Common tail pipe size for multiple PSVs

[Sawsan311]

Dear All,

for a system have 2+1 standby PSVs with staggered arrangement as per ASME U-134. the first PSV is set at MAWP and the second is set 105% MAWP. Bot PSVs have a common inlet from the vessel which will be designed as per API 520 PII such that the area of this inlet pipe shall be at least equal to the sum of the inlet flange area of the twp PSVs i.e. section 5.2 and such that PSV inlet non-recoverable losses of 3% of pressure drop is not exceeded.

For the sizing of the individual PSVs, the rated capacity of each will be used for calculating the outlet pipe such that the area of the outlet individual pipe is also equal to or greater than the outlet flange area of the PSV. However, if the two outlets combine into a common collection tail pipe before joining the main flare subheader/header.. what should be the capacity basis for that common collection pipe backpressure estimation and hence its sizing:

a- the maximum rated capacity out of the two knowing that rated capacity is a function of the overpressure of the PSV and its set pressure.
b- the sum of the rated capacity of both of the PSVs even if the opening is sequential. We understand that when PSV opens it goes to full lift at its rated capacity as long as the bore area exceeds 1/4 of the nozzle diameter. Is therefore required to consider both rated capacity of both of the operating PSVs since Engineer can't have assurance of the set pressure tolerances of the second PSV and whether it would sequentially open following the sudden full lift opening of the first PSV?? Do you think that option b is clarified by API 521 section 4.9.4.3 of the need to have an aggregate flow consideration irrespective to the staggering due to the above mentioned tolerances

thanks

Regards,

 

[LittleInch]

I think b) is the option to use for the reasons you state. You need to consider the worst case which is both going at the same time. Otherwise why have two PSV's?
The only other option is to consider what the maximum possible flow from upstream is at the lower rated PSV.

what is meant by 2+1 - what setting is each? You say first and second, but seem to have three PSVs?

Some numbers here might help.

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

 

[Sawsan311]

Thanks LittleInch for your reply,

the 3 PSVs are too staggered operating and one standby PSV. for Design pressure =100 barg, first PSV would be set at 100 barg, second PSV at not more than 105 barg. Despite being staggered I think it is safer to consider both PSVs popping at their full lift rated capacity as stated in API 521 4.9.4.3 since we have uncertainties on the set pressure tolerances as +/- 3% as per ASME SEC VIII UG-134 and UG-126...

 

[LittleInch]

I agree but how is this third (standby) PSV set?

Is it isolated or something?


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

 

[MJCronin]

Consider reposting in the "Safety Relief Valve" forum ....

MJCronin
Sr. Process Engineer

 

[Sawsan311]

the Standby PSV is set at a set point equal to the main first PSV and is normally isolated during normal operation. a key interlock is required as per API 520 PII to ensure chanegover from the operating PSV which is to be taken out for maintenance and this spare PSV which needs to be put online.

 

[georgeverghese]

Presume you are aware that the 2nd PSV with the higher setting must reach full relief capacity at 110% of process design pressure, ie its permissible overpressure is only 5% as compared to 10% for the PSV with lower setting.

 

[don1980]

Presume you are aware that the 2nd PSV with the higher setting must reach full relief capacity at 110% of process design pressure, i.e. its permissible overpressure is only 5% as compared to 10% for the PSV with lower setting.

This is correct if you're designing according to PED rules, but it's not if you're designing according to ASME Sec VIII. Sec VIII allows a 10% pressure rise above the setting of the second relief device (the one set at 1.05*MAWP). Sec VIII does this by allowing an accumulation of 16% when multiple relief devices are in service.

The PED's accumulation restriction of 10%, which applies regardless of whether you have one PRD or more than one, effectively negates any incentive to set a second PSV at 1.05MAWP. That's because one is left with the task of calculating the relieving capacity of the partially open second PSV. The vendors' flow coefficients are based on a fully open (full lift) PSV which occurs at 1.1*Set Pressure, rather than 1.05*Set Pressure. I think this detail must have been overlooked by the authors of PED.

 

[Sawsan311]

Thanks MJCronin, Don 1980 and georgeverghese, I think georgeverghese was referring to the second PSV which is set at not higher than 105* Vessel design pressure to have its overpressure estimated roughly as 10% while the first low set pressure PSV is set at the vessel's design pressure and having an overpressure as 16%. In both cases the accumalation of 16% is maintained as per ASME SEC VIII. Refer to API 520 PI section 5.4.2.4 table 3 where details are shown for how to estimate the overpressure values for each of the PSVs such that the upstream relieving pressure for both PSVs is the same and corresponds to 116%.

Mr.Don 1980 , for the PED standard you are referring to.. the set pressure tolerance between the second PSV 1.05 * MAWP and the 10% accumulation restriction makes staggering of the PSVs to avoid chattering or mitigate a large relieving capacity quite challenging? In general perspectives, for the design of multiple PSVs we usually consider both PSVs identical in size and operate initially at the full lift where the bore diameter exceeds 1/4 of the nozzle diameter and hence both would discharge their rated capacity. How can someone calculate the capacity of a partially open PSV unless it is equipped with a mean of restricted lift. Does PED state that the second PSV is set at 110% of the design pressure?? I think the resulting PSV sizing following this rule would be comparatively give a larger PSV than ASME code.. do you agree?

 

[don1980]

For non-fire cases, PED limits the accumulation to 10%, even in cases where multiple PRVs are used, and one is set at 1.05(MAWP). That means the PRV set at 1.05(MAWP) never reaches it's rated capacity - it doesn't achieve full lift. How does one calculate the flow capacity of a partially open PRV? You'd need the vendor to provide a flow coefficient for this partial-lift condition. The published coefficients can't be used because they are determined based on testing at the full-lift condition (10% overpressure). Possibly you can find a manufacturer that has done testing to determine the coefficient at a partial lift condition, but I'm not aware of any manufacturers that have done that. So, if you're designing based on a harmonized EU standard, the option to set the second PRV at 1.05(MAWP) is technically allowed but it's infeasible to actually apply.

 

[Sawsan311]

Thank you Don1980 for your reply..

I have a small query.. when you stated that the PSV will be partially lifted.. was the sizing of the PSV already catered for the fact that the second PSV which is set at 105% of MAWP would be provided with the sufficiently high area to enable its full lift. I understand that the full lift as described by API 520 is when the bore diameter exceeds one quarter of the nozzle diameter, but does increasing the area help.. at the same time I am doubting it will mitigate this since in both of the first and second PSV cases, the relieving pressure would not exceed 110% of the PSV set point,

thanks

 

[don1980]

Surface area isn't relevant in my explanation. I'm just talking about the relationship between overpressure and valve lift, regardless of the area of the disc or nozzle. Pop-acting PRVs require overpressure in order to achieve full lift. When PRVs pop open at set pressure, they don't pop fully open. PRVs pop open to only about 40-50% of their full-lift position. Additional force is needed to lift the stem further, and that comes from the increase in pressure at the valve inlet. As the inlet-side pressure rises above set pressure ("overpressure"), the valve stem rises further. Once the overpressure has risen to 10% the valve has reached its "full lift" position.

Certification testing, to determine the flow coefficient, is done at 110% of set pressure, the point at which the PRV has attained full-lift. When a PRV in a PED jurisdiction is set at 1.05MAWP, that PRV is only allowed 5% overpressure, which is less than the amount necessary to get the valve to full lift. So you're not allowed to take credit for the "rated capacity" of that PRV, because rated capacity is the capacity at 10% overpressure (full lift position).