Lube oil carry over

[B62plant]

I export gas via wakashau compressors which need lube oil injection. I have a filter coalescer fitted to take out lube oil carryover but I am only removing about 90% of the liquid. Can anyone recommend a better method or better system than the PALL unit I am currently using. The client stipulates zero liquid in sales gas.

 

[romke]

Maybe you can consider using a different kind of lube oil that does not easily take the gas to be compressed into solution. what kind of lube oil are you using and what type of gas are you compressing?

 

[georgeverghese]

There may be different types of liquid in this sales gas:
a) Carryover lube oil due to poor operation of this Pall coalescer as you say
b) High dewpoint hydrocarbons ( butanes and heavier) and may be even water in the vapor phase of this sales gas stream which is condensing out due to ambient or ground cooling. What is the max condensing temp of this gas at the max built up pressure in this sales gas line ( which may be line pack pressure at zero flow??).
Which one do you have, or do you have both ?

 

[B62plant]

Tried different viscosity lube oils, no effect. (Chevron 320 currently) I'm open to suggestions of any other lube oil or any advice on additives.

Carryover is 100% lube oil, no hydrocarbons. Cricondentherm is 2C and export temp over 20C (I'm in middle east so ambient temp not an issue).
PALL assure me the coalsecer they have sold me is 99.9% efficient according to their LASE test but I am injecting 17 litres of lube oil in to the compressors and getting about 2 litres a day of lube oil carryover in sales gas line.
I'm considering a second coalescer further downstream but PALL insist I am doing something wrong and are stumped as to the carryover volumes. Inspection shows correct fitting of elements.

 

[B62plant]

Does anyone have any experience of cyclone coalescers and if fitting may help ?

 

[EdStainless]

I suspect that you have line sizes, pressure drop. or velocities wrong.
Those filters are very effective.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[georgeverghese]

Ok, could we now see a PID of this coalescer vessel with level instrumentation /controls and surrounding equipment?

 

[B62plant]

PALL tell me the same, these are "state of the art" filters. They are baffled by the carryover.

 

[LittleInch]

OK, how about a data sheet / sectional drawing of the filter with actual conditions marked next to it?

mounted horizontally or vertical?

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

 

[B62plant]

It's a PALL Seprasol Plus® (CS605LGH13)
Vertically mounted

https://shop.pall.com/us/en/oil-gas...eprasol-plus-liquid/gas-coalescer-zidgri78m1q

 

[B62plant]

90 mmscfd
1200 psig
88 F
dew pointed gas

 

[georgeverghese]

Top compartment of this filter unit level draining should be automated, since this is where most of the lube oil in the gas will drop out. Right now, looks like this compartment is drained down to the plant closed drain header by operating procedures, presumably when operators see the 80% LAH alarm in DCS. Can you rely on the operators to do this all the time/ all year round?

If plant operators have left the ball valve on the top compartment drain line open continously, (also with BV232 at the bottom compartment either open or partially leaking) to allow a small amount of gas to bleed out the lube oil into the CD header (hence doing away with the neccesity to visit these filters every now and then), then lube oil and gas from the bottom compartment would flow in reverse up through the top compartment drain into the top compartment, bypassing the coalescers altogether. My guess is this is how you are losing lube oil into the sales gas line. A spade in the closed position at BV232 may help.

 

[B62plant]

I will look into how they are operating and get back to you, thanks

 

[B62plant]

The LCV from bottom chamber is kept closed (manual and auto valves) also BV 232.
The top chamber takes about 12 hours to build up to 20% level and is drained once a shift.
I'm satisfied they are not leaving the valves bleeding off from the top chamber or that the valves are passing.
I am not seeing any oil in bottom chamber at all.
I know the elements are fitted correctly because PALL witnessed the operation.
The unit is correctly sized for flow, temp & press because PALL designed it.
I am stumped. There is a zero tolerance for liquids in pipeline and I'm being threatened with closure if I don't resolve this.

 

[georgeverghese]

If you are confident of plant operations, and the filters are all fitted well, and gasket seals at each filter element are not leaking, that doesnt leave many other possible suspects:

a)These level transmitters are external displacer type, so presume you've cross checked its readout against the LG for accuracy.
b)You say lube oil loss is 2litres/day, but how do you know it is through these filters and not at the compressor unit lube oil degasser vent? Do you have an oil/hydrocarbon separator in the lube oil circuit?
c) Has the downstream user confirmed this is lube oil carryover and not hydrocarbon condensate? There are many gas plants which unknowingly produce offspec high dewpoint sales gas because some of the separators in the dewpoint control unit have unacceptable liquid carryover into the gas stream ..
d) How is the pressure at these sales gas filters maintained at a constant 1200psig? Do you have a backpressure controller / control valve at the final plant exit to maintain this pressure? Low operating pressure = high velocity at these filters = high liquids carryover.
e) Believe pulsating flow is detrimental to the operation of all types of vapor / liquid separation devices, including coalescers, which would be applicable in this case with these Waukesha recips. If PALL say this is likely, did they apply a derate on the permissible gas flowrate for these filters?

 

[B62plant]

Thanks for reply
a) yes, levels are accurate
b) 60 litres a month (2 litres a day) is being recovered at the pig receiver. No degassing at machines, all lube oil is carried over into discharged gas. Inject 17 litres a day. Mass balance based on what we drain at coalescer plus 2 litres adds up.
c) FTIR fingerprinting has confirmed it is lube oil
d) yes, pressure control at export line
e) no mention of pulsating flow by PALL in any of their reports but they state the flowrates are well within design spec

I will ask PALL about pulsating flow, thanks

 

[georgeverghese]

If pulsating flow from the recip compressors does have a negative effect on filter handling capacity, then one way of confirming this would be to run both coalescer units in parallel together to see if this LO carryover is reduced. Presume you have 2x100% units here.

 

[B62plant]

Already tried that,not because I suspected pulsation problems, I just tried it anyway as am clutching at straws now !! No effect. Got meeting with PIKO next week to discuss their elements, maybe PALL are just not compatible with Mapara 220 lube oil ?

 

[georgeverghese]

Another suspicion I have is that some of the lighter components in this degraded lube oil may have vaporised into the gas phase of this low dewpoint gas at the compressors. But the amount vaporised is still not enough to bring this new gas mix critical pressure above 1200psig when this gas is cooled at the compressor aftercoolers. So the LO remains in the gas phase at the filters. But at the other end of this pipeline where pressure may have dropped below this new critical pressure, this lube oil has condensed. In other words, the new gas mix Pc is <1200psig but higher than normal destination pressure.

With some assumptions, perhaps you could model this on your simulator ( say 2litres a day of say octane or C9 mixed with this total gas stream) and see what the new phase map shows (i.e. whether this mix has a critical pressure above or below 1200psig and also at the destination pressure).

If this replicates what is seen now, then I would see if there is a lube oil with less light ends in it, but that might change the viscosity characteristics required for lubrication.

 

[B62plant]

I think you are probably correct as I am taking samples at 3 points along the export line 1km and 2 km (total length 6km). I am finding about 500ml at each mid point. I don't have access to the end of pipeline as it is the client's plant but it is at this point that they are picking up the lube oil after pigging.
I think there is aerosol carryover and it is condensing the full length of the pipe system.
PALL insist though, that the filter/coalescer should take out 99.9& of all entrained lube oil down to 0.1 micron. I've calculated the concentration to be 7.56 ppm w/w.
I'm now considering adding a PEKO supplied additional coalescer at the pipe termination. The lube oil concentration is too low to use a cyclone system.