7 replies to this topic

[Shine KK]

Hi all, I am relatively new to this forum, I am process engineer.
can any one help on the following.
A relief valve tail line need to be sized, which is routed to a drum operating at 345kPa.
The RV is installed on a plunger pump disch (Chem injection) and it is sized for blocked outlet scenario.
The relief is assumed to be happening at ambient temperature (less than 35deg C), and the relieved fluid is in liquid state.
I know, generally we choose one size bigger than the RV outlet size.
I need to come out with a calculation to show the selected pipe size is adequate to handle the RV's rated flow.
I went thru API 520/521 but liquid relief is not mentioned there (API mentioned only thermal relief).
Is there any guideline available for this situation.. like max velocity/max pressure drop for tail line?
What equation shall i use to size this tail line? (btw i am familiar with sizing tail lines for gas relief)
can i use eq: 2.2 fr API 14E to find out the pressure drop of the tail line?
Any help will be much appreciated, thanks in advance.
Shine

Edited by Shine KK, 18 April 2012 - 03:11 AM.

[fallah]

Shine,

The superimposed back pressure in discharge line is 345 kPag. You can calculate build up back pressure by back calculating the pressure loss (using classic equation of liquid pressure drop calculation) from the drum (pressure=345 kPag) back to the PRV downstream (flowrate=rated relief load). If superimposed back pressure plus build up back pressure would be equal or lower than 10% (suppose the valve is conventional type) of the PRV set pressure, the selected pipe size is adequate to handle the PRV's rated flow otherwise you should increase the pipe size and check again till you meet that criteria.

Fallah

[jrtailor09]

Dear Shine ,

As per API guideline tail pipe shall be sized for the maximum mach no. of 0.7.

This is only criteria for the tail pipe line sizing.

correct me if i am wrong.

Regards,

[ankur2061]

Fallah,

Thanks for the nice explanation and I totally agree with your explanation.

Shine K K

Use the Darcy -Weisbach equation for frictional pressure drop at the rated capacity of the relief valve for a selected pipe size. Add this value to the superimposed backpressure of 345 kPag and check if it is less than 10% of the relief valve set pressure if your relief valve is a conventional relief valve. If it exceeds this value go to the next higher line size and recalculate. Repeat this exercise with bigger line sizes till you get the superimposed backpressure + variable backpressure (frictional pressure drop due to flow) less than 10% of the PSV set pressure.

Jatin,

PSV tail pipes are not sized based on Mach No. and definitely not for liquid relief. The backpressure criteria is the one that applies for sizing of tail pipes. In most cases where there is practically no backpressure in gas relief (atmospheric discharge or atmospheric flare), the velocity goes to sonic anyway since the actual pressure ratio (downstream pressure divided by the upstream pressure) is less than the critical pressure ratio where critical pressure ratio is defined as:

P_crit = (2 / k + 1)^k/k-1

where:
k = isentropic expansion coefficient

To conclude the sizing of PSV tail pipes has to be done on backpressure at the outlet of the PSV nicely explained by Fallah.

Regards,
Ankur.

[CMA010]

A lot of companies also size PSV tail pipes for vapours/gases based on mach numbers besides the back pressure criteria. As pointed out earlier a criteria of a Mach no. of 0.7 is often used (or 0.3 for atmospheric relief).

[kkala]

Well, concerning Mach number not higher than 0.7 for vapors/gases: This criterion aims at reduced noise levels during any emergency discharge to flare. So this is additional to basic requirement that (for a conventional type PSV) gauge downstream pressure should not be higher than 10% of PSV set (gauge) pressure.
"Do not exceed 0.7 of sonic velocity at point of lowest pressure", according to what we apply here.
Atmospheric relief of 0.3 mach has not been met before, some more data (rationale, supporting practice, etc) would be welcomed. For flare tips look at http://www.cheresources.com/invision/blog/4/entry-89-40-cfr-section-6018-us-epa-flare-tip-exit-velocity-and-flare-tip-diameter/.

Editing note, 13-May-12: "10% of" added.

Edited by kkala, 13 May 2012 - 05:38 AM.

[ankur2061]

kkala,

Noise is of the least concern during a discharge to the flare. The most important concern is to ensure that the backpressure criteria for the various types of PSVs (conventional, balanced-bellows and pilot-operated) is not exceeded which may prevent the PSV to discharge at its rated capacity thereby aggravating an overpressure situation in the vessel / equipment to which the PSV is connected.

Regards,
Ankur.

[kkala]

Local refinery practices call for 0.7 of sonic velocity (max) at point of lowest pressure, to reduce noise levels during an emergency discharge to the flare. This concerns discharge PSV lines and main header. And this is supplementary to the main requirement of max back pressure, depending of PSV type. This 0.7 Mach (max) is also reported in posts No 3 and 5.
We normally use Flarenet to check the tree of discharge network. The 0.7 Mach (max) seems to be also an additional safety margin to avoid sonic flow, despite calculation uncertainties.
Concerning the main flare header in refineries (and subheaders), I believe that extra margins in diameter had better be considered in design phase; new discharges will be introduced in the future, not probably covered by existing scenarios. This problem has been faced here, see http://www.cheresources.com/invision/topic/13556-designing-offsites-to-be-used-for-multiple-plants, post No 8, para 7.

Edited by kkala, 13 May 2012 - 06:29 AM.