15 replies to this topic

[Sherif Morsi]

Hi All,

 

I am currently sizing a PSV installed on the discharge line from a methanol pump. I considered two cases for sizing: Liquid relief and fire case.

 

The methanol pump is located around 125 meters away from the destination point (distance between to ESD valves).

 

In case of fire, should I account for the whole 125 meters or only the portion of the line directly exposed to the fire? Also, for the wetted area calculation, should I consider the line completely filled with liquid?

 

Regards,

Sherif

[Zauberberg]

Answer on both questions: yes. If you think about these two scenarios, you'll come to the same logical conclusion.

[fallah]

Sherif,

 

At first please upload a simple sketch of the system; then specify if the mentioned piping, that is exposed to fire, can be isolated. If so, please specify the weight of the liquid methanol would be trapped after isolation...

[Sherif Morsi]

Thanks for the replies, please see attached sketch

 

 DOC2013Nov13021018.pdf   27.28KB   345 downloads

[Sherif Morsi]

Zauberberg,

 

Is it mentioned some where in API 520 or 521?

[fallah]

Sherif,

 

As per the sketch appears the piping system that can be exposed to fire can be isolated between two SDVs along with pump trip. Then with such low pump capacity appears the piping size would be around 1" will lead to small weight of the trapped liquid. Therefore, in general, a TSV would be sufficient to release the liquid of overpressure consequence due to fire as an external heat source...

 

Then no need to wetted area calculation in order to consider the PSV for fire case...

[Sherif Morsi]

Ok,

 

The line is currently 2" Sch.40. The total volume inside the line is around 0.3 m3. So it's a small volume. Also, the pump is a positive displacement pump, so there is no way the liquid will flow across the pump.

 

I checked API 521 section "5.14.4 Piping" and only the thermal relief is considered for piping under certain conditions.

 

I guess I shouldn't consider the fire case for the line. Please confirm

[fallah]

 

 

Sherif,

 

It is confirmed...Just thermal relief would be adequate...

[Zauberberg]

Fallah, correct me if I am wrong but there are two valid cases to consider for the relief:

 

- Blocked discharge of the positive displacement pump

- Fire case, which is different than short-term thermal expansion of fluid inside the piping (liquid-full system)

 

Getting back to the original post, why the relief valve is deemed as not required?

 

Regards

[fallah]

Zauberberg,

 

If blocked discharge of the PD pump would be a credible scenario, an external PRV is needed in addition to pump's external PRV...

 

But due to very small trapped liquid in subject system, fire case cannot be a credible scenario and a TSV would do the job: Liquid relief due to thermal expansion...

 

Finally, if a PRV has already been instaled to handle the relief due to the blocked outlet case, there would be no need to a TSV to handle the liquid relief due to thermal expansion because the PRV can handle both liquid relief cases... 

 

Please let's know if it still isn't so clear for you...

[Sherif Morsi]

Thanks guys for the feedback.

 

Much appreciated

[ChemEng01]

The only case here is blocked outlet, using the rated flow of pump and thermal expansion.

 

Generally you don't apply fire case to piping or tubing because the pipe/tube is more likely to rupture before relieving conditions are reached (Unless you have a fire covering the whole length of your tubing (unlikely). It states this in the shell DEP and in API.

 

As fallah said thermal expansion would be another case, but definitely not the governing case. (RV sizing is not required for thermal relief)

[Zauberberg]

Fallah:

 

 

Thanks, I've got it. My understanding was that you recommend no relief valve to be installed at all, while there is one already at the pump discharge.

 

Where would you install a TSV? Upstream or downstream of the pump, or both?

[fallah]

Zauberberg,

 

Thanks for your attention...

 

Referring to the sketch of the OP, if pump would be tripped after line isolation, TSV can be located either upstream or downstream of the pump between two SDVs. But access for inspection and maintenance is an important point and also liquid static head due to elevation should be considered in specifying TSV set point...

[Sherif Morsi]

All,

 

Happy New Year!

 

Just an update, we have come to a conclusion that the PSV downstream the pump will be for blocked outlet case. The thing is we are using a diaphragm pump and we are limiting the design pressure of the system to 7 barg which is the design pressure of the vessel and tank connected to the system.

 

What do you think of that?

[fallah]

Sherif,

 

To modify my previous post:

 

Because of being a positive displacement pump between two SDVs, after fire occurence and subsequent pump trip, there would be no way the fluid flows across the pump. Then the pump will separate the line between two SVDs and the PRV at the pump downstream which has already been considered for blocked outlet case will handle the liquid relief due to thermal expansion of trapped liquid at pump's downstream. Obviously, another TSV should be considered at pump's upstream to handle the liquid relief due to thermal expansion of trapped liquid between the first SDV and the pump...