Thermal expansion on cooling water pipes 1

[abcmex]

I would like to know the experiences with the design of cooling water pipes on pipe racks specially the effects of thermal expansion.

I heard different opinions:
"...no need, the pipe operates most of the time at the same temperature." others "...take at least 50 deg C due to summer/winter effect"

If the design would have free pipe-end I would not consider this a problem but this is not the case. Do I need to install expansion loops?. For 40" it will cost some money!

PD. Do not worry I will get this calculated with some stress analysis program later. ;-)

Kind regards

 

[pmover]

simply stating: one should consider the installation or initial fill temperature of the pipe and the operating temperature of the piping system. naturally, for higher temperature differences, there will be thermal expansion (heating of pipe) or contraction (cooling of pipe) of the pipe system.
any engineering reference will provide you the necessary data to conduct thermal growth/contraction analysis.
-pmover

 

[Thomasjl]

Pipes do expand and subtract. Allowable stress on nozzles of coolers, pump and other equipment are limited. To avoid such stresses and/or to keep them under control you need anchors. Between two anchors the pipe must be able to expand and subtract. So if your system is short and has many (ninety degree) elbows you might not need special precautions. Otherwise you need expansion loops, no matter the pipe is cooled, heated, from inside out or due to external (sun) conditions.

It is very wise to have your system checked with stress analysis, don't wait for later, do this as soon as you can, before the rest of the design is too far to find nice alternatives.

Good luck.

 

[25362]

If by chance the pipe may be blocked in with valves, a water heat up in a blocked-in pipe, may develop pressures of about 5 atm per degree C, or more. Thus the installation of a pressure relief valve may be advisable.

 

[abcmex]

Thanks for your reply!

The pipe is going to be installed in a pipe rack and therefore the the inherent flexibility is probably not available " ..short and has many (ninety degree) elbows".

The piping design is an iterative process:
(1) assumption (geometry, w/loops or w/o loops, fix/slide supports) etc.
(2) calculation (Caesar II ect).
(3) stress value adequate Yes/No go to 1.

Therefore what I would like to know is if in a straight pipe rack between two fix points for a cooling water pipe usually a expansion loop is installed?
Kind regards

 

[StressGuy]

If you've got an anchor at each end of the rack, you'd better have a loop in between then.

Depending on the length, it is more common to anchor the middle of the run and use the changes of direction at each end to provide flexibility.

40" water pipe is definitely something to take seriously.

While not related to your thermal expansion question, you are also going to want to be mindful of local stresses at the support points when you've got a line that big full of liquid. You may need bearing pads on the bottom of the pipe at the supports.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

All opinions expressed here are my own and not my company's.

 

[jay165]

Listen to StressGuy. If it's in the sun and painted a dark color, that can have an effect, too. Do the calcs and you'll be surprised how much force a 40" D pipe can exert even when there are "small" temperature changes.

 

[DLANDISSR]

abcmex:
Why are the ends fixed? Are these flanges to equipment or are they fixed on the pipe rack? How are the take offs to equipment arranged?

If the ends are fixed on the pipe rack and not flanged to equipment the ends could be left unfixed axially thereby allowing some movement. If they ends are fixed by connection the equipment an expansion loop may be the only answer.

Assuming this is carbon steel pipe on a cooling tower the water temperature could be anywhere from 60 to 95 deg. The pipe if in the sun could be over 100deg and at night could be 40deg. This would mean a temperature swing of 40 deg. (I am only suggesting a possible situation.)

Carbon steel pipe will expand or contract .0045"/foot. If this 40" dia pipe is 100ft long it will expand or contact .0045x 100 = .45".

The question is can what ever the pipe is connected to absorb .45" of movement without damage? Do the branches off the main line have long enough offsets to absorb the movement so that the forces from the bending of the branch will be low enough not to damage anything they are connected to?

The bottom line is the system should be carefully anmalyzed to assure that the expansion and contraction of the system is properly absorbed and not transmitted to the supports or the equipment to which the piping is connected. It cannot be assumed, given the size of the pipe, the straight run of the pipe and the end conditions that there is not need for an expansion loop.

As mentioned above if the water in this pipe is blocked in and expands the resulting pressure could rupture the pipe. If the water is blocked in a closed portion of the pipe a relief valve should be provided.

Given the size of the pipe reinforcement of the piping may be required at the support points.

Also if this pipe can be subjected to vacuum conditions additional reinforcing may be required to keep the pipe from collapsing.

Also have you considered the effect of turning on and off pumps or closing valves. This may cause water hammer. The pressures and forces generated by water hammer could damage the piping, supports or equipment to which the piping is attached.

If this is an existing pipe rack, has it been designed to support a pipe of this size. A 40" pipe (.5"wall) would weight about 330lb per linear foot. A 100' long pipe would weigt 33,000lb. Also what seisemic Zone are you in. Is the pipe rack designed to support an additional load of 330lb per foot during an earthquake?

And then there is wind load also? Is the pipe rack designed for the additional load, earthquake load and wind load?


All things considered this design should be looked at very carefully!!!

 

[abcmex]

Yes, its a big pipe and has its troubles, we study measures against the described problems already.

The question here is regarding the usual handling of thermal expansion in a pipe rack if you have long straigth runs of such a big cooling water pipe (more than 1000 ft).

Any experience with sliding expansion joints?
kind regards

 

[LSThill]

abcmex (Mechanical)

The City of Houston, Texas has the experience with sliding expansion joints.

Leonard@thill.biz

http://www.thill.biz

 

[DarthDick]

abcmex,

regarding your question "the usual handling of thermal expansion in a pipe rack if you have long straigth runs of such a big cooling water pipe (more than 1000 ft)"....

Add a loop. Maybe two. 1000 ft is quite long. However, for 40" D pipe, the loop leg may be unwieldy.

I'm leary of sliding expansion joints, especially if The City of Houston uses them (I live in Houston and worked for them for a VERY short time).

 

[DLANDISSR]

abcmex:
Sliding expansion joints would be a joint like the packing around a valve stem. The pipe will slide through the packing when it expands.

When considering this type of expansion joint keep in mind there are pressure forces for which to provide. For example
a 40" dia line @ say 50PSIG would have an axial pressure thrust: 50 pounds per square inch times the area of the pipe 1256square inches = 62831lbs. If a sliding expansion joint is used this in a straight pipe that is anchored at each end, for example, the anchors would have to resist 62831lb pressure thrust. Even if tie rods are used between the flanges across the joint, the pressure thrust will have to be overcome before the joint would move to allow movement for expansion. The minimum force that anchors would have to resist is the pressure thrust.

You should calculate the pressure thrust first. Then add the friction of the slide joint (get data from the manufacturer), then add the friction of the sliding pipe supports. This is the force that each anchor must be designed to resist. I would use a 3 to one safety factor.
Given the data in the example that would mean that the anchors would have to resist in excess of 200,000lbs.

Do you really want to do this????

If you are looking for a short easy answer there is none.

As I said before this system needs to be looked at very carefully. You cannot use a general statement or rule of thumb like "expansion loops are or are not normally used on cooling wate systems"!!! This system must be engineered!!!!