Reciprocating compressor suction blockage

[AMXX]

Hi;

I am investigating a reciprocating compressor suction blockage case. In case of the suction scrubber demister blockage about 1 minutes takes till the compressor driver stops. So during this one minute, the compressor cannot have any feed gas as the recycle valve is recycling the gas to the inlet of the suction scrubber. It is clear the demister is damaged because of the vacuum at the compressor suction side and maybe compressor is damaged by pieces of demister. If we ignore the demister damage:
1) How long a reciprocating compressor can run without feed gas?
2) What types of damage is expecting for the compressor?

thanks

 

[pmover]

why in the heck operate a compressor w/out feed gas? that is the purpose of a low pressure shutdown trip.

1 - too many variables to consider, but the larger the bore, stroke, and # of cylinders result in less time.
2 - no flow may result in heat damage to cylinder/piston and other components. think about the heat of compression (PV diagram) and what happens to zero flow.

 

[AMXX]

The one minute is the time from activation of trip function by LL pressure transmitter (<20sec) till the driver stops as per the trip sequence (40sec) . In normal case the gas is circulated till compressor trip sequence can be completed however in this case as the demister is blocked there is no possibility to feed the gas to the compressor during trip sequence.

1) I need just a rough estimated time (based on operator / manufacturer experience) to figure-out how fast the trip sequence should be, to avoid the compressor damage. This gives an idea if it is possible to reduce the trip sequence time to protect the compressor. However if the compressor damage is happened in a few seconds then only have to avoid the demister blockage (e.g.: addition of a differential pressure trip) .

2) The case is for no feed gas condition so I believe the heat source mostly should be the friction by piston moving. I am not sure what can be considered during 1 minute trip period as heat of compression( suction pressure and gas rate are dropped) and friction.

 

[georgeverghese]

The process gas itself provides little cooling to the piston / cylinder walls. Most recips have water cooled cylinder head jackets to keep piston liners cool. The compressor should have a PSLL on suction to trip the compressor. Else the compressor will trip later on high discharge temp.

 

[AMXX]

Thanks Georgeverghese;

yes all these trips are available however activation of them takes times and trip sequence till drive stop step takes about 40 second. In normal shutdown sequence the compressor is received recycle gas but as I mentioned before in this case the compressor cannot receive it as demister is blocked so compressor is run about 1 minute in this situation if we ignore demister damage.

Anyhow this is an incredible scenario as demister cannot be blocked suddenly without notice so it can be preventable by a PDT around the demister. I just tried to find what can be ultimate consequence of this case that it seems it can be compressor damage by demister particles.

 

[georgeverghese]

Agreed, bits of the demister SS filaments may get extruded out through the top hold down grating of the demister and carry over into the compressor suction as a result of this high dp across the demister. Other machine integrity and process safety consequences may be
a) If you have vacuum on stage 1 suction in the compressor, there may be reverse flow of air through the piston packing into the process gas.
b) There may be lube oil also from the compressor packing also leaking out in to the process gas.

Why does it take 20+40sec for this compressor to stop on PSLL stage 1? The compressor should begin coasting down in less than 5seconds on detection.

 

[JJPellin]

Depending on the process conditions, this could also result in high compression ratio and rod overload. If this exceeds a rod load limit, the piston rod could fracture in a few seconds. Consult with the compressor manufacturer to determine rod load limits.

Johnny Pellin

 

[AMXX]

Dear georgeverghese;
Yes you are absolutely right about a & b . I did not think in this way thanks for your kind reply.

about the time : 20 second is the maximum time to reach from compressor normal discharge pressure to the trip pressure. Minimum time can be about 5 second.
40 second is the time after activation of trip sequence that the compressor motor is stopped. As per the compressor stop sequence it happened at stage S40 and operator told me that this means after 40 second the motor is stopped. This compressor doesn't have a separate emergency trip sequence. Normal trip and ESD trip takes the same time and I think 40 second is an acceptable time to stop the compressor.

 

[AMXX]

Dear Johnny;

Thanks for advice. I consider this one too.

 

[georgeverghese]

If you compressor suction drum demister has high dp frequently, then would suggest installing a filter coalescer upstream of the drum - wiremesh demisters dont work well with fouling gas, and on the other hand, the liquid entrainment from a vane pack type scrubber may be too much for a recip compressor. If the gas is clean, then it is unlikely you will see high dp across the stage 1 suction drum.

 

[AMXX]

No. There is no high demister DP at all. This is an extreme case which was defined during HAZOP meeting and I need to define its consequences.