2-phase Separator and Coalescer Sizing Help 1

[SEN_CHE]

I am in a project engineer role for my company. We typically use EPC firms to design and install new gas plants. I don't directly specify equipment sizes, but I have to live with the consequences. On some compressor skids, the separators/coalescers are designed by the skid manufacturer. It seems that regardless of separator vendor or EPC firms, we run into liquid carryover problems with the gas/liquid separators/coalescers, and I can use some practical advice for avoiding these problems in the future.

The examples listed below occurred for new plant construction (versus a modification to an existing plant).

We had an issue with a propane refrigeration compressor (screw compressor) skid - the horizontal oil separator on the compressor discharge had significant carryover issues. We worked with the skid manufacturer, who insisted that the separator was sized properly, trying all sorts of process changes to pressure/temp, until the manufacturer ultimately relented and installed a larger vertical coalescer downstream of the first one.

In a second example, we had a coalescer downstream of a reciprocating compressor to separate compressor oil from the gas stream. This one was specified by the EPC firm, and bid out to several companies which are commonly known vendors for our industry. The EPC purchased a vertical coalescer. This coalescer showed significant oil carryover. We worked with the EPC and coalescer manufacturer to resolve this issue, all the while the coalescer manufacturer insisted that their sizing was correct. Ultimately, we were not able to resolve this problem with the original manufacturer, and we purchased and installed a horizontal coalescer from another manufacturer.

In both cases, the carryover occurred when gas flows/pressures/temps were steady - we weren't having multiple start-ups/shut downs of the compressors where you might get oil slugging. In both cases, the amount of liquid carryover was significantly larger than the expected based on filter efficiency.

What do we need to do differently in the future? Should we ask the EPC firms to automatically add a significant overdesign for gas flows? Should we insist on a specific maximum velocity through the separators?

 

[pierreick]

Hi,
To support your work a few documents .
Pierre

 

[pierreick]

hi,
One more document.
Good luck
Pierre

 

[georgeverghese]

What was the separating device used in the liquid phase ( between oil and propane ) in the C3 compressor discharge V-L-L separator ? Cartridge type coalescers?

Problems with L-L separation in this service are notorious in the industry with screw compressors. Believe the problem is to do with thermal degradation of the oil due to excessive recycling within the compressor body when running at turndown with slide valve for most of the capacity regulation operating range. Which then results in high oil solubility in the propane phase ( for an oil which should have been immiscible in C3). Solution is to use VFD for most of the capacity recycle and only rely on the slide valve for the last few % of the turndown (which normally should be infrequent) or not at all, if the VFD can handle the entire turndown. This will then minimise lube oil thermal cracking / emulsification in C3. Most package vendors reluctant to offer VFD for screw compressor turndown to minimise dependence on the slide valve, probably due to added cost.

There may be retrograde condensation downstream of the filter coalescer in multicomponent gas separation ( due to pressure drop across the coalescer)? Also, there are some additional flow capacity margins to be included in recip compressors with pulsating feed flows (I dont have these design margin values with me now). In any case, a guard suction scrubber is necessary upstream of the compressor. And suction line d/s scrubber to the compressor cylinder should continuously slope back to the scrubber.