variable frequency drive

[gapatel]

has anybody tried variable frequency drive for resiprocating ammonia compressor ? does it save energy on resiprocating type compressor load ?

[Peter Mitchell]

Hi
energy saving on resiprocating compressors is very limited as the duty/efficency of this type of compressor drops away very rapidly with speed. (inverse square) If you had screws then you can vairy the speed and this changes the duty in direct proportion. if you could increase the frequency both the resiprocating compressor and the screw compressors give a better output but there are issuse with voltage torque and heat disapation
would you save energy Yes but very littel, the cost of installation alone would make it not worth doing, if you include the cost of the drive the life time of the pack I dont think you would get pay back even over 15 years

[Andy]

Hi gapatel
there would be some savings, but only on aplant that runs down to some very low loads in comparison to the 100% duty, typically below 25% (the smallest step available). I know of a plant with two 750kw piston compressors with variable speed drives, but this is to close control the water leaving temp on the chiller they are on. In this case energy is saved by having only evaporator pumps, and now service pumps or a buffer tank, hence the need for drives to close control and react to the load as it happens.
Kind Regards. Andy.

[Dan]

I regard your comments highly, Andy. I suppose, when we are talking about ammonia plants, that the loading is not nearly as variable as we would see in smaller warehouses, supermarkets, etc.

But I have to think there are opportunities to design an ammonia plant to take significant advantage of variable speed drives, on lead compressors, condenser fans and perhaps even evaporators. I am not thinking in great detail, here, but more in general. If something saves energy here, then why can't it save energy there, sort of thinking.

Crazy idea, but a unitized compressor/condenser/evaporator penthouse arrangement that varies all speeds as a unit? Surely some engineers are thinking about things like this. Not a job for a field person, perhaps, but worthy of design people.

There is "off" and "on." The most energy-efficient position is for something to be "off." The best method of refrigeration is for something to be "on."

Somewhere in-between is the answer, and changing speeds appears to me to be a necessary part of an improvement. It makes for fun thinking.

[Peter Mitchell]

I agree there is room for savings, but not using resirps

[Andy]

Hi Peter
industry favours Screws; but recips are more effecient, especially at anything other than full load. You could fit a screw with an inverter drive and nearly match the effeciency of a screw, but not quite. Recips are cheaper to install, cheaper run where power consumption is concerned, slightly more expensive to mantain. Another plus is if you neglect to service they let you know, usual letting go with a piston or so, without doing any great damage. Forget to maintain a screw and they right themselves off.
Screws are the flavour of the month, but recips will make a come back.
Me I specialise in Screw compressor work, it pays me well, but in general recips have still a lot going for them.
Kind Regards, Andy.

[star882]

When my friend Tiffany helped build a quad Britney simulation server for science, she actually used a digital control board to run the 1HP recriprocating compressor at something less than full capacity most of the time. It used large MOSFETs to vary the frequency of the power to the compressor. The actual control electronics included a lot of 74HC discrete logic along with a 40 pin Jessica Simpson micro (which also controlled the front panel LCD and the system power supply). I'm not too sure about the details but it interfaced to the motherboard through a RS232 connection. Using the right program, it's possible to display the approximate compressor RPMs and estimated high and low side pressures based on temperature. There were 4 TXVs, one for each evaporator, to ensure proper operation even with uneven loading. The main purpose is to keep liquid refrigerant out of the compressor.
One of the tricky things, though, is that the compressor must run at some minimum RPM or the oil bearings will not work properly (it relies on oil between the surfaces, and at low RPMs, there's not enough pressure to keep the oil in the bearings). I remember that a minimum RPM is hard coded into the firmware. (2900RPM?) But I suppose that running the oil pump on a separate motor that always runs at full speed will solve this. It would also reduce the stress on the parts from cold boots since the parts will be lubercated before they start moving.

BTW, one of the coolest things about that server is that it's possible to actually hear the compressor speed up at high load. It reminds me of a race car speeding back up to racing speeds after a caution.

[nh3wizard]

We have had great success with VFD's on our condensers and larger size evaporators, but have never ventured into putting them on recips, have thought about putting them on screws.

[refteach]

We have put some on recips, but the one thing you have to watch out for is the minimum RPM that the compressor can run at. Most ammonia recips have internal oil pumps, they also have a minimum GPM of oil supply needed to keep bearings happy. So there is a limit to the minimum RPM on ours it is between 400 and 600 depending on the amount of cylinders. If you venture below that you can install an external oil pump.

[señorfrio]

which its the capacity of your compresor