Desalting Natural Gas 1

[geooffy]

Hi, I was wondering if anyone has any experience removing salt from natural gas.

In my situation I'm seeing salt depositing in a compressor downstream of an inlet separator. The inlet separator is quite large for this application so I shouldn't be seeing any free water carrying over.

I also conisdered that the salt might be carrying over as submicron aresols, but the compressor feeds a glycol dehyrator after which there is a sales meter which also regularily plugs up with salt (there isn't anywhere but the inlet gas that this salt could be coming from).

Does anyone think that salt in aresol form could carry right through a dehydrator? Any other ideas on what could be happening here?

Thanks.

 

[gr2vessels]

geoofy,
The inlet separator, I believe is meant to be a gas filter coalescer, to separate liquid of any form present in the gas stream, except the vapours. However, the separator is not 100%, so you could see disolved and suspended salts carried over to the dehydrator and having the water removed;- but there, some of the salt will/could be transferred into the pure glycol and some salted glycol droplets be carried down to the meter. It normally shouldn't all this occur, if the initial separation would remove ALL the salts disolved / suspended in the water and if no glycol would be present in the downstream gas.. Just be aware also that you could have the liquid separation almost 100% in the upstream of the dehydrator, but carry oner more pure glycol droplets than initial liquids into the meter, depending on the design and operating conditions (that's again, another issue). In conclusion, it looks to me that the inlet separator is letting way to much liquid carry over to the dehydrator, where the salt could contaminate the dehydrated glycol and droplets of this glycol end up in the meter, perhaps even further, downstream in the sales gas piping...
cheers,
gr2vessels

 

[zdas04]

The mist extractors in mechanical separators tend to coalesce water droplets greater than about 20 microns into drops that are affected more by gravity than by buoyant forces. The droplets that get through can represent a large mass of produced water. If the produced water is around 10,000 mg/L TDS, then each bbl of water that is evaporated in later pressure drops will deposit 3.5 lbm of salt which represents a substantial volume in control devices and compressor valves. Recip compressor valves are especially vulnerable because the surrounding metal is so hot that any droplet that touches bare metal will boil rapidly.

I've been fighting this problem for a decade in wellsite compression. Sometimes forcing a pressure drop in a convenient location can prevent deposition in an inconvenient one. I've occasionally had good luck with witch's hat strainers (or duplex basket strainers for more important streams) creating a dP that is large enough to force a phase change, but most times the dP is inadequate.

My best luck has been in designing facilities that it is reasonably easy to flush the accumulated salt with hot water (for NaCl) or strong acid (for Nacholite). I've never found a reliable way to keep it out of control valves and compressors.

David

 

[geooffy]

Thanks for the help. Do either of you have any experience with desalting this kind of natural gas?

I'm thinking that if we were to install a filter coalescer it would plug off all the time. But what about some kind of water wash in a spray tower? I'm sure it would use a tremendous amount of water, but has anyone given it a try?
Thanks.
Geoff

 

[dcasto]

At our underground gas storage in embedded salt we saw what you described. We iject fresh water upstream of the coalesoors and had a continous blowdown of brine for disposal.

 

[chief]

I have found cyclone separators effective in removing entrained solids in gases, probably would be just as effective removing salt particles from natural gas. Likely to separate out some condensate as well,

Offshore Engineering&Design