## Featured Articles

Check out the latest featured articles.

## New Article

Product Viscosity vs. Shear

## Featured File

Vertical Tank Selection

## New Blog Entry

Scrubber Design for Desulfurization- posted in Ankur's blog

# Thermal Expansion Relief Valve

This topic has been archived. This means that you cannot reply to this topic.
31 replies to this topic
|

### #1 Sunil Pal

Sunil Pal

Brand New Member

• Members
• 4 posts

Posted 04 July 2007 - 09:51 AM

Dear All

Greeting.....

I want to calculate relief quantity due to solar radiation only. In this case i required area which is exposed to solar radiation. According to me area will be foot print area. Area (A) = Line Outer Dia (D) x Length (L) between two valve. Although the value of the solar radiation is generally in the range of 250 to 330 BTU/hr/ft2 (0.79 to 1.04 kW/m2). as per API 521, Pg. no. 41. In this case total total heat transfer rate will be = Foot print area x 330 BTU/hr. (maximum). Accordingly we can calculate capacity.
My reservation is, in actual semi cylinder exposes the solar radiation, in any case the exposed curve surface area does not more than half because solar radiation will be only one side even vertical pipe or tilted from y axis, in this case the half curve surface area will not be equal to foot print area then curve surface area will be = 3.14xDxL/2. But in this case the solar radiation will not be equal to every point on the curve surface area and radiation will be maximum only on perpendicular to the sun so we can not take 330 BTU/hr/ft2 value for whole curved surface area. So better to take foot print area. Although it will not make much difference in relief capacity. In API 521 page 13,14 The total heat transfer rate (H) BTU/hr. For calculating H, API considered volume expanded by the solar radiation due to radiation , conduction and convection, but i am not sure for the conduction which area they have taken, whether foot print or semi cylinder.

Your's comment will be always expedite.

With Regards

Sunil Pal
EEC, Mumbai

### #2 Art Montemayor

Art Montemayor

Gold Member

• 5,528 posts

Posted 04 July 2007 - 10:51 AM

Why do you want to calculate the relief quantity due to solar radiation only?

And what is it that you are calculating? Is it a pipe? Is it a 100% liquid-full vessel?

Is this a thermal relief valve application?

If it is a thermal relief application, you normally do not have to calculate anything. Simply install a conventional 1/2" or 3/4" diameter thermal relief valve and you're done.

Please explain what it is (or what you think) that you are confronting.

### #3 Sunil Pal

Sunil Pal

Brand New Member

• Members
• 4 posts

Posted 06 July 2007 - 03:08 AM

It's vendor requirement, Vender does not accept Material Requisition (MR) for the Thermal expansion relief valve without relief quantity. However 1/2" or 3/4" is the inlet size of the of the pipe and for the calculation of orifice area we have required relief quantity. I admit mostly the valve size 3/4" or 1/2" inlet but being a design engineer, it does not necessary the valve always will be 1/2" or 3/4" only. After the calculation whatever quantity comes the size should be based on that. API does not says that for the thermal radiation due to solar radiation for sizing the valve relief quantity is not required. Primiraly this problems comes in pipe only. In the last post we have calculated for consideration of pipe.
In same time without relief quantity, we can not find out the number of devices if the line is having higher dia and long length or we can not justify provided valve adequate or not.
Yes , this is thermal relief valve application.

### #4 Art Montemayor

Art Montemayor

Gold Member

• 5,528 posts

Posted 06 July 2007 - 06:38 AM

Sinil:

You say it's a thermal relief application; then, it only requires a nominal ½" or ¾" thermal relief valve. That, in USA practice is the accepted, legal engineering norm. I don't know what the requirements are in India. We don't calculate the relief quantity here in the USA, and much less the relief rate.

