Discharge Temperature of Screw Compressor 4

[EmmanuelTop]

Good day to everybody,

In many discussions we have had in these forums, I picked up a recommendation to keep the discharge temperature of an oil-flooded screw compressor above 100 degC so as to avoid any possibility of water condensation and contamination of circulating oil.

I have a case here of compressing 98% Methane-rich natural gas, saturated with water at suction conditions (25 degC, 3 barg), to the discharge pressure of 15 barg. The offers that we get from different vendors (e.g. Mycom, Howden) are quoting 80 degC as discharge temperature of the compressor, at 15 barg discharge pressure.

The questions are - should we accept these discharge conditions, or should we insist on minimum 100 degC discharge temperature? And how and at what cost/trade-off this can be achieved?

Many thanks for your time.

Regards

http://antwrp.gsfc.nasa.gov/apod/astropix.html

 

[zdas04]

The trade off is really easy to calculate. When I had about 90 C discharge temperature on a fleet of screws my replacement oil costs were on the order of $20k/month. Raising the temperature to around 100 C (98 C to 103 C is target I recommend) lowered that to about $2k/YEAR. At 80 C outlet, I would expect 30-40% down time (so a 1 MMSCF/day machine would have over $400k/year of lost revenue at $3/MSCF), and probably $3-5k/machine/month of make-up oil (call it $40k/year/machine). A proper oil-temperature management system (INCLUDING an oil pump, the manufacturers often say that the dP across the skid is adequate, it isn't) will add something like $10-15k to the capital cost of a skid.

Plant guys find it nearly impossible to accept how difficult the fluids encountered on a wellsite really are. 80 C is a very common setting for a 3-way TCV that controls the temperature into the process and doesn't look at the temperature out of the process at all. This is the normal plant design. Remember that inside plants (e.g., a refrigeration loop): (1) the mix of gases is very constant; (2) the suction and discharge pressure and temperature are very constant; and (3) the gas is dehydrated. Are ANY of these true on a wellsite? Sometime put an online GC on a well stream and plot the mix of components on a minute-to-minute scale, you'll be amazed at the variability of a "mostly methane" stream. As water ebbs and flows in the near-wellbore, pressures change dramatically. As flow rates change, the exposure time to uphole aquifers changes and the temperature of the gas changes. A temperature control scheme that only looks at the inlet oil temperature is grossly inadequate to the task of cooking the absorbed (NOT condensed) water out of the oil. In a few days at 80 C inlet your oil will look like coffee with extra milk, the level will increase in the sight glass and then drop rapidly as big droplets start to fail to coalesce. Then you'll lose a machine to bearing failure. Then you'll notice that your mechanics are carrying drums of oil on their trucks.

At one point you were considering scheduling my class for your field. If you are talking to compressor manufacturers it may be too late to get maximum good from the class. One of the things I teach in that class is that the screw manufacturer is nearly the least important decision that you will make with regard to screw compressor deployment. Mycom and Howden are both fantastic machines that will provide absolutely horrible service if not properly packaged. The packager is the key. Having independent expertise working with the packager is the only way to keep the packager honest. I do these sorts of design reviews all the time and while they are kind of expensive, following my design recommendations has resulted in a number of screw compressor installations whose primary LOE component is energy input not replacement oil. You need to change your focus immediately from manufacturers to packagers and find someone who can keep the packagers honest.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[EmmanuelTop]

Dave, thank you for the reply.

1. I do not understand the role of a packager in this case. Isn't it the manufacturer who designs the system and sets the discharge temperature, requirement for the oil pump, etc.? What is the role of a packager with regards to this subject? I apologize if this is a dumb question, but I have very limited experience in design and procurement of screw compressors.

2. How the discharge temperature can be effectively controlled? By the oil flow, oil aftercooler temperature, artifically raised discharge pressure, or something else?

3. How can we be sure that the selection of lube oil is done correctly? The Company has had some really bitter experience with operating screw compressors in offshore fields, where they process rich (condensible gas). That was different application though - our gas is not condensible. Are the screw compressors compatible with rich gases?


Regards

http://antwrp.gsfc.nasa.gov/apod/astropix.html

 

[zdas04]

I've never seen a screw compressor packaged by a screw manufacturer. Some (like Ariel) have a packaging standards document that defines what must be done to protect the manufacturer's warranty, but most don't and even Ariel's is pretty generic. The Packager gets the compressor, the driver, the coolers, and (often) even the vessels from manufacturers and then builds the skid, assembles the components, and does the piping and instrumentation. The way the bits are assembled and connected is absolutely the key component to success.

