another link where the term "double column" comes from. Keep in mind that it doesn't define the distance between double columns. It is related to buildings but some of same theories can be applied to pipe rack design too.

https://www.aisc.org/globalassets/modern-steel/steelwise/052011_may11_steelwise_web.pdf

 

[BigInch]

What industry he's in I'm not sure. Sounds like pretty close. Process, piperacks and pipe supports, or pipe rack if you prefer (either means the same to me).

Find what you like to do, earn a living at it, and then make your lifestyle fit your income. — Chuck Yeager

 

[BigInch]

Just noted your link.
Petrochem very seldom requires any PE stamps. Nonpublic structures on private property, highly restricted access. Once during the occasional blue moon you might see a stamp on structural dwgs, most likely because some guy just got his PE and wants to see how many drawings he can stamp and sign at one sitting. ASME stamps are on the pressure vessels, etc. The only saving grace for structural engineers is that petrochem plants usually go down for some reason other than structural faults. Can you imagine, saying "Ya I stamped the __________ (Bhopal, PiperAlpha, BP Texas City Refinery, fill in the blank) drawings". Once I stamped some site permit conceptual plot plan drawings for a town council meeting, but never anything else. Why take that liability away from the shareholders, if you don't have to.


Find what you like to do, earn a living at it, and then make your lifestyle fit your income. — Chuck Yeager

 

[Rabbit12]

Thanks for all the responses guys. I'm also in the oil and gas industry but unlike BigInch I end up stamping quite a few drawings. Our clients mostly require it.

 

[TLHS]

I've been in upstream, midstream and downstream stuff, and almost everything gets stamped in Canada.

 

[IFRs]

Texas requires almost everything to be stamped. I've done a few insulated steam and hot oil lines, used mostly steel on steel shoes until the pipes got large ( 24" and 36" ) and were being retrofit on existing racks. We would cross brace a set of columns, no cross-brace, cross brace a set of columns, etc. There were always anchors and pipe loops on the theory that the pipe will do what it wants unless you tell it what you want it to do. Thermal loads were included at 100% because they were known to occur. Wind loads were factored but never in the direction of the pipe. We used a lot of guides on anchored pipe runs to keep them from bowing sideways and interfering with other pipes or stressing valves, etc. Just the sun on empty white pipe would cause them to bow sideways far enough to be off the shoes. T-posts were trouble if there was thermal displacement and the post was far from an anchor. I've certainly seen many pipes slipped off their shoe for lack of restraint or consideration for thermal expansion, then when the pipe cooled the shoe dragged the top of the support way off vertical. Do what you think is proper, usually the steel and foundation are not huge cost factors compared to the field labor or the cost of fixing some lame design...

 

[BigInch]

I haven't practiced in TX since '91. Back then NOTHING was stamped. I'm curious. When, and how, did that change. Has it been via law (which ones), or by client demand?

Technology is stealing American jobs. Stop H1-Bs for robots.