7 replies to this topic

[ne2kal]

I'm trying to make sure that the relief valves on a glycol storage tank are sized properly. There are 2 valves on the tank, one set at 25 psig relieving to flare, and the other at 7" water column ( relieving to atmosphere).

Using hysys, I modelled the flowrate of each individual source feeding into the tank to calculate the critical flow. I used gas blow by as the worst case scenario, and used an upstream source with the highest pressure & largest line size to model the flow rate using hysys (utilizing the adjust variable in the simulation).

Using API 520, I calculated the capacity of the relief valve for the 25 psig. I used a relieving pressure of 25 + (25*0.1) + 14.7 = 42.2 psia.

How do I take into consideration the vent valve in the calculation?

Also is my thought process ok?


I would appreciate your input.

[Art Montemayor]

ne2kal:

Your query doesn't make sense. Why use 2 relief devices on the same tank? The first device - 7" water column - will be the only one that operates. The other - 25 psig - WILL NEVER OPERATE.

There is no simulation program required (or any "modelling") in order to define the maximum, credible, & possible relief flow rate. This is a blow-through case and is fixed by the upstream process.

Please define the details of your glycol tank. What is the Maximum Allowable Working Pressure (MAWP)? This value is what is required to set the relief device in order to protect the tank.

Please explain how you can justify 2 relief devices, each set at totally opposite pressures. How do you come up with te following: 25 + (25*0.1) + 14.7 = 42.2 psia. The arithmetic is flawed.

[JoeWong]

You low pressure relief device implied that your tank is low pressure atmospheric storage tank. Have you refer to API 2000 ?

Why not submit a sketch or drawing to check why 2 devices are provided. Please all the details.

[ne2kal]

Gentlemen,
I'm attaching a drawing along with this reply.


I do not know the flow rate feeding into the tank. I used hysys modelling to figure out how much gas from upstream can get blown into the storage tank. For that I assumed a gas composition, used the operating pressure of the upstream source 642 psia @ 88 F, and simulated the scenario to get the critical flow that is attained by reaching the relieving pressure of 42.2 psia. [I had the piping iso to create the model] . Since 25 psig is the set pressure, 10% for overpressure, and 14.7 for atms conditon = 42.2 psia.


Joe,
The storage tank is not under atmospheric condition. The MAWP = 25 psig for the glycol storage tank.

Attached Files

•  Glycol_Storage_Tank_Drawing.doc   226.5KB   123 downloads

[CMA010]

Gentlemen,
I'm attaching a drawing along with this reply.


I do not know the flow rate feeding into the tank. I used hysys modelling to figure out how much gas from upstream can get blown into the storage tank. For that I assumed a gas composition, used the operating pressure of the upstream source 642 psia @ 88 F, and simulated the scenario to get the critical flow that is attained by reaching the relieving pressure of 42.2 psia. [I had the piping iso to create the model] . Since 25 psig is the set pressure, 10% for overpressure, and 14.7 for atms conditon = 42.2 psia.


Joe,
The storage tank is not under atmospheric condition. The MAWP = 25 psig for the glycol storage tank.

The P162 relief valve seems to be for vacuum protection rather than overpressure?

[JoeWong]

Please clarify doubt raise by CMA010. Is this vacuum breaker or pressure-vacuum relief valve (PVRV) ?

For a gas blowby, critical flow may be attained with a limiting device. What is the limiting device in you piping that you have considered ?

How you model this limiting device in HYSYS ?

[ne2kal]

I just confirmed that it is for vacuum protection.


The PSV protected against overpressure is P161 SV set at 25 psig.


I don't understand what you mean by limiting device and modelling it on Hysys.

I modelled the gas blow by scenario using a given gas composition, an upstream source pressure of 642 psig @ 88 F, and downstream relieving pressure of 42.2 psia. I used a pipe segment from Hysys to input all the piping isometrics.

Using the adjust variable, I let hysys adjust the flowrates to get to the relieving pressure of 42.2 psia.

[JoeWong]

I don't understand what you mean by limiting device and modelling it on Hysys.

Limiting device is the the device within the line or loop that limiting/bottlenecking the flow during gas blowby. Other fittings/line/device will allow higher flow. For example, liquid line with level control valve (LCV) feeding to another separator. When the LCV FULL open with maximum CV, normally it will limiting the flow. Another example, globe valve (or reduced ball valve) in the drain line would possibly the limiting device. Past experiences shown that the pipe/line flow is excessive and hardly the limiting device.

I modelled the gas blow by scenario using a given gas composition, an upstream source pressure of 642 psig @ 88 F, and downstream relieving pressure of 42.2 psia. I used a pipe segment from Hysys to input all the piping isometrics.

Using the adjust variable, I let hysys adjust the flowrates to get to the relieving pressure of 42.2 psia.

The procedure is right (if you use HYSYS). However, some points you have to look at :

i) you have defined the pipe as your limiting device. As mentioned earlier, the flow will be excessive. Again, the limiting device may be the control valve, globe valve, etc.

ii) review the applicability of correlation that you used in the pipe segment compare to the flow condition, phase, etc. The condition likely to see two phase flow.

iii) check along the line if you experience sonic / choke / critical flow prior to reach 42.2 psia. The flow will not further increase if the choke condition is achieved with further pressure drop at the downstream pressure.

iv) From 642 psig to 42.2, the pressure drop is excessive. Density of fluid may change drastically. You may have employ more pipe sections.

v) Back flow gas blowby is event that many engineer missed out. Please ensure this is analysed.