Frick Screw Compressor Variable Vi

[AlaskaLNG]

Hello,

I have a compressor I'm having some issues with. It's a Frick oil flooded screw compressor in Ethane/Ethylene mixed refrigerant service. I actually have two of them running in parallel on the same system. These are variable Vi machines and while I'm familiar with how the variable Vi works and what its function is, these two have me stumped. My understanding is that the slide stop is used to match compressor discharge pressure as close to system discharge pressure as possible to prevent under/over compression. The thing that baffles me is that these machines don't even have pressure transducers to read system discharge pressure, only compressor discharge. On one of these compressors in particular, the slide stop appears to have a mind of its own. It will be running great and out of nowhere the Vi will drop from 5.0 to 2.2 and it causes extremely high discharge temps when it happens. I've tried troubleshooting it, but without knowing what exactly is controlling the slide stop and telling it where to go, I've hit a dead end. Is there someone out there that can shed some light on what might be going on?

 

[zdas04]

Frick is not the best screw compressor on the market (not the worst either), but they absolutely have the best marketing. One of their selling points is that stupid variable VI. They like it so much that it is often the first change that is made in a rate-shed scheme. I hate the variable VI more than anything else on the skid, not because it is a bad thing to match VI to suction/discharge pressure, but that in applications like yours, the suction and discharge pressure never change. If the optimum VI is 3.1, then going to 5 will cause the machine to compress the gas to a much higher pressure (high temp) just to dump the excess across the VI plate. If it goes to 2.2 then at the end of the screw you are less than required discharge pressure and the gas has to compress the gas the remaining pressure (very inefficient).

Much is made of the inefficiencies caused by over compressing or under compressing. Here is a slide from my course that shows that over compression can cause a big reduction in efficiency, but under compression not so much.

If I was you, I'd figure out the optimum VI (P_disch=P_suct*(VI)^k), set the VI there lock it down, and remove any actuator that lets it (or forces it to) change. Make sure that the code in the PLC isn't trying to drive the VI (it probably is and you'll need a programmer to exorcise that function). In air conditioning service, variations in load are easily handled by the slide valve if the driver is not variable speed.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[AlaskaLNG]

Thanks for the reply David. This refrigeration system is actually used as a pre-chiller in a small scale LNG plant. Suction pressure is set manually via plant computer and the slide valve loads and unloads accordingly to meet the suction pressure. The load we put through the tube side of the ethane evaporator varies depending on whether we're trying to make more or less LNG at that time. Discharge pressure can also vary greatly because it uses a propane refrigeration system to condense the ethane. Being located in Alaska, ambient temperatures fluctuate greatly and have a profound impact on discharge pressure. The propane side of this equation works great and I never have any issues with it. It even has the separate pressure transducer for reading system discharge pressure and is plumbed in to the discharge pipe downstream of the oil separator.

With all that said, I could easily manually set the slide stop position and pull the fuse for the solenoid valve that runs it. Under most conditions, I think it would run fine like this and if I felt the need to make a change I could put the fuse back in and change it to what I need to change it to. I would just like to figure out why it seems to have a mind of its own and jump around for no apparent reason. The only thing I can think of is it's written in to the logic, but I don't see any kind of process parameters change that would cause it to make that adjustment.

 

[georgeverghese]

Am no expert at this, but I've read that it would be better to use variable speed on oil flooded screw compressors to enable capacity control in order to avoid this high internal temp problem ( also reduces thermal degradation of the lube oil), and avoid using the slide valve to the extent possible. With VFD, it is possible to go down to less than 10% of rated speed on an electric motor if an aux motor cooling fan is included.

The statement " these machines don't even have pressure transducers to read system discharge pressure, only compressor discharge." - is confusing.

 

[zdas04]

Skids I design always have capacity control that: (1) Only uses the slide valve when speed is at min or max; (2) has speed changers (either VFD or variable governors) for primary capicity control; (3) when the speed changer is at a max/min point the slide valve comes off shut to try to match the screw with the demand; (4) has a suction control valve to prevent high suction pressure that is not part of the PLC (it is local sensing); and (5) have a low suction makeup (recirc) that is a last resort and is not controlled through the PLC. I've built several hundred of these skids and they seem to work really well.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[georgeverghese]

From your description, the compressor slide valve operates in mix C2/ C2 = suction pressure. Since you have 2 of these compressors running in parallel, could it be that these machines are running on split range suction pressure control ? That may explain why one of these machines drops down in capacity at low LNG flow / low refrig demand.

In principle, the control concept David has outlined would be preferable.

It does sound like you dont have a variable speed driver at the moment, hence your thought to disable the actuator that operates the slide valve. One interim measure (till a VFD is implemented, assuming this is not on split range control, no mechanical faults etc ) would be to enable low suction pressure recycle from the C2 / C2= condensor ? That may not be easy either. Running start / stop may ruin the driver / compressor shaft bearings.

 

[AlaskaLNG]

It does not have a variable speed drive. They're GE 450HP, 3650 RPM electric motors. They are not run on a split range suction controller. Each compressor is controlled via PLC to meet a certain suction pressure set point, which can be changed individually by the plant operator at any time. I'm not having any issues with the slide valve. It appears to work as it was intended to do. The slide stop is another issue. This place is run on a shoe string budget and there's zero chance of upgrading these machines to a VFD, but I'm going to try setting the slide stop manually and pulling the fuse to see if that solves some of the issues.

George: From my understanding, a slide stop is used to essentially change the location of the discharge port within the compressor. It's done to be able to match the internal compressor discharge pressure to the discharge pressure of the rest of the system to prevent over or under compression. In order for the PLC to control the slide stop, I would assume that it needs a pressure transducer to read the internal compressor discharge (which it has) and one to read the system discharge pressure (which it does not have). Unless of course there's some sort of volume ratio code written in to the logic in the PLC, which it appears may be the case. Hopefully that clears up some of the confusion with my original statement.

 

[georgeverghese]

Okay, guess that is another way to operate these 2 screws in parallel.

Must say I dont know much about this slide stop, but perhaps it is required for startup operations when there is low external discharge pressure, and there isnt enough dp between internal discharge and external discharge to keep the lube oil recirculating? Hence the discharge slide stop moves to a choked position to create sufficient dp (to keep the lube oil running to all LO injection ports)?

So perhaps this slide stop is operating to maintain a predetermined fixed casing discharge pressure at the moment all the time, which is high enough to maintain or exceed this required dp even when the external discharge is at a presumed high pressure.

So perhaps the slide stop is going into this throttle position when it sees lower than set discharge pressure, which may be when the C2 / C2= condensor is at a lower condensing pressure?

Maybe a better way to operate the slide stop is to maintain a fixed DP between external discharge and internal casing discharge pressure as you suggest. Presume this slide stop can be continously varied in position rather than in large steps.

Putting this slide stop in a fixed position as you suggest now may work for high volumetric throughputs but not at lower throughputs ( at lower throughput, the choking effect at the slide stop would be less, and LO recirculation rate would be less)?.

This doesnt sound like a difficult problem - getting the Frick compressor vendor to work with you on this is half the work done already I'd think.

 

[zdas04]

There are two different adjustments on the Frick that can be programmed. The one we were talking about is the variable VI. That is the one I suggested disabling.

The other is the unloader which is called a "slide valve" on the Frick (and most process screws, it is either a turn valve or a poppet valve on air-derivative screws). The slide valve should be left under PLC control.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist