3 replies to this topic

[B-2 Spirit]

Dear All,

 

I am currently working on a gas plant as process design engineer.

 

(1) The Gas Compressor is two stage centrifugal compressor. I have designed the suction side for settle out conditions. The discharge sides have been designed for settle out pressure + 1.3 x normal differential pressure. that is the discharge and all the downstream equipment have been designed for Compressor shut-in conditions.  Normal compressor dicharge pressure is 24 bara and calculated compressor shut-in condition is 37.5 barg (this is yet to be confirmed by compressor vendor)

 

I believe that in this case I don't need to consider "blocked outlet case" for compressor discharge PSVs, so fire case will be the only case for PSV sizing. Is my undertstanding correct?

 

(2) The Offgas compressor is a screw compressor. In this case the compressor discharge PSV has been designed for blocked outlet. I am assuming that 1.3 times the normal flow rate is a good assumtion for the blocked discharge case sizing. Is this assumption valid?

 

thanks and regards,

B-2

 

 

 

[Justin P]

B-2,

 

I was just about to pose a similar question for a similar, existing 2-stage centrifugal compression system.

 

On point (1):  

1. I am assuming that there is an interstage cooler and interstage knock-out drum between stages. Is this correct? If so, then yes, I would consider the fire case on the KO drum.
2. Are there any intervening valves between the first and second stages?

 

On point (2): Multiplying the normal flow rate by 1.3 may or may not be a good assumption (depends on how far along the design is). If possible, use the performance curve of the selected screw compressor and flow rate at the maximum possible speed.

 

To reiterate your question(s) regarding blocked outet (these are questions to the audience):

1. Assuming zero credit for anti-surge (aka minimum-flow, aka spill-back) valves and zero credit for upstream blowdown valves (those that would otherwise dump the feed to a flare), what happens if the discharge of the second stage is blocked in?
2. What exactly will happen during the transient state of a compressor being blocked in on its discharge? As the compressor flow stagnates and surge begins, what happens to the interstage pressure if the machine continues spinning 
3. For a fixed speed compressor, does raising the suction pressure of the first stage suction increase the discharge pressure, or does it decrease the compression ratio?

regards,

Justin

[Robert Montoya]

B-2 Spirit; 

 

PR valves are required for any compressor where the maximum pressure which can be generated during surge or restricted discharge conditions exceeds the design pressure of the discharge piping, downstream equipment compressor seals, or compressor casing.

For centrifugal compressors, If the design pressure is higher than the discharge pressure of surge point at maximum speed, no relieving is required. If relieving is to be considered, the relieving rate should be evaluated considering the design pressure, shut-off pressure, suction pressure, surge control line, maximum speed for turbine drive and capacity of anti-surge recycle line.

 

It is usually economical to set the design pressure lower than the maximum possible pressure that the compressor can develop, and to provide appropriate PR valve protection on the discharge. In some cases (e.g., where the flow through a PR valve would be the release which establishes the size of a closed disposal system), it may be advantageous to set the design pressure of the compressor casing and downstream equipment equal to the maximum pressure that can be generated at the surge point, assuming the most severe combination of speed, molecular weight, suction pressure and temperature conditions that can occur as the result of a design contingency.

 

For positive displacement compressors, discharge PR valves are nearly always required. Reliance on stalling of a reciprocating compressor is generally not economically attractive, since driver stalling pressures are usually quite high in comparison to operating pressure.

 

[Bobby Strain]

It's difficult to follow much of this discussion when only the compressor is discussed. One must consider the whole system to which the machine is attached when determining relieving requirements. When in doubt, look at your company or client design guides. But maybe they don't exist. So get one of your experienced colleagues to review the system with you. And you should have a P&ID for any meaningful discussion. Good luck.

 

Bobby