Discharge temperature is a function of: (1) gas suction temperature; (2) gas compression ratios; (3) gas flow rate; (4) oil flow rate; and (5) oil inlet temperature. You have zero control over 1 and 3. You have some control over 2, but the wasted energy of throttling the suction or discharge is pretty high. Your big levers are oil flow rate and oil inlet temperature. I've been working with a client in South Australia this week on a temperature control scheme that is new to me and really exciting. They have a glycol loop that provides gas cooling, oil cooling, and engine cooling in sequence. Since gas outlet temperature has a max, the gas cooler is first in the loop. The engine is next in the loop, and finally the compressor oil. All three coolers have a 3-Way bypass to send more or less glycol around the cooler. The glycol out of the oil cooler goes to a fin-fan air cooler. This scheme provides a constant heat sink to the oil in virtually all weather. On the oil side, the flow splits right after the oil pump to send constant flow to the bearings. On the injection side it goes through a (PLC controlled) valve that sends more or less oil to the screw based on compressor discharge temperature. I really like the thermodynamics of this design. I really hope that it works like the models say it should, the first three are about to be built.

Most of the compressors I operated were in CBM and the fluids were CH4, CO2, and water vapor. I sent the gas analysis to an oil manufacturer and they recommended an oil that worked well. When a colleague complained about the oil performance in rich gas we sent his gas analysis to the oil manufacturer and their recommendation was VERY different from what he was using. He changed to the recommended oil and the contamination problems went away (the recommended oil was a semi-synthetic blend that was not quite unique for him, but pretty close). So yes, oil-flooded screws work in rich gas as long as the oil is compatible with the fluid components (you do not want your screw compressor oil to act like the Lean Oil in an NGL extraction process).

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[EmmanuelTop]

With whom we are supposed to work on the discharge temperature subject? With the manufacturer, or with the packager?

If the manufacturer has come to us with 80 degC discharge temperature, my guess is that we have to speak with these guys.

Thank you

http://antwrp.gsfc.nasa.gov/apod/astropix.html

 

[poli60]

Hi, interesting discussion.
With my (limited) experience in this sector, I share Dave opinion... but I've seen a Manufacturer (Howden) making (good) packages, not on exclusive basis (ie, they also work with selected packagers). They have good experience but it's not very easy to approach their technical people, screened by the sales men... and you, as "technicians", could easily imagine which kind of troubles can create a "vendor"
In which area is the project? In which stage? Are there many Manufacturers/Packagers involved?
Good luck!

 

[apetri]

I think with both manufacturer and packager,
the usual trick is to maintain the oil at temperatures
ABOVE those of condensed fluids to facilitate
evaporation of everything which may produce degradation,
the details (the additional temperature, the maximum
temperature etc.) must be verified with manufacturer
keeping in account the characteristics of selected oils.
The selection of a proper oil is also very important,
there are many different types mineral,
synthetic oils etc. with different
properties (densities, viscosities, resistance to oxidation tests,
pour point tests etc.) and (of course) costs,
different oils will require different drain intervals
and may produce different results in compression cycle,
hence it needs and expert to put all together.

 

[zdas04]

The times I've gone to manufacturer's facilities I've become convinced that they are very well divorced from any understanding of how their machines are being used. I've seen Howden screws packaged by Howden and and I wasn't very impressed. I like Howden screws a lot, but as I said, they don't know enough about the end-use that their packages will be put to and they are quite inflexible about meeting your site needs. Their compressors work a lot better when they are packaged by others. I try to let the packager deal with performance-envelope problems.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[EmmanuelTop]

Any packagers in Europe that you can recommend? We have received offers from Howden and Mycom dealers in Europe.

http://antwrp.gsfc.nasa.gov/apod/astropix.html

 

[zdas04]

Sorry, I've never been in that market. I know some good ones on the other side of the world, but that doesn't help you much. Exterran has offices in Moscow, Italy, and Switzerland, I think they are all sales offices and the actual equipment would be built in Houston or Singapore, but that is not all bad. They have packaged some good machines for me, but I still feel a strong need to approve pre-construction drawings and I like to visit the fab shop to make sure that they know I'm going to notice smoke being blown up my pant legs.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[poli60]

I'd recommend GEA Refrigeration people in Italy (aka Technofrigo).
They are very expert in this field.

http://www.gearefrigeration.com/