2 replies to this topic

[Marco78]

Dear all,

I have to install two pressure relief valves on the cold side of a condenser (1) and a heater (2).
Causes: exchanger blocked-in on the cold side with flow in the hot side.

1) CONDENSER: vapours temperature is equal to 120°C. The fluid of the cold side is water at 35 °C (design temperature). Considering the pressure design of the vessel (12 bar) I chosed a Pset for the valve equal to 7 bar (common in our plants)

2) HEATER: on the hot side we have saturated steam at 5 barg (159 °C more or less) and the cold side is water (normally at 65°C). As for the condenser the PSet is 7 bar (design pressure of the heater is 9 bar)

My questions are:
1) is it correct design the valve for hydraulic expansion?

2) is it true that, since the vapour pressure of the water at the maximum reachable temperatures in the two cases (120 and 159 °C) is less than the Pset of the two valves vaporization will not occur.
(API 521 3.14.3 states that vapour generation rate has to be consider only if the Pvap is higher than the relief design pressure)

3) What will be the relief temperature? I expect 120°C and 159°C and consequently flashing during relief, is it right?

Many thanks in advance
Marco

[paulhorth]

Marco,

Answers to your questions are:

(1) Yes. The relief valve will lift due to hydraulic expansion. The flow will be very small, so even with the smallest relief valve, the capacity of the valve will exceed the expansion and the valve will close again after releasing a few drops. This opening and closing will continue as long as the heat input continues (that is, till the water reaches the same temperature as the heat source). The valve will open first after only a small rise in temperature. I think (from memory) that the rate of pressure rise of a blocked-in volume of water is about one bar for every two deg C (approx).

(2) Yes. No vaporisation upstream of the relief valve.

(3) Yes, there will be flashing on the outlet side of the valve, but only after the contents has heated up to above 100 C, as described above.



Paul

[GS81Process]

Marco, My thoughts on your questions:

1) I would have to know more about your process to confirm that the governing sizing basis for the PSVs is the blocked in water case. Most importantly, are you certain that these are the governing cases? DId you perform an overpressure scenario analysis as outlined in API 521 and look at whether other cases are possible?


2) Your question can be answered by referring to the steam tables. The saturation temperature at 7 barg is around 170.5C (I'm not sure of the barometric pressure at the location you are referencing so I used 101.325 kPaa). So, yes since the vapour pressure is less than the set pressure then you will not have steam ENTERING the PSV. As noted by a previous poster, there will be flashing as it is depressurized, presumably to atmosphere?

You noted that you chose the set pressure as 7 bar because it is common in your plants. I think that you need to be able to provide better justification as to why you chose this set pressure. It is ok to choose a set pressure below the design pressure of the vessel you are intending to protect, but you have to understand why.

3) If the water side is blocked in, it will begin to relieve at temperatures not much above 35C and 65C, respectively. This is because it takes very little temperature rise for a blocked in fluid to generate pressure, as noted by a previous poster. The temperature will quickly rise to that of the heating side, i.e. 120C and 159C, respectively, as relief occurs.

Can you prevent the blocked in overpressure case on the water side by locking the valves open and using administrative controls?

Edited by GS81Process, 11 July 2011 - 02:12 PM.