12 replies to this topic

[shahidulislam48]

Hi,

 

At first, I would like to apologize for my long post.

 

I would like to seek some suggestions about a system that I am having trouble to handle with my calculations. In this regard I have attached the drawing of the system.

 

For the sizing of PSV-100, I have found the required relieving load to be 19000 Kg/hr (Fire Case).

 

The purpose of EX-100 (Emergency condenser), is to condense the unexpected vapor generated by overheating of the vessel during our operation. It is important to mention that the vessel is subjected to heating and cooling by internal and external coils for maintaining desired temperature.

 

In the case of finding required relieving load for PSV-200, I have considered V-2, V-6 and V-7 will remain closed while V-1 will remain open. If these things are assumed, then required relieving load calculation leads to the consideration of feature like that I have done for PSV-100 and results in 19000 Kg/hr (Fire Case).

 

 Vessel protection.pdf   94.64KB   56 downloads

The main problem is that EX-100 has been designed for 46600 Kg/hr (vapor). It indicates that vapor will be generated in the vessel at the maximum rate of 46600 Kg/hr. Which is 2.5 times larger than the vapor generated during fire case (19000 Kg/hr). This flow rate could also be the flow rate that is faced by PSV-100 and 200. Even if I take whole vessel’s wetted surface, it doesn’t yield that amount of relief.

 

In that case, would I stick to my relief load calculation for fire case or I should consider 46600 Kg/hr for required relief load for both PSVs? Of course later will result in so much bigger PSV.

 

Any suggestion is highly appreciated.

 

Many Thanks.

Shahidul

[flarenuf]

hello

ok  let me ask a few questions here

 

1) in the event of cooling water failure i guess the vessel will overheat ( i am assuming the cooling coil you mention uses water)
this is a scenario in itself and you should determine the load for this  relief at accumulation pressure
 is this  the figure of 46600kg/h is from a runaway reaction when the vessel cooling coil and overhead condenser fails ?
 

2) for the fire case you are assuming that the cooling water valves are closed as well as V2 being closed (why), this is double jeopardy and not a valid scenario
    just because there is a fire the CW can still flow
 

3) if there was a fire around the vessel then you say you the load is 19000 kg/h, this will be relieved through PSV 100 and 200 , the flows being determined by the set point and size of the valves

4) I assume EX 100 is above the vessel , which is higher than 8m  so the wetted are of EX100 doesn't play a part in vessel fire area calc
 

5) so why have you a PSV on EX100? i dont have the full details but i assume its a full condensing shell with water in the tubes
    the condensate free drains back into the vessel. EX100 is above 8m in the air and presumably is therefore above any fire sustainable surface.
    so is not EX100 protected by PSV200 ?

 

[shan]

46600 kg/hr is the governing case to size the PSV-100 when V-1 fails.

[shahidulislam48]

hello

ok  let me ask a few questions here

 

1) in the event of cooling water failure i guess the vessel will overheat ( i am assuming the cooling coil you mention uses water)
this is a scenario in itself and you should determine the load for this  relief at accumulation pressure
 is this  the figure of 46600kg/h is from a runaway reaction when the vessel cooling coil and overhead condenser fails ?
 

2) for the fire case you are assuming that the cooling water valves are closed as well as V2 being closed (why), this is double jeopardy and not a valid scenario
    just because there is a fire the CW can still flow
 

3) if there was a fire around the vessel then you say you the load is 19000 kg/h, this will be relieved through PSV 100 and 200 , the flows being determined by the set point and size of the valves

4) I assume EX 100 is above the vessel , which is higher than 8m  so the wetted are of EX100 doesn't play a part in vessel fire area calc
 

5) so why have you a PSV on EX100? i dont have the full details but i assume its a full condensing shell with water in the tubes
    the condensate free drains back into the vessel. EX100 is above 8m in the air and presumably is therefore above any fire sustainable surface.
    so is not EX100 protected by PSV200 ?

 

Hi,

 

I would like to answer your questions by corresponding points.

 

1)  Overhead condenser does not fail. yes, it is the possible cause of 46600kg/h due to vessel coil failing and overheating. will that lead to 46600kg/h relief load as I am not getting so much by using conventional fire case relief load equations?

 

2) Yes, I can go with your second point.

 

3) I want to clarify this point by mentioning that I have calculated relief load for PSV-100(19000 kg/hr) and by considering the scenario I am assuming that there will also be the same relief load for PSV-200. I do not have size of the valve. Just calculated the relief load by conventional fire case equation.