If you look at the size (volumetric capacity) of your "equipment" – which you don't identify – you will probably find out that the theoretical relief quantity due to radiation is probably so small (a few cm3/min) that a formal calculation is considered a ridiculous waste of time. However, we don't know the size of your "equipment", so there is a remote possibility that you may have need for a larger relief valve or rupture disk.

### #5 pleckner

pleckner

Gold Member

• ChE Plus Subscriber
• 564 posts

Posted 06 July 2007 - 05:38 PM

@Sunil Pal

It is still not clear if you are talking about the thermal relief for a pipe, pipe line or for equipment. Please clarify.

For pipes that are not considered pipelines of considerable length and diameter, and considering only solar heating, then I will disagree with you, a 3/4" x 1" PSV will always be more than ample for protection. As Art said, you are not required (at least in the U.S.) to show a calculation to prove this. The orifice calculated from the required relieving rate will not even come close to the smallest PSV you can buy so why bother? If you want or need to consider fire, then yes, you can possibly require a significantly larger PSV than the 3/4" x 1".

If you are so intent on calculating the heat requirements, then you only need to grab a copy of Perry's Chemical Engineering Handbook. I believe there is a procedure given for solar heating.

I don't understand your vendor as they are not responsible for determining the PSV size required, you are. They should supply what you ask for. If your vendor insists on a relief capacity and to save you some time, then take the stamped capacity of the particular 3/4" x 1" PSV you want to purchase and use that.

You also have to understand what is happening here and why you want a PSV for thermal relief in the first place. If the pipe is completely full, there is no place for liquid to go whan it begins to increase in volume as the temperature goes up. Liquid expansion is very slight but creates a tremendous amount of pressure when it does expand. The PSV allows a certain amount of liquid to burp out of the pipe. Once this "burp" happens, the pipe is no longer completely full. Therefore, the odds of a second relief happening is essentially nill after this since the liquid will not expand enough to refill the pipe. This is why for a scenario such as solar heating can be protected with just a minimum 3/4" x 1" PSV.

### #6 JULYSAN

JULYSAN

Brand New Member

• Members
• 1 posts

Posted 12 July 2007 - 03:57 PM

Greetings to all.

The following may be used to approximate relieving rates of liquid expanded by thermal forces where no vapor is generated at relief valve setting and maximun temperature. These calculation assume the liquid is non-compressible:

gpm = ( B )( Q )/500(G)(S)

typical values of the liquid expansion coefficient B are:

API Gravity*********Specific Gravity*********B, 1/ºF
water***************1.000****************0.0001
3 - 34.9*************1.052 - 0.850**********0.0004
35 - 50.9************0.850 - 0.775**********0.0005
51 - 63.9************0.775 - 0.724**********0.0006
79 - 88.9************0.672 - 0.642**********0.0008
94 - 100*************0.628 - 0.611**********0.0009

Q, BTU/hr = solar radiation (250 - 330 BTU/hr * sq ft) aplied to surface area
G = specific gravity relative to Air=1.0 or water=1.0
S = Specific heat (BTU/lb ºF)

I hope that this information help you. usually this value is accepted for completed the MR's

### #7 JoeWong

JoeWong

Gold Member

• ChE Plus Subscriber
• 1,223 posts

Posted 13 July 2007 - 06:31 AM

JULYSAN,
Good information...

The formula seem to me was extracted from API RP 1997.

I hope you are using latest API Std 521 Fifth edition, 2007.

JoeWong

### #8

• guestGuests
• 0 posts

Posted 02 August 2007 - 09:29 AM

Hi

I am using API 521 2007 ed

I have been asked to justify installing a thermal expansion relief valve in a gas condensate line

I have therefore tried to use equations 3, 4 & 5 to determine the increase in pressure if the line becomes boxed in.

