Question on centrifugal compressor startup sequence 2

JustSomeRoark · Nov 27, 2022

Gents:

When starting up a centrifugal compressor, do you normally allow the discharge isolation valve to open before or after the discharge pressure of the compressor matches the discharge header pressure?

For instance, consider a single stage centrifugal compressor driven by a fixed speed syncronous electric motor. Capacity of the compressor is controlled by suction throttling. Lets assume that the motor is switched on with the inlet isolation valve fully open, the suction throttle valve at its minimum opening and the antisurge valve fully open. After starting up the motor the compressor will develop a pressure lower than the discharge header (i.e. the antisurge valve is fully open). It is advisable to open the discharge valve at this stage, assuming that the check valve at the discharge will hold? Is there any risk/concern associated with the check valve passing gas from the discharge header to the compressor discharge?

If such operation is not sound, when/how would you recommend/advise to open the discharge isolation valve?

Regards.

georgeverghese · Nov 27, 2022

Open the discharge SDV only when the developed compressor pressure is almost the same as discharge header pressure for obvious reasons.
You can set up a dP permissive to open this valve - compressor pressure is known at DCS , and so is discharge header pressure. Send this dP signal to the control system dedicated to the compressor. No need to set up a dedicated dP transmitter for this.

pmover · Nov 27, 2022

another possibility . . .
Assuming all pre-start interlocks are in safe state for startup.
for startup, both suction and discharge valves are open (confirmed by valve operator limit switches). the check valve in discharge pipe prevents any back flow. once the unit develops operational speed, gradually close anti-surge valve which will develop the head to open the check valve. use/open suction throttling to meet process demand.

LittleInch · Nov 28, 2022

The risk would be that the check valve doesn't hold and that gas passes backwards into the suction system, possibly over pressuring the upstream system.

This sort of thing is often looked at in HAZOPs (reverse flow) and can be quite problematic. If you are relying on this check valve then you often need two of different type of design. The axial plug type can be very effective, but not cheap.

Thus I would normally expect that the discharge pressure was built up to a point where it was higher than the discharge pressure to avoid this point.

This assumes of course that the inlet pressure is less than the discharge. Booster compressors along a pipeline have an issue whereby initially the compressor essentially free wheels until a small differential pressure grows as the flow rate increases.

Equally hard to start sometimes

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Question on centrifugal compressor startup sequence 2

JustSomeRoark

Chemical

Recommended for you

georgeverghese

Chemical

pmover

Mechanical

LittleInch

Petroleum

Similar threads

Part and Inventory Search

Sponsor