Centrifugal compressor start-up challenges

[Sawsan311]

Dear All,

I would like your views on the start-up conditions of a compressor which operates under discharge pressure master controller (with override under low suction pressure).

Starting the compressor with its anti-surge valve initially open is mandated while bringing the speed to the minimum speed prior to the RUN state and loading the compressor via the performance controller automatically. However, I have seen that some studies consider the discharge valve initially opened initially, will this endanger the compressor to stonewall specially when speed increases.

Other start-up scenario, consider both suction and discharge valves closed in order to prevent the instability caused by reverse flow through the common suction manifold specially in the case of parallel trains operation and in absence of check valves at the suction. Is this start-up methodology practically followed?

Also, a general question,will stonewall operation be more prominent at high speeds or low speeds fora VSD machine?

Thanks

 

[rotw]

However, I have seen that some studies consider the discharge valve initially opened initially, will this endanger the compressor to stonewall specially when speed increases.

There still is the check valve at compressor discharge that compressor can start against until the cracking pressure of check valve is reached.

 

[Sawsan311]

Thank you very much Mr.Rotw for your reply,

I would like to explore in compressor's stonewall operation knowing that it can lead to motor overheating with ultimately tripping the compressor due to high vibration loads.

Compressor endangered by stonewall would be caused by abnormal failure such as anti-surge valve or hot gas bypass valve failing inadvertently in open position causing an increase in suction pressure and even with the increase in speed by performance controller, compressor can choke at high speed limit. This is mainly driven by the changes in the system curve. Can you confirm my understanding?

Additionally, do you think the tendency to stonewall can't be generalized to be more prone at high speed versus low speed because it depends on the shape of the system curve, performance curve as well as slope of the choke line?

Appreciate your views.

Regards,

 

[rotw]

Compressor endangered by stonewall would be caused by abnormal failure such as anti-surge valve or hot gas bypass valve failing inadvertently in open position causing an increase in suction pressure and even with the increase in speed by performance controller, compressor can choke at high speed limit. This is mainly driven by the changes in the system curve. Can you confirm my understanding?

In general, anti-surge valve fully open is not likely to cause choke flow for the compressor. The valve is normally sized to cover the surge limit line flow (minimum). Plus a margin is added to accomodate for dynamic response effect. This margin normally does not extend to the choke region. In some case it can happen, but these situations are marginal and it should be flagged at design stage. On top of that, compressor shall be mechanically sized to cope with choke (not specifically because of anti-surge valve consideration but due to consumer demand / throughput increase for instance).


Additionally, do you think the tendency to stonewall can't be generalized to be more prone at high speed versus low speed because it depends on the shape of the system curve, performance curve as well as slope of the choke line?

As far as thrust bearings are concerned in respect to verfications for stonewall operation, the consideration that prevails is that operating points (wherever these are located) must be verified as well as map limits so to make sure there is always sufficient margin between thrust bearing capability and axial load - but there is no reason to oversize a thrust bearing neither. An oversized thrust bearing will consume more lube oil, generate more power losses, and create an overhung weight/load not helping rotordynamics.

 

[Sawsan311]

Dear Mr.Rotw,

Thank you for your reply,

In my second question I was mainly asking about the tendency to approach the choke line is expected to be more at high operating compressor speed or low speeds?

thanks

 

[rotw]

I read your question well. But I just skipped the use of terminology "tendency to approach" choke. Too vague.