Tuesday, July 04, 2006

Using pressure control valve as blowdown valve

Q) I need some opinion on the blowdown system for liquid petroleum gas storage. Some designer do not used any BD mechanisme except the PSV. However if I used PCV as BDV for blowdown any restriction from API 521?. or other codes?. Please

A1) The PSV is not my choice for the blowdown valve. In low pressure classes such as 150 or 300 consider a shutdown valve with an orifice plate to control the release rate. In class 600 or higher consider a shutdown valve and choke to take the drop. I normally use an API 6D ball valve for the shutdown valve.
I think that API requirements for the blowdown duration exist in the API 14 series offshore standards. However I did not find it in a quick word search. Perhaps they use depressure or something else in the standards.

A2) I'm no engineer but 30 years experience. PSV's are PSV's. They are not intented to be used for blow down valves. PSV valves are normally very expensive and have an almost zero bubble shut off. So, if used for more than a PSV, and lets say it leaks for any reason. If there is no other PSV that will relieve the operation the process will have to be shut down until the leading PSV is repaired. In addition there is no way to do a controlled blow down (restricted flow for whatever reason) with a PSV. PSV's are designed to open fully when reaching their set pressure.

A3) Modulating pilot operated PSVs do not go fully open when the set pressure is reached. Rather, the pilot adjusts the pressure on top of the dome to maintain, essentially, the inlet pressure at set pressure, opening and closing the disk as needed. In effect, they act akin to a control valve.
Anderson Greenwood has a feature that will allow you to dump the pressure on top of the dome remotely causing the PSV to go fully open which would allow the PSV to function effectively as a blowdown valve (they call it a 'remote unloader'). However, blow down valves are usually quite a bit smaller than PSVs in my experience as they aren't typically sized for the same flow rates. A PSV could be sized for a large vapor flow rate as a result of a fire or gas blowby while a depressurizing valve is typically sized for the gas inventory in the unit that needs to be vented as a result of an ESD (emergency shut-down). Common requirements that I've seen for a blowdown valve is to depressure a unit to 100 psig (or 1/2 of design pressure, whichever is lowest) in 15 minutes.
Remember if you use the PSV as a depessurizing valve to consider low temperatures in the system if you have any liquids that will autorefrigerate. That comment also applies to a depressuring valve.

A4) I don't see how you'd use a PCV as a blowdown valve. Acting as a backpressure device, the PCV would be designed to maintain pressure in the system. It would maintain the normal pressure and any higher pressure that were still below the point where you'd want to move into isolate and blowdown mode. If this is true, the setpoint would be up high - at the point where you're running into trouble. When the PCV starts to open and bleed off pressure, it will soon close. It will not lead to a low pressure condition, which is what I assume you want to achieve. Perhaps you'd better tell us by what you mean by blowdown.

HYSYS - Depressurizing

Is depressurizing of a vessel designed for fire case only?
Is it correct to think of fire when designing the depressurizing of a system?

As per API RP 521, 3 factors need to be considered during depressurizing.

1) temp increases due to heat input to the vessel causing liquid to vaporize.
2) density change of the gas in the vessel due to the decreased of pressure during depressurizing.
3) liquid flash due to the decreased of the pressure in the vessel.

However, I can find only the heat input to the wetted surface (liquid) in HYSYS -depressurizing- under 'heat flux' under 'API521' (by default-"adiabatic". How about the heat input to the unwetted surface (how to specify in HYSYS)?Notice that there is another option under 'heat flux', which is 'fire'it shows the equation as below (anyone know about this equation? from which reference?)

Q = C1 + C2*Time + C3*(C4-Vessel Temp)+ C5*Liqvol(time=t)/liqvol(time=0)

Yes, you'll need to take fire heat input into account when doing a depressurizing case. Just think about when you would be emergency depressurizing - in a fire event, right?
The heat input to the wetted surface is much higher than the unwetted surface - see API RP521.
You can account for it, and do other customizations, by using the general equation (heat flux - fire).
You can specify C1 thru C5 to come up with your own heat flux as a function of Time, Vessel temperature, and Liquid volume.
If you like to compute the (initial) heat flux yourself, for example, set C1 to your value, and set C2 - C5 to 0 (zero).
We have done this many times.
Hysys will then do multiple flashes to determine the temperature, pressure and compositional variations over time (as suggested by your items 1/2/3)For a good reference, read the material on support.aspentech.com
There is also a training module on dynamic depressuring there (as part of the Advanced Modeling course).
Also, I recommend that you check the Known Issues section for depressuring.

PH function in HYSYS

I use HYSYS to simulate my TEG dehydration project. This is a new design plant. My customer wants to know how much chemical (for PH control) to be initial fill to the system. As such, I have to estimate the PH value of the solution in order to determine the amount to be added.I tried to add "PH" function under "property" in HYSYS. However, it showed " cannot be added". Someone told me PH function can only be used for certain fluid package. if true, what are the available fluid package? are the fluid package suitable for TEG dehydration system?FYII am using Peng-Robinson as the fluid package for the system.

See http://www.olisystems.com/Application-fecl3.htm.
Explore that website for further info.

You cannot really simulate the pH your glycol will get to as it is effected by all the following1) contamination from non synthetic lube oils and chemicals from upstream2) oxidation3) CO2 in solution 4) TEG degradation5) glycol losses and replacementWhat I can tell you is on all the glycol systems I support the average injection of pH correctors is very small,in most cases zero and probably 12 litres a year would be an overstatement

With respect to the original question: pH is available in HYSYS only when the OLI-Electrolyte Property Package is used. No other package allows for the ionic dissociations needed for these calcs. Unfortunately your TEG is probably so concentrated (i.e. not aqueous) that the electrolyte package will be useless.best wishes,sshep