Inlet Line Sizing Upstream of Blowdown Orifice

[bth1]

I am currently sizing the blowdown orifices and associated inlet/outlet piping for an offshore oil separation process.

For sizing of the orifices I am using HYSYS's depressurisation utility.

For critical flow, HYSYS estimates the orifice size using the orifice equation stated in Perry's (Pg 10-15):

w = Cd.A.P1.(k.(M/(R*T1)).((2/(k+1))^((k+1)/(k-1))))^0.5

This equation is derived based on the assumption that the beta ratio (orifice diameter / upstream pipe diameter) is < 0.2. When the beta ratio is < 0.2, i.e. the orifice diameter is small in comparison to the upstream pipe diameter, frictional effects can be ignored.

For the inlet piping upstream of a blowdown orifice, I normally size based on a rho.v2 criteria and ensure that the velocity and pressure drop are sensible.

I am concerned that this is not sufficient though. For example, using the above criteria, I may end up with a 4" pipe for the maximum mass flowrate predicted. However, for the orifice size calculated by HYSYS, a 4" pipe may mean that my beta ratio is a lot higher than 0.2. This in turn means that frictional effects become a factor and in reality my orifice may not actually pass as much mass as HYSYS has predicted, thus the time to blowdown increases.

I would welcome any thoughts on the above and would also be interested on how you typically size the inlet lines upstream of blowdown orifices.

 

[djack77494]

My thoughts on this topic-
The beta ratio of 0.2 is a very rough guideline. I would think that oftentimes when beta>0.2 the inlet friction loss is still small. Likewise, under some circumstances (e.g. long suction line) the friction loss might be large. Until you get a good feeling for when frictional losses are/are not important, analyze the entire system.
Doug

 

[JLSeagull]

For AMSE Class 600 and higher, consider using fixed chokes for blowdown instead of orifice plates. Try to limit the blowdown valves and associated inlet line around NPS 2 or NPS 3. The full-port blowdown valve opens and with most drop taken via the choke.

I was slapped around a bit in another thread for this suggestion by one of the premier process engineers, perhaps Montemayor.