 

4) Yes, of course EX-100 is above the vessel. Since I have to calculate the relief load for Fire Case for PSV-200 hence I need wetted area. But from the system it might be possible for both PSVs to have same relief load. For the sake of calculation if I assume that shell side of EX-100 is filled with liquid then relief load for PSV-200 will still be too much lower than the vapor capacity of exchanger.

 

5)Yes, you are right again. PSV-200 is protecting  EX-100. I have to protect the exchanger from any undesirable situation. And it has been considered for Fire Case. Usually we use PSV at tube and shell side of exchangers. There is also a PSV on cooling water side.

 

Many thanks.

[J_Leo]

What about V-1 and V-2 both closed, for example during maintenance, and there is a fire? I think this is a valid scenario. In this case ,you relief load would be 19,000 kg/hr.

[mirandomka]

Hello,

There are couple of things here. First you don't need to design both PSVs for 46600 kg/hr. Only the PSV on the vessel would need to. As I assume it is also "blocked outlet" situation for that PSV to be designed for 46600kg/hr.

The reason that the PSV on the cooler does not need to be designed for 46600kg/hr is because the setP of the psv is same as the psv on the vessel and your source of pressure is the vessel, so you won't overpressure your cooler shell for losing cooling water or upstream additional vapour generation as it would be accounted by the psv on the vessel.

[mirandomka]

Secondly, fire case.

In my opinion, fire case is not applicable for both PSVs as both the vessel and exchanger cooler elevations would be higher than API fire height(7.6m).

You won't have wetted surface area for calculation.

[mirandomka]

Thirdly, cooling water side psv-200 is not related to 46600kg/hour of vapour generation at all as I later realized it's on the cooling water side not the process side.

[mirandomka]

Lastly, internal coil failure case relief load may not be 46600kg/hr it's minor but req relief load will be the steam leaking rate to the vessel.

Reason is your cooler will be assumed working properly condensing the 46600kg/he of vapour, other wise it is double jeopardy


BTW, I really seriously recommend you to go with some experienced EPC for this task. PSV sizing needs system design experience otherwise you can be killed/kill someone.

Edited by mirandomka, 10 February 2017 - 12:11 AM.

[mirandomka]

Sorry I missed one important factor here. Please check if your steam coil/heat source max supply temp is above your boiling point of the fluid at relieving pressure or not. If not, you don't even have a cooling failure/blocked outlet case here.

However, please also check the upstream pressure to the vessel. If it is higher than vessel psv set p, your blocked outlet case is still valid and required relief load would be vapour entering the vessel.

[shahidulislam48]

Hi,

 

Thanks mirandomka for your valuable time and concern in this matter. I would also like to thanks other members.

 

At first, I assume that you are agreeing with the required relief load to be 46600 kg/hr rather than 19000 Kg/hr for PSV-100 (if not please correct me).

 

According to your first reply, if I can only design PSV-100 for 46600 kg/hr then what could be the relieving scenario I should consider for PSV-200? As my perception I can come out with the idea that the worst case PSV-200 would face will have to relieve 46600 kg/hr.

 

I would like to mention that PSV-200 is on the shell side which is handling process fluid. (Might be according to your 3^rd reply)

 

According to your second reply, I have been recommended to size both PSVs for fire case. Both Ex-100 and vessel are on different platforms. We could assume that flammable mixtures could spill out on the platforms and will lead to firing. That leads to the fire case consideration.

 

According to your 4th reply, to avoid double jeopardy we could assume the all the four valves of EX-100 are closed when there is no operation in the process is going on (as it is a batch process). Some significant amount of fluid condensate might be contained in the shell side of the EX-100 which could cause over pressure during firing case. Still for maximum worst condition we could assume that PSV-200 should handle amount that is equivalent to the exchangers maximum process fluid capacity.

 

At last, heat source max supply temp is above the boiling point of the fluid at relieving pressure and upstream pressure to the vessel is lower than vessel psv set p.

 

Regards.

 

Shahidul

[mirandomka]

For sure both PSVs has to be designed for fire base on your comment then.

But I don't think they both have to be the same size.

PSV100 should be bigger than PSV200.

The double jeopardy I'm talking is not fire. I'm talking about the abnormal heat case. You don't have to assume the cooler is down and abnormal heating at the same time.

If no other scenario leads to larger PSV, PSV100 to be designed for 46600kg/h. PSV200 to be designed for fire case and consider wetted surface area equal to exchanger shell area+connected piping up to isolation valves.

[shahidulislam48]

Thanks mirandomka. That was helpful.