Oxygen in gas stream in natural gas compressor

[engr2GW]

Hello,
I have a chicken and egg situation with Gas lift well and gas lift compressor. Oxygen is being detected in a gas stream, this is not a one off or one month few spikes situation, it have been on and off for months. The well produces the gas, the compressor compresses some of the produced gas and sent back to the well for gas lift.

Is there any circumstance under which air (oxygen) can contaminate or go into the gas stream through the compressor?

Is a compressor known to have operating inadequacy that can introduce oxygen into the gas stream that it is compressing? If the well as the source of the oxygen is ruled out.

Thanks for your help.

As much as possible, do it right the first time...

 

[rotw]

- You did not indicate which type of compressor is this ? (centrifugal, recips, other ?).

In case this is a centrifugal, unless gas lift compressor suction pressure is low/negative I may be wrong but I can't see how ambiant air could be ingested into the compressor due failing element, seal or gasket or a leakage of some sort. As for dry gas seals, in typical tandem type arrangement with or without intermemdiate buffer, the primary seal gas is usually process gas itself or another source of clean gas - but certainly not air. Intermediate buffer gas, if present, is generally Nitrogen. Even if for some reason, air is used as separation gas and on top there was no intermediate buffer, air is not supposed to pass into the compression compartment because positive flow in primary seal is always maintained unless something is wrong with the sealing system ; that in itself would represent a hazard and explosive mixture could occur for a compressor. Instrument air (for instance control valve actuation) should not flow into the process gas stream too, should it be in the sealing system or in process system (anti-surge valve) unless there is a failure of some sort; cannot say more on that due to my experience being limited. However I suppose instrument air flow is generally extremely low comparatively to the main gas stream. If for some reason air as well as any contaminant is leaking into process gas stream from compressor, and that process gas stream specifically is recirculated in a much bigger process loop, there could be a risk that the quantity of air or contaminant will keep on increasing until this becomes critical, but I am not aware of such configuration occuring in gas lift processes, generally speaking.

"The well produces the gas, the compressor compresses some of the produced gas and sent back to the well for gas lift."
Not clear without a process diagram. Where is the Oxygen being detected? Is this detected in the line downstream the compressor going to the well, if so are there any other process gas streams which are branching into that line and what is their "significance" vs. the main stream coming out of the compressor ?

 

[georgeverghese]

Some other part of the process in this plant is recovering LP gas at subatmopheric suction pressure ( LP flare gas recovery compressor) which is recycled back into lift gas?

 

[engr2GW]

Thanks rotw and georgeverghese for your responses.
I have attached the PFD.
First thing is that Oxygen was detected in the inlet separator (1), the only feed into this separator are well head production and return of the gas lift gas injected from the compressor.
The VRU which could be a source of the O2 by pulling ambient air have been turned off and no longer in use for a Long time.

Oxygen was also detected at the facility where the sales gas (2) is going to.

That is what lead me to ask about the compressor WHICH BY THE WAY is a 3 stage reciprocating compressor. from the process flow, it is almost narrowed down to the compressor or the wellhead, but I don't know wellhead to have a level oxygen for a long period of time.

As much as possible, do it right the first time...

 

[rotw]

I cannot answer for recips type.

Just a very speculative remark otherwise: I suppose it is not a straighforward analysis to isolate compressor as root cause because in the middle of this there is the well which has its own dynamics.

 

[georgeverghese]

a)What happens to wellhead casing gas in your plant?
b)The PFD says suction pressure to 3stage compressor is regulated to 30psig - how well is this suction pressure maintained - were there incidents where pressure dropped below 0psig?

 

[engr2GW]

a) the discharge from the compressor that feeds the wellhead casing comes back through the production tubing mixed with reservoir fluids, so it comes back.
b) The suction pressure is maintained, not only by the valve set point, but by a recycle line. meaning, some of the discharge gas are re-cycled on occasion, when the suction pressure drops a bit to maintain a positive pressure.
the only point where gas may exist at sub-atmospheric pressure is at the tank, and the tank vapor have no interaction with this process.

As much as possible, do it right the first time...