Hi
This might help:-

http://www.thermacor.com/wp-content/themes/thrma/f...

It doesn't give any indication of how many or where to position them related to a given pipe run, its seems to me you need to do an analysis of the pipe runs and decide whether you need an expansion loop or not.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

We ran some pipe a while back and didn't provide any expansion provisions, here is what I learned (after hacking it all apart).

Pipe will expand 15mm for every 100m of straight run, every 100 degrees C delta.

Loops need to be wide enough to be flexible, something like 1/2 meter in perpendicular plane for every 1" in NPS.

So in our case, we had equipment with rigid nozzles at the end of the pipe runs, and therefore we had to anchor a run at several places and make it take the movement at the loop.

We designed the loop to move based on length and operating pressure.

The 15mm/100m for every 100 deg C is just an approximation of the thermal expansion coefficient of steel.

Contact Perma-Pipe Inc. 7720 North Lehigh Avenue Niles, Illinois 60714-3491 Phone 708-966-2235 Fax 708-470-1204 and request that they send you their slide chart of pipe expansion and pipe flexibility. One side of this slide chart has steel or copper thermal expansion rate in inches per 100 feet from 50F to 640F for steel, 50F to 406F for copper based on installation done at 50F. The other side has configurations for 90 degree elbow or 90 degree Z bends or loops with recommended locations of moment guides between anchor points to provide flexibility in piping such that stress level in piping at maximum service temperature of 650F will be below 22,500 psi. Basically you have to determine where you want piping anchored. This typically would be at the connection to equipment anchored and locations where you do not want piping to move because it will hit something. Then you use the chart to determine expansion rate. Calculate the net expansion of the longest leg for 90 degree elbows, the expansion between anchors for Z bends or loops. Then you locate moment guides per the slide chart. Check the internet for pipe anchors and pipe guides. Coordinate with structural location of anchors, pipe expansion, filled and insulated pipe weights and have them calculate the additional load to be imposed on the structure so they can check or design the building structure accordingly.

Did you try searching for your answer in the code? ASME 31.3 Part 5, 319 Piping Flexibility.

If have questions feel free, but you have to do a bit of reading and there is no easy answer here.

There are manual calculations for stresses in expansion loops. eg. Kellogg, Design of Piping systems. But one you starting using a programme like Autopipe you will never go back to manual calculations.

RMDaniel
You posted the following questions & statements:
1. I am looking for a reference on when an expansion loop is required.

2. I mean, I know that it is necessary for high temperature service piping and for straight piping to avoid too much stress.

3. But I cannot find any specific values regarding the minimum temperature and minimum distance such that an expansion loop will be needed in the line.

4. If any of you have guidelines for this, it would really help me in my current project.


Comment:
#1 This is actually two requests in one; location of any/all References related to Expansion Loops, and when are Expansion Compensators required?
So first, you use any search engine on the web and type in “Expansion Loops” and you will get 17million sites. Some might help you but most will not. You first need to know this basics relating to the subject.

#2 Expansion of piping materials (Metallic and non-metallic) will happen to all systems and as the result of all temperatures. You first need to consider the location of the Job Site and the atmospheric temperature at the time of the initial installation as well as the potential temperature for later shut-down events.

Then you need to consider the projected actual maximum (or minimum) Operating conditions. Don’t make the mistake of using the system “Design” Temperature. The Design Temperature is only a theoretical number for use in calculating pipe wall thickness. The pipe will never actually ‘See’ the Design Temperature.

#3 When considering the actual Operating conditions you may have lines with an elevated temperature above the installed temperature or you may have some lines with a negative temperature far below the installed temperature. Examples include, Carbon Steel lines installed at ‘0’ degrees (C or F) and then operating at 600 degrees F (315.5 C) will expand a great deal. The same line in Stainless Steel at the same temperatures will expand almost twice as much. Pipe materials will also react to artic site temperatures and severe negative operating temperatures as well. Sub-Zero temperatures inside or outside the pipe material will cause “Contraction” (the opposite of Expansion) to the same degree as a high temperature.

