Pressure rise due to simultaneous heating of the pipe and blocked-in liquid

[crispe07]

Hi,
I am working in a project where there is a pipeline filled with water and traces of glycol at 50ºC. That pipeline can be blocked and the client suggests a thermal relief valve. I am trying to justify if this thermal relief valve is needed or not.

As per API 521 (6th edition, 2014) section 4.4.12.4:

Where the system under consideration for thermal relief consists of piping only (does not contain pressure vessels or heat exchangers), a PRD might not be required to protect piping from thermal expansion if any of the following apply:

c) The fluid temperature is greater than the maximum temperature expected from solar heating [usually approximately 60 °C to 70 °C (approximately 140 °F to 160 °F)] and there are no other heat sources such as heat tracing (note that fire is generally not considered when evaluating pressure-relief requirements for piping);
If we consider 70º as the maximum temperature expected from solar heating (our pipeline is not in the upper part of the rack, but we cannot say which is the maximum T expected), we can't consider this point as our case.

d) The estimated pressure rise from thermal expansion is within the design limits of the equipment or piping.

I used the equation (3) from API 521 (6th edition, 2014) section 4.4.12.4 for pressure rise due to simultaneous heating of the pipe and blocked-in liquid and I am getting a pressure rise of 49 kg/cm2 by the increase of 1ºC of temperature.

I also calculate the temperature rise due to solar radiation (taking 1.04kW/m2 for intensity of solar radiation in accordance with API 521) and I am getting a temperature rise of 5ºC, resulting in a pressure increase of 245 kg/cm2 based on the results of pressure increase I mention previously.

This does not make much sense so I am wondering if this is correct.
Does anyone have any thoughts?

 

[Latexman]

My company design manual says ambient heating of a 4" Sch 40 pipe at 20^oC will rise about 47 psi/^oC when filled with water, and about 197 psi/^oC when filled with ethylene glycol.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[LittleInch]

Looks to me like you have a factor of 10 error unless you have a very temperature sensitive fluid. A figure of 5 bar per deg C isn't unexpected.

see

http://www.eng-tips.com/faqs.cfm?fid=1339

for further details.

The issue of solar radiation is always a little odd as you need to somehow account for heating of the fluid over the period it is exposed to solar radiation. Most people don't bother and just stick in a thermal relief valve.

At 50C, it might not actually gain much temperature, but depends on size, insulation (if any) surface length exposed to the sun, surface length not exposed to the sun, diameter, pressure rating and amount of pressure you can spare.

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

 

[bimr]

Here is a article from Hydrocarbon Processing for reference:

 

[georgeverghese]

Since this is a nonflammable/ non toxic liquid, I wouldnt bother with a TRV - most block valves are not strictly leaktight when closed, especially metal seated valves, and more so if they are gate valves.

 

[crispe07]

Thank you all for your answers . We finally decided to install de PSV.
One more question: Would it be the same conclusion if the fluid is only water (with no traces of glycol)?

@georgeverghese: the large of the pipe is about 250m... you still think that the PSV is not necessary? If it is not necessary, how could I justify it? The problem is that as per API's formula, the PSV is needed (althought maybe it is not), but I need to justify it with some calculation or bibliography.

 

[LittleInch]

Only water will give you a smaller pressure rise, but we don't know how much "spare" pressure capacity you have.

I rarely disagree with George, but stating that you don't need a TRV because the valves will definitely, 100% leak a bit (enough) is not, IMHO, a good idea and not one which would get past most authorities / reviews. As it's water glycol, I can only imagine this is a pretty low pressure system?

BTW, thanks for coming back and telling us what you decided - the " thanks for all the information" one line response doesn't really add anything, so well done on your first post.

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

 

[georgeverghese]

You only need a fraction of a drop to relieve in order to dissipate the pressure built up from solar radiation. You got tiny leak points at the valve packing stems, flange gaskets, and bigger leak points at the valve trims, especially if these are not soft seated. My previous company standards tell me there is no need for a TRV in this non flammable service for any line size.

 

[mk3223]

It could be a Client call for deciding no TRV for the system. But, IMO, it's a good engineering design with a TRV for protecting the blocking system. According to the Code, the valve may be fabricated with an allowed leakage rate, but it doesn't intent for the system overpressure protection.

 

[LittleInch]

George,

The OP has 250m of an unknown pipe size. That's going to take more than a "fraction of a drop"...

I've seen guidance where if the volume or length of pipework is below some magic number then they don't install TRVs, which probably rely on the same principle.

Water glycol mix expands more per degree C than plain water.
I've sat next to class 600 piping manifolds in the sun with the pressure guage at 100 bar and constant hissing nose from the TRVs. It was very quiet at that point but it took me a while to work out that there wasn't a leak somewhere.

I accept that few valves seal 100%, but my point is that I would not like to rely on it.

Just because your previous company standard says it is ok for non flammable service doesn't necessarily make it correct and after one failure I would imagine they might change their mind. Just because it's non flammable doesn't make it non lethal or non important to the operation of the plant. One failure could cost 1000 TRVs. I know which one I would choose.

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

 

[don1980]

If you have an isolated liquid-filled system, then the only question you need to answer is whether that fluid can be heated. If the answer is yes, then over-pressure will occur. Attempting to calculate the amount of pressure increase is an academic exercise that has no practical real-world value. If the fluid can be heated, then you need to provide an outlet path for the liquid expansion, or provide a relief device.

The best solution is to ensure that liquid-full systems are not completely isolated.

 

[georgeverghese]

@little Inch,
Those hissing 600lb TRVs' were probably leaking for some other reason - the fact that were hissing for some time says there is something else going on there..

BTW, is the expansion volume of this blocked in volume less or more than the volume expansion of the pipe ? If it is less, then you dont have a overpressure problem at all.