5 replies to this topic

[mert]

Dear all,

I am designing an vertical oil separator tank for oil-injected screw compressor. I have done detailed literature review about it but I couldn't reach enough documents about compressor oil tank. Most of the articles are about the cyclones or other tanks that are being used for chemical processes which are very large in dimensions. Can you suggest me some articles about this issue? Also, if you have any design spreadsheets or documents about the sizing, separation efficiency and pressure loss of the separation tank I will be glad to recieve them.

[Art Montemayor]

Mert:

Designing a vertical oil separator for the discharge system of an oil-injected screw compressor is not a big problem. This subject has been discussed, debated, argued, expounded, and explained many times in our past Forum threads. Various design spread sheets have been developed and uploaded into our Forum threads – including mine. To obtain this valuable information you should spend time researching our Forums by using the SEARCH engine feature in our Forum website.

However, the subject of a warranted and guaranteed design for an oil-injected screw compressor is a totally different subject. This application is a very specialized and unique one. I would not attempt to do it myself unless I had the use of lab and plant tests done on my design products – and those would have to be specific to the actual application. A lot depends on the quality and type gas you are compressing and that is why I know of no one that would take on this application except for the screw compressor manufacturers themselves. That is the way it is done in industry – and it is for a very good reason. The proper functioning of such a device has a marked effect on the efficiency and quality of performance of the screw compressor itself and may even be a jeopardy to the machine if not done correctly – or applied correctly.

I presume you are in Turkey and you want to (or have been assigned) design and fabricate such a separator. My recommendation is: Don’t. You probably have an oil-injected screw compressor already and if so, then you should refer this application to the compressor manufacturer.

[mert]

Dear Art,

Thanks for your reply. I have checked the forums and I have found some documents before I posted my topic. I checked them again after your advice but I still couldn't find a document that is exactly about my issue. Or as being a newbie here I don't know where to look exactly.

Let me tell you my problem again. I want to design an Oil/Air separator tank for oil injected rotary screw compressor. I have drawn a basic separator tank model for you to understand my problem more clearly. The separator should use centrifugal and gravitational method together. I have only found documents about design of gravitational separators in the forums. As you know compressors are not very very big machines so my separator tank should be small in dimensions (1-1.2m height , 350-450 mm diammeter). And I want to know how to make the design and sizing of the tank.

Edited by mert, 06 December 2012 - 10:58 AM.

[Art Montemayor]

Mert:

Attached are some workbooks for calculating vapor-liquid separators.

I don’t know whether you took the time to read my previous response, but you should heed the advice I offer there.

One of the main problems related to oil-flooded screw compressors is that the oil employed to seal the rotors and lubricate the machine has hydrophillic properties. It has a tendency to absorb any water moisture present in the gas being compressed. That changes the characteristics of the oil in a really bad way. It causes oil breakdown, corrosion, and bearing failures and further damage.

Additionally, the separation efficiency has to be very high in order to capture and return oil droplets back to the circulating system and not have them depart with the compressed gas. An internal filtering mesh device is almost certainly mandatory. Extra instrumentation and alarms have to be incorporated with the compressor in order to allow it to work normally and continue its lubricated and flooded service.

The actual, real separator design is based on empirical factors and experience. Calculations alone are not going to permit a successful sizing and operation.

I asked several direct questions which you have not addressed. You say you “want” to design this separator. Is this because you have been assigned to do it? Or is it because you are bored and want to have the experience? Or is it a school assigned problem? I can understand an academic assignment; however, to do it for a real-life application incurs a lot of responsibility and liability. That is why I don’t recommend doing it.

Attached Files

•  Art's Liquid-Vapor Separator.xls   116.5KB   226 downloads
•  Sizing Cyclone Separators.xls   109KB   187 downloads

[mert]

Dear Art,

Thanks for the reply. I have been assigned to design a separation tank with the theory behind it. I am working at biggest compressor manufacturer in Turkey. We have been manufacturing several types of separation tank for years however, the designs are only done by experience and tests so I would like to make a design which is based on theoretical calculations and CFD analysis. I may do the whole job by CFD but I want to learn the theory also. I will look at those spreadsheets in detailed I hope they will be enough to understand the theory. I will let you know and discuss the theory with you if I complete my design.

[ColinR33]

Dear Mert,

I have designed a number of these units as I used to work for a packaged equipment designer/manufacturer. We typically used a vertical two compartment design. The bottom section is for bulk separation of the lube oil and vapour, and provides residence time for the lube oil pumps. The upper section contains coalescing elements to remove finely dispersed droplets of lube oil from the vapour. There is a tubesheet between the two compartments and the vapour flows upwards through the tubesheet and through the elements in an inside-out flow. Any condensed lube oil collects below the coalescing elements (they are on risers 8-12" high to elevate them above the liquid level) and is drained back to the compressor suction through 1/2" tubing with a needle valve. A level gauge is provided to monitor the L.O. level in the upper section and allow the operator to adjust the needle valve to get a relatively stable level. The bottom section has at minimum a low level switch, level gauge, temperature transmitter and fill connection. A mist eliminator can also be provided and in colder climates a heating coil is provided for startup as there is typically a minimum lube oil temperature permissive for compressor startup. Art's typical separator sizing info is applicable to sizing, CFD is not necessary, and the coalescing element vendor will tell you how many elements to use. In my experience the bottom section usually drives the vessel diameter based on vapour velocity and liquid holdup (30sec to 1 minute for a small closed loop system such as this).

Cheers,
Colin