Next you need to consider the pipe material itself. Different materials have different coefficients of expansion. As noted in the paragraph above Stainless Steel has a higher coefficient of expansion than Carbon Steel. CPVC and Fiber Glass pipe also has a higher coefficient of expansion than Carbon Steel.
See:
https://www.google.com/#q=expansion+rates+of+diffe...
Or
http://www.engineeringtoolbox.com/linear-expansion...
Use the correct operation temperature minus the correct installation temperature for the proper expansion/contraction rate times the segment length.

#4 Next you need to consider the options available to compensate for the expansion/Contraction. Loops are a form of configuration and work fine but there are also other ways to solve the same problem. These other ways include other configurations and Engineered Expansion Joints (EEJ).

Other Configurations include:
“I” – With this configuration you have only a straight run but only the one (inlet) end is anchored. The far end is unrestrained and allowed to grow (expand) with the operating temperature. A typical example of this would be a Utility Steam supply line on a loading dock or wharf. The piping at the supply or land end would be anchored but the far end out at the end of the wharf is not anchored because the connections at the far end are just hoses used for cleaning decks and machinery. Between the anchor end and the far end the pipe is guided to maintain position.

“L” – With this configuration you have a natural “L” shape because of the pipe routing. The two legs of the “L” do not need to be equal in length but should be at least 1/3- 2/3. With this configuration it is important to recognize that the ends of the pipe of both legs need to be anchored and any guides on the pipe between the anchor and the connection point of the two legs need to allow for the expansion of the other leg.
“Z” - With this configuration you have the equivalent of two “L” shapes but you do not normally anchor the center of the middle leg. Guides are required for all three legs but again you need to allow for the displacement of the pipe due to the expansion of the three legs.

“U” – This refers to the Loop in a long straight pipe run. At the far end of the two legs attached to the “U” you must have Anchors. A Loop will not work if the ends of the straight run of pipe are not anchored. Between the Anchor and the Loop there needs to be Guides to control the movement of the pipe during the growth and shrinkage during operation.

Engineered Expansion Joints (EEJ):
These come in different styles from very simple to very complex. There is no reason for me to try to explain the different types and how they work. The web site below does that. The point I want to make here is based on my own opinion. EEJ’s are very costly and have inherent problems and can result in high maintenance and costly shutdowns. EEJ’s should never be the first choice and should be avoided when ever possible.

http://www.usbellows.com/products/metallic_expansi...

Sometimes its possible to do all the right things and still get bad results

Pennpiper does a great job in existing the issues.

Point 3 above - no such values exist as there are too many variables such as weight of pipe, weight of contents, friction of support, pipe material, types of guides or supports, installed temperature, operating pressure etc etc.

You can often find that the worse case is actually the empty solar case when expansion is high and friction low.

There is no real alternative to a proper design analysis.

As a real basic number for installation in ambient temps say15 C, you probably won't need expansion loops below 50C. But you need to do the work to confirm this.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Thank you all guys. I apologize for responding late. It was a holiday yesterday in our country so I was not able to check my mails.

All you guys have given great inputs. I appreciate your willingness to help. I am new in piping analysis projects so I just wanted to check for design practices that I may use as a basis for my initial calculations. Again, this is only part of my preliminary studies. I have read other references online and some design textbooks yet I cannot find a minimum (or maximum) values for the subject. It seems to me now that I need to perform the calculations to confirm.

Thank you guys!

rmdaniel,
You ask for the minimum temperature for when an expansion loop is necessary. I know that LNG lines at circa (minus)196 Deg C (lowest temp I have come across)normally require expansion loops.

[maxdistortion, while I would not say there is necessarily anything wrong in application with your practice, what you have typed twice in the 13:08 post does not appear to be a conventional expansion calc result - I think per conventional calcs and with a coefficient say of 0.000012 m/(mdegreeC), 100 m of e.g. welded steel pipe would really appear to change in length about 120 mm with the substantial temperature change of 100 degrees C - could this be typo and you meant e.g. 1.5 mm per 100 m per degree/delta C?)(See e.g. table on page 59 of file at http://www.american-usa.com/system/assets/265/orig.... Actually, I think the coefficients of thermal expansion of e.g. steel and ductile iron material are very similar, and for that matter quite similar to that of common steel and reinforced concrete buildings and support structures, that may be a convenient synergy in some applications.)]