Vacuum Pumps

[SeanB]

I am looking at installing a Vapor Recovery Unit at a marine terminal. We have two options available for the vacuum pump, a dry vacuum pump and a liquid ring vacuum pump. I was hoping someone could comment on the reliability of both and any preference of one over the other.
Thanks.

 

[sshep]

SeanB,

Liquid ring pumps (LRPs) are very reliable and forgiving, and where you have a choice these are my first prefference. Usually what drives the decision away from LRPs is the depth of vacuum needed (in which case these can often still serve as one of the stages), and the availability of a suitable sealing fluid and a place to send it when contaminated. An LRP will require more ancillary equipment if the seal fluid is via recirculation loop (i.e. seperator drum, seal liquid circulation pump, filter/coalescer, and controls), but the ease of operation and reliability still make LRPs my first choice. This is my experience from numberous types of vacuum system installations for columns, others may have different experience.

best wishes,
sshep

 

[quark]

The main drawback with liquid ring vacuum pumps is that you should have some quantity of condensable gases present in the process otherwise they malfunction. This quantity depends upont the sealing fluid properties and capacity of the pump. Secondly, corrosion of internal parts is an issue. Temperature of the fluid is very critical with respect to level of vacuum required.

Though costly, dry pumps overcome these difficulties. They are quite reliable if alignment is proper.

There is no apple to apple comparison in this case.

 

[rmw]

I don't understand Quark's comment about LRVP's malfunctioning if there is no condensate from the vapors, but it does bring up the thought that you will have to be careful in you selection of the seal fluid. Depending upon the liquid you choose, the process, if it does have any condensable vapors will either contribute to the liquid, meaning that you have to deal with the overflow, or it will contaminate the seal liquid, again potentially causing disposal problems.

rmw

 

[quark]

rmw,

Thanks for the correction. It should be non condensable gases. The idea is to take care of pump cavitation and to provide minimum gas flowrate into the pump.

 

[pmover]

any possibility of using an ejector (thermo-compressor)? a source of motive gas will be needed and then a location for the exhaust.

good luck!
-pmover

 

[SeanB]

Thanks for all the useful replies thus far. The vacuum pump will be required as part of a Vapor Recovery Unit - recovering gasoline vapors at a marine terminal. So there will be vapor condensing in the pump. This will be vendor supplied skid - they offer the two types of pump we have been discussing. On the liquid ring pump they provide a 3-phase separator to separate the condensed gasoline from the seal fluid, which will be ethylene glycol. I fear have the seal fluid contaminated with gasoline - thus the reason why I am leaning towards a dry seal pump.

 

[sshep]

SeanB,

By using an aqueous seal fluid and liquid-liquid seperation, your seal fluid becomes saturated with gasoline but is not really contaminated in a way that would require blowdown and make-up of seal fluid. Water, soluble alcohols, etc would be the problem contaminates for an aqueous seal fluid. Where to route the wet gasoline phase, and possible phase seperation problems may still be important seal fluid considerations, but I wouldn't completely rule out an LRP yet. Incidently my experience with distillation tower liquid ring vacuum pumps is that the seal fluid can usually come from a suitable tower bottoms and contaminated seal fluid sent back with feed to the appropriate tower- such infrastructure often makes LRP seal fluid decisions easy.

As quark pointed out, dry vacuum pumps are reliable if they have proper alignment (or more generally if you have a maintenance department which can properly support this type rotating equipment). Despite a personal liking for LRPs, the plant where I am working now has over a dozen rotary claw vacuum pumps and we recently came up from a TAR with only one vacuum pump failure and a few minor trips. My concern is that a marine terminal application seems like it might require alot of starting and stopping and also be short of qualified maintenance staff needed to insure reliability.

There are control considerations which also factor into the decision. My experience is that control systems including start permissives and trips protecting against conditions of backflow, explosive mixtures, temperature, pressure, vibration, etc, often cause more vacuum system cut-outs and start-up problems than actual rotating equipment malfunctions. You would think with so many thousands of systems in operation that there would be a standard implementations of controls and safeguards, but I find every installation seems to be a custom job.

Lastly, all vacuum systems that are actually controlling pressure at a setpoint needs some "noncondensibles" to chew on. Regardless of what spillback, speed, or throttling control that was originally designed, if there is no steady flow of noncondensibles then I find that most systems (dry or wet) end up getting a supplemental bleed of N2 (or CH4, air, etc) installed later.

best wishes,
sshep

 

[oxilume]

SeanB,

pmover recommended using an ejector. This is a step in the right direction. However, use an eductor with a low HP exposion proof pump and you have a simple system that is virually maintenance free.

The condensible vapors will be condensed by the motive fluid. The non-condensible gases will have to be vented.

However, a properly setup venturi scrubber will take care of vapors plus scrub the off-gas.

If you decide to use any type of dry vacuum pump, ensure that the lobes (pd type blower) are made of non-sparking material.

Also, check all applicable Coast Guard Regs!!!!!

I have a copy of the most current regs regarding explosive vapors for vacuum service or conveying material offshore.

DO NOT USE any type of plastic piping or hose unless it is grounded and bonded. Non metalic hose certified for this type of use will have a wire exposed inside to allow dissipation of static electricity.


Todd

http://www.oxilume.com

 

[MedicineEng]

CEPIC makes steam ejectors systems for vacuum production. The feedback that I have from these systems is that they might take sometime to achieve the vaccuum required (specially if near -1bar),but then they are reliableand trouble free.

 

[sshep]

A steam ejector will produce (a potentially large amount of) gasoline contaminated condensate. This will have to be steam stripped or otherwise treated. I loved the reliability of ejector based vacuum systems, but now days they usually are environmental dinosaurs in these type services.

A pump-eductor system has essentially the same seal fluid considerations as a liquid ring pump system. The LRP is pump like equipment with similar reliability to an API pump in the same service. The other parts of the system will be similar (vapor-liquid-liquid seperator assuming aqueous seal fluid). We bought such a package (

http://www.hijet.com)

several years ago for hydrogen degassing service. The system works pretty well, but I understand their North American subsidiary is no longer in business. This technology for vapor recovery has not really caught on yet, but does seem to be viable.

best wishes,
sshep