6 replies to this topic

[BJMAC]

Dear All,

I am a process engineer, work for an oil and gas company, at the engineering design sector, and I am new at PSV sizing. I need to size a PSV (define the criteria) that is located at the hot water return line of a waste heat recovery unit (WHRU) that is located together with a gas turbine, at a FPSO.

According to API 521 I should consider thermal expansion. On the other hand, WHRU supplier commented that code requirements for boiler safety valves normally considers zero water flow with full heat input to the heat exchanger. They also commented that ASME codes normally considers sizing of PSV´s for maximum flow that can be supplied to the heat exchanger.

In the system mentioned above, from my Company experience, we shall consider: hot water pumps stop, with WHRU dumper failure, and with WHRU dumper control failure (Gas Turbine won't stop, won't receive a signal to inform that the dumper failed), so full heat input to the heat exchanger.

My questions are:

1 - Which criteria should I use, thermal expansion or full heat input to the heat exchanger ?

2 - How to define the flow rate for the criteria full heat input to the heat exchanger ?


Thanks in advance for the help,

BJMAC

[Art Montemayor]

BJMAC:

I moved this thread to the Relief Devices Forum where I believe it belongs.

Everyone and every industry has its acronyms and pet names for machinery and equipment. What is a “dumper”?

What would be even better and more correct input of basic data is your P&ID – or at least a schematic drawing showing where, exactly, you propose to install a thermal expansion valve. It is very difficult and tiring trying to understand your description. A P&ID says it all and explains why a PSV or a TRV would be needed.

Be careful with API 521. It is a RECOMMENDED PRACTIC – not a mandated code or even a direct and specific recommendation to your problem. It only tries to cover the GENERAL cases – not your specific application. It carefully explains that you must have the ability to block-in a 100% liquid line or vessel in order to consider thermal expansion. With your supplied P&ID, this possibility should easily be apparent.

Can you please furnish the related P&ID?

[BJMAC]

Art Montemayor,

thanks for the notification. Dumper is the gas diverte valve. I already have the P&ID, a pdf file, but I didn't find how to upload/attach it.
Can you help me ?

Thanks,

[BJMAC]

Dear All,

please find attached the P&ID related to the query.

Rgds,

BJMAC

Attached Files

•  PSV WHRU.pdf   242.66KB   56 downloads

[Art Montemayor]

BJMAC:

Thank you very much for the P&ID. Now, this clears up a lot of the doubt, confusion, and questions on your query.

As I mentioned before: be careful in interpreting API 521. It is not a “Code”. It is meant to guide you with recommendations – but you must interpret the type of process you have at hand. Now that I see your P&ID, I note that you use the term “hot water” for the return line out of the Heat Recovery Unit (HTRU). This, to me, tells me you left out some very important basic data in your initial post: you are HEATING WATER not GENERATING STEAM. Normally, I would expect a steam generator as the HTRU. This is very important process basic data because it reveals that you don’t expect – nor design for – a CHANG E OF PHASE within the HTRU. If my observations are correct, then you have made a major process error in evaluating the application: you don’t have a thermal expansion case alone. You could easily have a credible steam generation case – and, therefore, a 2-phase PSV relief scenario within the HTRU coil.

Certainly, API 521 may recommend you to consider thermal expansion. But that is a general recommendation – never can it be considered specific. API 521 is powerless to be specific because it doesn’t know or have your specific P&ID.

Although you show hot water circulating through the coil in the P&ID, you mention that the “WHRU supplier commented that code requirements for boiler safety valves normally considers zero water flow with full heat input to the heat exchanger. They also commented that ASME codes normally considers sizing of PSV´s for maximum flow that can be supplied to the heat exchanger.” The WHRU supplier is correct – if you have a BOILER (a phase change heat exchanger). If you have a hot water heater, it may be another case altogether. How is the unit design classified? The important point here to bear in mind is that, regardless whether you are creating steam or just heating water, you may have the ability to create steam if your water pump fails and your turbine’s exhaust continues to input flue gas energy into the HTRU coil. If this is a possible and credible scenario, then you have a steam generator case on your hands and you have to design for a PSV to relieve the steam – not a thermal relief valve to relieve expanded liquid. A Thermal Relief Valve is incapable of handling a vapor flow safely – much less a 2-phase flow that can easily occur.

I strongly advise you to hold a mini-Hazop on this specific node and determine if you have a credible and possible scenario for blocking in the hot circulated water with the continuing input of turbine exhaust into the HTRU. If so, then you have, in my opinion, a PSV application (not a TRV). Additionally, why is the coil PSV shown at the OUTLET of the coil and not at the INLET (where it can more surely protect the coil)? It is considered standard practice to locate the PSV at the entrance of the coil – not at the outlet.

I anticipate and look forward to other Forum members joining in on this very important topic.

[fallah]

Dear Art

IMO, uploaded PID with lack of, at least, below information cannot so help to evaluate credible scenarios in order to find a reasonable basis for mentioned PSV sizing:

-Sizes/calsses of inlet/outlet lines
-Full cycle of inlet/outlet lines including the equipment such as pump, drum,...
-Operating/design pressure and temperature of the cycle
-Values of full heat input and maximum circulating flow
-Control philosophy and layers of protections of the system especially against the causes will lead to circulating flow stop as an subsequent effect...

Fallah

[BJMAC]

Dear Art and Fallah,

Please find below information requested:

We consider the following scenario can happen: Stop of the hot water water circulating pumps (emergency shutdown, keeping main electric power supply on), and a Double failure, failure of the dumper (dumper stiffened, rigid) and dumper signal failure (the dumper limit switch does not confirm that it is open, so the GT do not stop).

Regarding the PSV position, in general we locate the PSV at the outlet.

The WHRU is considered a water heater as per ASME Section VIII Div 1 & 2 Boilers and Pressure Vessel Code: Pressure Vessel.

Piping inlet/return 10”, class 300#;
design pressure (PSV set, the target of the query) 13 barg; water inlet/oulet: 125°C/95°C, design temp. 155°C;
Three pumps (3x50%, 850m³/h each), control valve recirculating from discharge to suction; pressurized expansion vessel (PSV set to 6 barg);
WHRU: 25,000 kW, circulating flow per unit (570m³/h);
In emergency shutdown (keeping main electric power supply on, only one of the four GTs, 4x33%), the pumps stop and the gases (gas diverter valve open) go to atmosphere. TSHH and a FSLL at hot water outlet open the gas diverter valve.

Rgds,

BJMAC