Thermal Expansion in pipes 1

[raingal79]

Dear all,

I would like to ask about liquid thermal expansion in a blocked section of pipe. I'm wanted to calculate the estimated pressure rise in a pipe and have searched through the internet for information. I found that we can do so by using:

dP = expansion coeff * dT / isothermal compressibility

However, it proved to be very challenging to find out the coefficient of thermal expansion (/degC) and isothermal compressibilty (/bar) values for common liquid e.g. acetic acid, ethylene glycol etc. IS there somewhere I could retrieve these info?

Another discrepancy - is there a difference between coefficient of cubical expansion (0.00021/degC for water) VS coefficient of thermal expansion (1/degC for water)? It makes a lot of difference to the estimated pressure rise calculation.

Appreciate some advice. Thanks.

 

[25362]

Why should it be in the internet ? Isn't the CRC Handbook of Chemistry and Physics, as an examplr, a sufficiently good source ?

 

[vzeos]

If you want to calculate the pressure increase, you may also want to consider the thermal expansion of the pipe material and its elasticity.

 

[raingal79]

I have looked through CRC handbook but couldn't find also. I looked under the appendix - for liquids.

Any advice on the 2nd question regarding the difference between coefficient of cubical expansion versus coefficient of thermal expansion.

 

[25362]

To raingal79, have a look at the table titled Pressure and temperature dependence of liquid density, page 6-139, on the CRC 77th edition.

 

[JStephen]

What I remember about the CRC Handbook of Chemistry and Physics...it has everything EXCEPT what you need to know!

 

[raingal79]

if that's the case, appreciate someone can share how to determine for a blocked section of liquid in pipe, whether or not thermal relief v/v is required. thanks.

 

[blackwed]

Once I saw the pressure in an 8" snow-white jet fuel line go from 0 to over 600 psi in about 10 minutes and thought, now there's a rabbit I can chase. I finally concluded that much more than a simple qualitative understanding of the issue was superfluous (at least, for me). You basically get a whole boat load of pressure increase for just a wee little increase in temp if your line is truly packed, i.e., 100% liquid filled. For an above ground pipeline the error in monitoring and/or controlling the temp introduced such a wide variation in the resultant pressure change that a quantitiative analysis was unnecessary, not to mention wildly divergent from field obserations. You'll need to be in some very closely controlled conditions for any calcs to be meaningful.

I think it's a good idea to employ pressure relief in lines that can be isolated and will be subject to temperature deltas.

DB

 

[MJCronin]

raingal,

A discussion of "Thermal relief" and the methods to size valves for blocked in piping segments is addressed in API-521.

See:

http://www.eng-tips.com/viewthread.cfm?qid=124051&page=5

blackwed,

Not only is thermal relief a "good idea" it is (indirectly) required during the Process Hazard Analysis (PHA) commonly performed on most chemical plant systems.

I have always found it difficult to determine what group of chemicals in which piping systems require thermal relief valves....for example, chlorine, LNG, and any refrigerated gas are no-brainers....however, try to find a definitive guidleine on liquids such as ethylene oxide, propylene oxide, etc...

My opinion only...

MJC

 

[blackwed]

"Good idea" was as far as I felt comfortable going with anything like a recommendation since my electrical background has so far allowed me exposure to the characteristics and properties of merely a handful of refined petroleum products. Y'all routinely work with myriad other liquids with which I have absolutely no experience.

DB

 

[EGT01]

Raingal79,

Here's some references I've collected....

"Calculate Thermal Expansion Coefficients", Carl L. Yaws, Chemical Engineering, August 1995, pages 98-104.
Contains a listing of thermal expansion coefficients for about 350 different liquids and suggests methods how to estimate thermal expansion coefficients.

"Decide Whether to Use Thermal Relief Valves", Fabio Bravo and Brent D. Beatty, Chemical Engineering Progress, December 1993, pages 35-38.
Another good article with some thermal expansion coefficient data and some reference to expected effects of solar radiation.

"Evaluating thermal expansion/overpressure protection", Chip Eskridge, Chemical Processing, August 15, 2001.
This article was written to address "Does this line [i.e., pipe] need a pressure relief valve [PRV] for thermal expansion/overpressure protection?". Equations are included to determine the pressure rise based on the liquid thermal expansion coefficient and liquid compressibility. There are tables of these values for a small number of liquids. Not long ago you could download this article from the Chemical Processing web site but I don't find it there now.

http://www.chemicalprocessing.com/cp/index.html

"Guidelines for Pressure Relief and Effluent Handling Systems", AICHE, CCPS, 1998.
Excellent all around reference for most all topics regarding pressure relief. In regards to your needs, there is a section discussing how to determine pressure rise from thermal expansion. They reference the Handbook of Chemistry and Physics (Lide 1997) as a source for liquid compressibilities.

Perry's 6th edition, page 3-119 has a table for Compressibilities of Liquids.

Last but not least...

"Relief Systems Handbook", Cyril F. Parry, Institution of Chemical Engineers, 1992.
This is another good text book for relief systems but is somewhat small considering the subject. Nonetheless, in regards to your needs, Parry offers a very detailed decision tree to follow for establishing whether thermal relief is required.

 

[handee]

Great references! Thanks.

raingal79: It's probably not worth doing the calculations. Just recongnize that the pressure rise can be hundreds or thousands of psig if a dead-ended liquid-filled line heats up. Use pressure relief. The expansion coefficient is handy to estimate the relieving flow rate to size the valve.

Fran McConville

 

[TSeener]

Raingal79,

Hi. Just a quick note on what I have used successfully in some of my previous projects. The key is not to calculate whether or not a relief is required, but merely to prove that the relieving capacity is going to be adequate. For the thermal expansion, we almost (there was one case out of hundreds) always select the smallest relief that we could find (3/4” x1/2”), and then the calculations support that.

If you are dealing with water... refer to the density / pressure data,i.e. steam tables. If you set your relief at 150 psig (defined by the weakest link in your system and the pipe specification), you can get the pressure, temperature, and density (or specific volume) from hundreds of sources (but I mention the steam tables, because it is so quick). The volume of your system, should be easily derived, which will then give you the total volume to be passed via the relief. You will use the starting temperature and starting volume (fixed) versus ending volume and temperature. You will see that there is usually a very small increase over a large system.
The next is the time aspect of this calculation, and I would refer you to NFPA 30. This will give you the heat generated from a fire case, which is usually the worst case. Use the heat to determine your heat up time in this isolated section of the piping system. Then you should be able to get your volumetric flow, which I would recommend adding 20%. With your volumetric flow, go back to the ASME code on sizing relief valves and determine your minimum required orifice size. The smallest is usually more than enough

The expansion of the piping system will occur as well, but this type of a calculation is done just to prove to the client that their interests are protected.

I hope that this helps you in your project. Sometimes using the simple basics are all that is required…


Best Regards,

TSeener