Crew compressor as make-up compressor 2

[NghiaPP]

In hydrotreating process, the make-up hydrogen is supplied by a make-up compressor with discharge pressure in the range 50 to 90 bar. Suction pressure is around 20 bar. Gas purity is 90% H2 or higher. All make-up compressors, I have seen are of reciprocating type with a stand-by machine. The stand-by machine is needed due to the frequently maintenance. I ‘m wondering why not use a crew compressor in this service. Time between maintenance services is much longer and the standby machine may avoid.
Is the reason that crew compressor can not designed for high discharge pressure or can not be oil free at the required operating condition? or any other reason?

 

[zdas04]

I have never heard of a "crew compressor" so I googled it, and Google asked "did you mean screw compressor"? So I'll ask the same question, are you talking about screw compressors? There was a link for a used "crew compressor" from Kaesor, but they are talking about "screw compressors".

David

 

[NghiaPP]

Sorry, I meant rotary screw compressor

 

[zdas04]

Let me get this straight, you are doing static tests with hydrogen at 50-70 bar(g) (725-1015 psig)? I have never heard of anyone doing that. I have often heard of people doing tests with helium at very high pressures, but never with a gas that explosive.

To address your basic question: Most screw compressors (flooded or dry) are built for much lower maximum pressure. The market that most manufactures see as viable is MAWP of their compressor case is generally under 300 psig. There are exceptions, and I can think of one flooded screw with a casing MAWP consistent with ANSI 600 flanges. None of the flooded screws I've ever seen packaged will give you 99+% oil removal, so you'd be putting some amount of oil into your hydrogen. I don't know how bad this would be, but it wouldn't be good.

On the other hand, industrial recips have oil injection into the gas to lubricate the cylinders. The amount of oil carryover can be quite large and is often bigger than a properly set up flooded screw. If you can tollerate the oil from a recip-lubricator, then a flooded screw shouldn't be a problem.

Dry screws might in CO2 service have had serious lubrication issues. I would think that in H2 service the issues would be worse.

David

 

[DLite30]

hydrotreating--hydrocarbon treating process...not hydrotesting, Dave.

Those pressures are probably at the very limit of what you can economically achieve (purchase and installation costs) with a screw compressor.

I would consider a centrifugal compressor before I would a screw compressor.

 

[zdas04]

In the immortal words of Emily Latella "Never Mind".

Once your mind starts down a path, it is hard to shift it to a sensible path.

I'd agree that this application is better for a dynamic compressor (centrifugal or axial) than a PD compressor.

David

 

[NghiaPP]

Gentlemen,
I don’t agree that centrifugal compressor would be suitable for this service. The high compression ratio and the low molecular weight (2-4) will require very high head compressor.
I give an example for the case with 99% hydrogen gas. The head is 1430 to 2650 kJ/kg or 480,000 ft to 880,000 ft. For centrifugal compressor, the number of impellers will be excessive (48 to 90 impellers, assumed one impeller provides 30 kJ/kg). This means 4 to 8 compresor casings in series are needed.

 

[zdas04]

He didn't say what the suction pressure was, but it is common to use a booster to go from 80% up to 100% of nominal system pressure. If the discharge is 90 Bar(g), then I would expect the make-up to come on when the pressure is 72-81 bar(g)--1.24 ratios, a simple single stage dynamic application.

David

 

[NghiaPP]

In my first email, the suction pressure was given at 20 bar. It is typical pressure from PSA units. One or two stages reciprocating compressor was used for the most of units, I have seen. I have just wondered that no of them was a screw compressor.

 

[DLite30]

Depending on how long ago the original make-up compressors were installed there may not have been a screw compressor available that was as economical (especially if the discharge had to be oil-free) or reliable as a recip unit...or maybe the people making decision on the equipment didn't feel comfortable or know enough about screw compressors to install them.

Even today you'll probably find that if you priced out purchase cost, and installation of cost of a screw compressor, driver, acceccory systems, etc, that a recip will be cheaper.

Initial installation cost go along way in swaying a decision between maintenance costs of two options.

There's a myriad of reasons of why someone may have made the decision that they made.

But physically, you should be able buy an off-the shelf screw compressor to do the job you describe.

 

[dcasto]

The big issues with screws are the slip and losses which make their efficency very low as the ratios go up. At 5 ratios, they are under 50% eff versus 80% for a recip.

 

[zdas04]

I've seen similar slip numbers for a dry screw, but for a flooded screw, the slip is at least an order of magnitude less. At 10 ratios I get about the same energy input per unit mass compressed with either a flooded screw or a 2-stage recip.

David

 

[albertgrigoriev]

We design screw gas compressors so for typical screw gas manufacturer it is not easy to work with light gases.
They have low viscosity and rotors have high rotation speed.
But majors as GE, Man Turbo, Kobe, Howden are able to offer both dry and oil-flooded options, Mayekawa and Aerzen are able to design oil-flooded one.
Max pressure for screw compressors is about 100 bar for oil-flooded and above 60 bar for dry type.

So it is quite possible to use screw machines for hydrotreatment applications.

Albert