Thermal Relief Protection for Short Sections of Pipe

[process440]

I have an interesting situation,
Heavy oil can be blocked in a 3 feet long section of 10" pipe.
Is thermal relief needed for such a short section of pipe?
I would really appreciate your input.

 

[Montemayor]

process440:

Why do you believe the short length has anything to do with whether the pipe/valves rupture under hydraulic pressure? The oil trapped between the block valves will expand with an input of heat and cause the pressure to increase almost exponentially and very quickly - potentially rupturing the pipe or valves (or both) if hydraulic relief is not obtained. That's the reason a thermal relief valve is installed.

Of course, you fail to tell us if there is a continuing heat input into that section of blocked oil. If there is no heat or energy input to cause the trapped oil to increase in its specific volume, then the thermal relief is not required and the trapped pressure remains the same.

Art Montemayor
Spring, TX

 

[Latexman]

For a constant heat input/foot, the length of pipe has very little to do with the pressure attained. A short run of pipe will reach as high a pressure as a long run under these conditions. However, the quantity of fluid relieved to maintain a safe pressure is directly related to the length of pipe.


Good luck,
Latexman

 

[unclesyd]

I would like to add one caveat to be on the look out for in addition to Montemayor's post.

This is an actual occurrence in a very similar situation where some liquid heat transfer fluid was trapped between 2 valves and for a time the line upstream and downstream were not charged and cold, no problems. As the line was being put back in service vaporized heating oil @ 300°C was valved into the piping on the downstream side of the closed valve. The operator decided after charging this line and while it was heating up went for a pit stop, then decided on a coffee break and probably stoped and killed several deer as it was opening of the deer season.
Being a vapor it readily condensed on the cold valve and you know the rest. The damaged section of the line also had a instrument (trashed) installed. The irony of this was that this section of line had a drain and this section was used as a double block and bleed system.

 

[25362]

Similar subjects have been raised many times in this and other forums. For a rough and rapid estimate of the pressures developed on heating a packed-in liquid divide the cubic thermal expansion, 1/^oC, by the isothermal compressibility, 1/bar.

For a heavy hydrocarbon oil one could expect a pressure increase of ~ 10 bar/^oC at ambient conditions.

If the oil is highly aromatic it would be less compressible, if naphthenic its compressibility would be higher, with paraffinics in between. Thus highly aromatic oils may have larger pressure increases for the same temperature rise.

Sometimes a high pressure may be accompanied by the separation of a solid phase, and as a result, the volume would shrink (i.e., showing a fictitious high compressibility) but I wouldn't count on this effect as a safety factor.

 

[Guidoo]

For me, best guidance on the subject is figure 6.1 in book from Cyril F. Parry: Relief Systems Handbook. According to this figure, you need a thermal relief valve assuming:

- source of heat is ambient temperature rise or solar radiation
- liquid is flammable (flashpoint below 100°F)
- system closure is frequent and/or system is located in unmanned off-sites area or on public property

 

[smckennz]

In some cases similar to what you describe, it is not necessary for one of the valves to have a tight shutoff. If this is the case, then a small hole bypassing one of the valves may be all you require. You do need to work it out though. There are countless examples of installations such as you describe and the only thing that saves them is seat or flange leakage or good luck.

Cheers

Steve

 

[25362]

Guidoo: I assume condition #1, doesn't preclude, for example, heating by heat exchange with a hot stream, and condition #2 is not exclusive, otherwise, water, among others, would be left out.

 

[Guidoo]

25362: Correct. Basically, what I did was mention the answers I gave when going through the flowchart as shown in the book from Parry. For example, when source of heat is process heat (as in a heat exchanger) or heat tracing, thermal relief is required, even when trapped liquid is not hazardous (for example water).

 

[dcasto]

Whatever case you select as decideing if a relief is required, skip the relief valve and put in a 1/4" tubing check valve around one end of the line. Most valve manufacturers of double block and bleed valves used to have body reliefs and today they use this check valve arrangement with a 1/4" tap in the body connected to a 1/4" tap in the flange.

 

[Latexman]

dcasto,

That'd have to be a spring loaded check valve, right?


Good luck,
Latexman

 

[dcasto]

Latexman, Yes and you can specify the delta P it takes to fully open the valve.

 

[jalvarez]

Latexman and dcasto:
Be cautious with these kind of solutions.
Normally, if you have two valves in a single pipe, they are used (as unclesyd described in his real life example) as double block and bleed to assure a positive blocking of a pipe or equipment. Using a check eliminates the risks of thermal expansions, but the possibility for an operative mistake increases. They can forget the check existence, get the false idea of having blocked the system with a single valve. The Operation Procedure must be revised, to assure the use of both valves, in this case. And take into account that a check valve never assure the blocking for a reverse flow.
process440: Another issue that was not mentioned here is the selection of the set pressure for the Thermal Expansion Relief Valve (if this is the final solution). It should be specified as high as possible, like 1.5 times the design pressure of the equipment being protected (depending on local codes). This is to avoid unexpected releases.
For your specific case, why do you have a double block? This is the question...
Regards
jalvarez

 

[Latexman]

jalvarez,

I don't have any double block valves in close proximity! The original poster does, process440. Follow the threads!


Good luck,
Latexman

 

[Latexman]

jalvarez,

Sorry, after re-reading your post more carefully, I realized that I misunderstood the first time I read it. My apologies. Your post is right on!

I'm curious though. Is english your mother tongue?

Good luck,
Latexman

 

[jalvarez]

Latexman: Signing jalvarez, what would be my mother tongue? You've caught me, it's Spanish. I was proud of my English... let's say till now...
Another issue related to thermal expansion, maybe for everybody discussion, is the one I've just posted in the forum, an extension of this one.
Have a safe day
jalvarez