Piping stress analysis

[francis2013]

Dear all,
I am new to write in this forum. I spent some time on it and I found very good quality and interesting discussions. I would like to open a discussion on piping stress analysis. Maybe there is already a discussion about what I am writing, but I do not have enough time to go through all the posts. I have to say that I am an engineer new on piping. My colleagues asked me to perform a pipe stress analysis and I do not have lot of time. I already read some books that suggest to divide the pipes in simple supported beams in order to simplify the calculations. My principal point is about the constraints induced by the support, in particular for U-bolt. My colleague suggested me to consider that the U-bolt will allow all the rotations. Than all the translations are also blocked. In particular regarding the longitudinal axis the friction load from the tightening of the bolts is able to constrain the weight and the seismic loads, but will allow the thermal expansion of the pipe. Do you think that these assumptions are correct? I ask this because I found on literature different contrasting questions, and I think that the experience could count even more than books.
Thank you all in advance
Best Regards

 

[BigInch]

U bolts won't do much other than keep the pipe aligned when there is no load. First sign of any significant load will normally fail a U bolt.

I hate Windows 8!!!!

 

[ConstantEffort]

If you're talking small pipe (NPS <=2) and cold temperatures (<= 200°F) where the consequence of failure is negligible, then go for u-bolts and assume whatever you want.

Since you mention stress analysis I'm guessing larger pipe and hotter service. U-bolts can be safely used still but some extra thought is required. Assume that u-bolts alone are not going to hold any longitudinal forces. If you do plan on tightening the u-bolts down, make sure that, if it does hold, you're not sending the growth back on anything sensitive to load. I would also carefully check to ensure lateral forces are within reason.

FYI, Anvil publishes a maximum side load value for their Figure 137 u-bolts.

For longitudinal stops with a u-bolt, you can weld a lug or lugs onto the pipe adjacent to the u-bolt. Looks ugly, can't take much load, but sometimes that's all you need.

 

[francis2013]

Thank you for all the answers. Probably I used some terms in not a proper way. The pipe I am going to use is a 2.5" (3.0" external) and the temperature increases from 20 to 30 Celsius in 30 m. I would not like to weld something on the pipe. Is it correct if I assume that the pipes are free move longitudinally on the U-bolts due to the thermal expansion while the loads such as the weight (15 kg/m) in a vertical pipe can be supported?

 

[ConstantEffort]

Leave a gap between the top of pipe and the u-bolt. 4 nuts per u-bolt, 1 on each side of the steel for each leg. Then you can assume it is free to move longitudinally on a properly supported line. Use a support span chart here and move on.

 

[francis2013]

Thank you again for the answers. I understand that stress analysis require little time to be improved and it is not an easy task as lot of people think!! I will try to be more specific. Unfortunately I do not have the possibility of doing drawings. I have a 2.5" water pipe with water from 20 to 35 Celsius. the pipeline can be divided in 4 parts:

1. A first vertical segment of 3 m.
2. A first horizontal segment of 30m.
3. A second vertical segment of 3 m.
4. A second horizontal segment of 10 m.

My colleague's idea is to use U-bolts for the vertical and horizontal segments, considering ASTM 31.x as guideline for the distances between the supports. You mean that if I have a longitudinal load (like a seismic one) the U-bolts on the horizontal segment do not provide reactions and only the supports on the vertical segments (like 13 kN!!) can bring it? It seems so strange because I have seen many pipelines even with bigger size than 2.5" supported by simple U-bolts. Are all these pipelines design wrong?

 

[ConstantEffort]

How many of those u-bolt piping systems have you seen go through a code-strength seismic event?

I think you're over analyzing a small bore water line at near ambient temperature. If the consequences of failure are so great, the higher cost a more robust support system should be easy to justify.

If you count the u-bolts as holding longitudinal movement to zero, what is the force required to constrain your 30m run as it expands with temperature?

I think what you want is a system that allows free movement in gradual (thermal) regimes but no movement in shock (seismic) regimes. There are supports that do that, and u-bolts aren't one of them.