Low Pressure Flare line pressure drop calculations 1

[almostbroken]

I have a case where we have an 750 bbl 8oz tank that goes to flare through a 2oz weighted check valve. There is a water dump into the tank that is calculated at 700 cubic feet per hour. The tank is gas blanketed witn natural gas at 1 oz.

From the check valve the gas goes through about 1000 ft of 6inch line through a new detonation arrestor and then into a 4 inch flare. (don't ask)

When the water is entering the tank the thief hatch start to weep, and as the water is sour it is very noticable.

My calculations using HYSYS, and a variety of flow equations from GPSA and course notes all seem to indicate that we should be able to flow 10X as much before the back pressure would exceed the 8oz setpoint.

This leads me to wonder if there are better flow equations for low pressure gas flow?

Anyone have any ideas?


Almost Broken

 

[TD2K]

Have you done a pressure survey at the tank, along the flare line, across the detonation arresotors to see if the pressures are inline with your calculations? I'd be suspicious of the check valve pressure drop requirements. I'd also be suspicious of the detonation arrestor pressure drop but you did say they are new, were they sized based on these flows. I haven't much experience with them but the last job I had we put a pair in and they had a much higher dP than say a simple flare arrestor.

Taken any samples to see if the gas densities are in line with your simulations?

At these pressure you would treat this as an incompressible flow so I doubt the equations are the issue here.

 

[CJKruger]

almostbroken, different issue - Most companies will not allow you to vent a non-pressurized tank to a common flare. They allow only pressure vessels to vent/relief to a common flare.

 

[almostbroken]

We are going to do a p

http://www.eng-tips.com/faq.cfm?pid=124ressure

survey. I had hoped to find the issue without going to the plant which is 100 miles past nowhere.
The gas composition we believe to have a handle on as we know the gas going into the plant. I believe HYSYS does a reasonable job with predicting the gas breakout of the water dump. I have checked with Raoult's law and found good agreement.
The dump line has had meter added to it to confirm the water flow and it agrees with the calculated 20 m2/h (700 ft3/h).
Like you I suspect it could be the weighted check valve. I always use PRV's but the client wanted this cheap alternative.

Thanks for your post, it confirms my thoughts. It is great to get that.

Almost Broken

 

[SNORGY]

Almostbroken,

Personally, I have no confidence at all in the repeatability of the "set pressure" in a weighted check valve. I also have less and less confidence in each passing year with the set pressures for thief hatches. I think they "burp" all the time at any variety of tank mean pressures. Further, I am not sure what your tank level indication is, but - for example - gauge board seals can also be major odor sources. If this applies in your case you probably have NACE seals so this would help.

I don't have the API-12D/12F/650 Standards in front of me at the time of this posting, but when I have a tank vented to flare I try to go to 16 oz design if I can. It usually forces some construction extras (notably anchor chairs) in the tank, but it does reduce odors to some extent by allowing higher set pressures for devices in the tank vapor space.

Per your request, I won't ask about the 6" line into the 4" flare header.

Is the detonation arrestor required or can you establish continuous fuel gas purge far enough upstream in your flare system and get rid of the arrestor altogether? Information in your post suggests that you have a gas blanket on the tank and the tank is quite a ways upstream in the system.

Coming back to the weighted check valve for a moment, my biggest issue with them is that you usually end up getting them from a distributor who is far enough down the supply chain that you can never really be certain how well they are calibrated, designed, or even if they are new or reconditioned.

As for clients wanting to be cost effective...people get "knocked down" that way. But...clients are clients. I suppose if they want "cheap" then they are prepared to live with the consequences, which could range from persistent odors and regulatory fines to knocking a guy down as he's gauging the tank.

Regards,

SNORGY.