However, the formulae do not converge

I have tried this graphically also

Regards

Ian

### #9

• guestGuests
• 0 posts

Posted 02 August 2007 - 09:33 AM

eqns from section 5.14.4

### #10 witono

witono

Brand New Member

• Members
• 1 posts

Posted 22 June 2008 - 02:59 AM

Mr. Art,
I agree with you. I have several TSV in our gathering pipeline (2 phase, gas and liquid), it is D orifice size. Problem is, I was requested to design a relief system to receive any relief from those TSVs? Does anyone has any experiences on how to designed a relief system for TSV? Should we provide a kind of vessel or can we release it to atmosphere?
Thank you,

Witono
- Indonesia -

### #11 anil seunarine

anil seunarine

Brand New Member

• Inactive Member
• 3 posts

Posted 07 July 2009 - 01:41 PM

Hey,

Has anyone been able to answer Ian_McQ post dated Aug 2 2007?

I have encountered the same problem.

Anil

### #12 fallah

fallah

Gold Member

• ChE Plus Subscriber
• 4,501 posts

Posted 08 July 2009 - 01:28 AM

Do you have in hand all data needed to calculate P2 by equation (3)?

If so,why didn't converge?

### #13 anil seunarine

anil seunarine

Brand New Member

• Inactive Member
• 3 posts

Posted 08 July 2009 - 12:28 PM

fallah,

The problem I have with equation (3) is that equation (3) requires the cubic expansion coefficient and the isothermal compressibility coefficient of the liquid. To determine these parameters using equations (4) and (5) the temperature and pressure is required for both the initial and final cases. But we do not have the final case pressure since this is what we are attempting to determine. Also density of the liquid for the final case is affected by the final temperature and pressure.

I have attached the information that I have for this calculation. Any assistance will be greatly appreciated.

Anil

### #14 latexman

latexman

Gold Member

• ChE Plus Subscriber
• 933 posts

Posted 08 July 2009 - 01:14 PM

Isn't the final pressure = sizing pressure = set pressure x 1.1 (10% overpressure)? Set pressure can = MAWP or less.

### #15 fallah

fallah

Gold Member

• ChE Plus Subscriber
• 4,501 posts

Posted 08 July 2009 - 01:33 PM

Anil:

Firstly you should be assured your case isn't among four cases defined in section 5.14.4.1 (as a,b,c,d),and also your system is subjected to simultaneously heating of the relevant pipe and blocked-in liquid.

Parameters in Equation (4) are pressure independent,and for calculation of x by Equation (5) you have to use trial and error procedure.Suppose a value for P2 and compare it with final P2 would be obtained by Equation (3).

With good estimations,i think an acceptable result (not accurate) would be obtained.

Personally,i am going to this area of calculations if i am being faced with a critical situation in safety point of view.

### #16 anil seunarine

anil seunarine

Brand New Member

• Inactive Member
• 3 posts

Posted 08 July 2009 - 01:56 PM

Thanks fallah,

My concern was that the final case pressure will affect the final case density and specific volume. Therefore, the final case liquid density and specific volume is determined at the initial case pressure and final case temperature?

Anil

### #17 fallah

fallah

Gold Member

• ChE Plus Subscriber
• 4,501 posts

Posted 08 July 2009 - 02:35 PM

QUOTE (latexman @ Jul 8 2009, 02:14 PM) <{POST_SNAPBACK}>
Isn't the final pressure = sizing pressure = set pressure x 1.1 (10% overpressure)? Set pressure can = MAWP or less.

P2 ( by Equation 3) plus a margin would be the set pressure.

Normally,overpressure in liquid thermal expansion is considered 25% of the set pressure.

### #18 jerald04

jerald04

Junior Member

• Members
• 17 posts

Posted 30 September 2009 - 06:53 AM

Hi all,

I've been looking through existing thermal relief valves datasheets in my plant.

I noticed that these TRVs are specified with 25% allowable overpressure, unlike 10% for non-fire cases and 21% for fire cases.

I've been trying to look up codes and B31.3, but couldnt find any information that states that TRVs installed on pipings allows an overpressure of 25%.

Does anyone knows where does this 25% comes from?

### #19 fallah

fallah

Gold Member

• ChE Plus Subscriber
• 4,501 posts

Posted 30 September 2009 - 02:30 PM

jerald04:

If you refer to API 520, part 1 (seventh edition), section 2.2.1.2.6, you will see the source of the TRV overpressure value.

### #20 jerald04

jerald04

Junior Member

• Members
• 17 posts

Posted 30 September 2009 - 07:08 PM

Fallah,

Thank you very much!

### #21 Nomanwer

Nomanwer

Brand New Member

• Members
• 7 posts

Posted 08 June 2012 - 03:10 AM

Hi all,

can any one answer me that i want to install a 3/4"x1" TRV on 10" piping. Pipe length is around 300m. Will it be sufficient or not?

Thanks

Noman

### #22 fallah

fallah

Gold Member

• ChE Plus Subscriber
• 4,501 posts

Posted 08 June 2012 - 08:12 AM

Nomanwer,

You fail to specify the state (guess to be liquid) and type of fluid trapped in the 300 m pipe segment and also the heat rate and source (may be radiation due to sunshine) by which the piping content would be expanded and needs to be released from TRV. By having heat rate and expansion factor of the fluid and other required information, it could be evaluated if the mentioned TRV covers the relief load due to thermal expansion or one/several 3/4"x1" TRV should be added along mentioned pipe segment.

Fallah

Edited by fallah, 08 June 2012 - 08:16 AM.

### #23 kkala

kkala

Gold Member

• Banned
• 1,939 posts

Posted 08 June 2012 - 03:35 PM

Referring to post No 6 (12 jul 07), we could assume B=0.0010 (supposing light hydrocarbon liquid), G=0.65, C=0.5 Btu/lb/oF.
Projected area of 300 m 10" pipe: (300/0.3048) ft * (10.75/12) ft = 882 ft2, Q=330 Btu/ft2/h*882 ft2 = 291000 Btu/h.
So discharge rate required B*Q/500/G/C = 0.0010*291000/500/0.65/0.5 = 1.8 gpm. Now the capacity of mentioned thermal relief valve can be preliminarily assessed, knowing its set pressure and orifice size.
Of course this would be more precise (without assumptions), if data requested in post No 22 is known and used.
It is noted that solar radiation input was considered on the projected pipe area (not on the lateral cylindrical pipe area), that is on a rectangle D*L, where D=external pipe diameter and L=pipe length. This seems reasonable. Clarification would be welcomed.
A liquid expansion formula quite similar to that already mentioned is reported in C R Branan's "Pocket Guide to Chemical Engineering", Gulf Publishing Co - 1999, Safety Relief Valve design (p. 31). Q is merely reported as heat input (Btu/h).
This formula can be used for sizing thermal relief valves (TRVs) for either equipment (e.g. exchanger with cold side blocked) or piping, even though piping (not process) engineers place TRVs on pipelines here by experience (without calculation).
A supplementary note, written on a local PID (evidently by Piping): TRVs shall be typically installed between two valves, in case that pipe length is longer than 30 m, or liquid volume is more than 1.0 m3.

Edited by kkala, 08 June 2012 - 03:46 PM.

### #24 Nomanwer

Nomanwer

Brand New Member

• Members
• 7 posts

Posted 11 June 2012 - 05:50 AM

Product is HSD and Mogas(petrol). Both in liquid form. Heat Source is sunlight. Do i really use the formula given in Post#6

### #25 fallah

fallah

Gold Member

• ChE Plus Subscriber
• 4,501 posts

Posted 11 June 2012 - 07:06 AM

Noman,

At first calculate the expanded volume of trapped liquid at the maximum temperature and then the time needed to raise to that temperature.

The relief load would be expanded volume devided by the time. If it is higher than existed TSV capacity you should add another TSV(s) till you meet the calculated relief load.

Fallah