Using VFDs with Refrigeration Compressors.

[itsmoked]

I'm curious if anyone has any experience with this.

If you just approached a robust system and installed a VFD to run it what would be things to worry about?

One reason would be if the compressor possessed an oil pump for the crank journals. How much could one expect to reduce the speed before one might start expecting an oil pressure issue?

Insulation/bearing issues? (40 feet between the VFD and the compressor)

Is there an issue with the rotation speed being too low to allow the pistons to eventually get across top-dead-center?

What else?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

The units I have seen have positive displacement gear pumps. There is a sliding valve plate that directs the flow in the correct direction for either rotation direction. That is why compressors may be connected initially for either rotation, but it is advised to not change the direction of rotation after a time in service. After several years of service the valve plate may no longer move freely and may not slide to the correct position.
The first issue that I see is lack of enough inertia to carry the crank past TDC.
Suggestion; Use an undersized compressor and run it near double speed. That will allow you to safely turn down to about 50%.
Any comments concerning efficiency at increased speeds?
If you can come up with a combination of sizing and speed that lets the compressor run 24/7 efficiently you may reap savings from reduced peak demand charges.
Some years ago I saw an "Economizer" marketed to supermarkets. It was basically an inline restricter that reduced to capacity somewhat.
The savings came from longer running at reduced loading and as a result, reduced demand charges.
No fixed price. They asked for a percentage of the savings on the power bill.
I had an idea that I never pursued; There is a correlation between peak sunlight hours and peak A/C loading.
My idea was to use a VFD to drive an A/C at rated speed with the output of a solar array fed to the DC bus.
Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

Hi Bill!

Interestingly this application has no power costs at all so there's not much interest in that regard.

The problem I'm trying to work on is that the cooling load varies wildly. It's A/C but the load is so high that it's full-on refrigeration. It is based on two 10hp compressors that are staged but run their own halves of each three evaporators. The compressors have head unloaders but only for 2 of the six cylinders. We're warned not to unload more than two cylinders or we'll start pumping oil.

The issue is that as you start to transition between two comps and one or worse, when you're near to shutdown an unloaded single compressor can pump-down and cycle-off, only to still need to come back on in just a couple of minutes. There it runs for 5 minutes and shuts down again and this energy transfer range can last for a couple of hours.

If I could modulate the capacity a little by slowing the comp down then combined with the unloader I'd probably have enough capacity range to eliminate this frequent quasi-short cycling.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Hi Keith. First a link to a refrigeration forum may be in order.
There may be a simple mechanical fix.
My feeling is add an oil separater and unload more cylinders.
Unlike air compressors many refrigeration compressors (but not all) are able to start loaded.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[djs]

I know that some modern residential A/C units use a VFD for the compressor to get the SEER rating up.

 

[MintJulep]

Oil pump performance, as you already note.

Reduction in velocity in the piping system and lack of return of oil to the compressor.

Compressor motor cooling (assuming hermetic compressor).

Bearing damage / bearing currents.

Your compressor manufacturer may have some existing guidance and restrictions for VFD operations.

 

[ME27272727]

All of the above, and make sure the motor is rated inverter duty.

 

[zlatkodo]

I can't see what would be a benefit of using VFD to supply motor of refrigerant compressor?

 

[wayne440]

This

...to eliminate this frequent quasi-short cycling...

will be one benefit if the idea works.

 

[itsmoked]

MintJulep and ME27272727 pretty much ruined the party...

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

How about a restriction on the suction?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[wayne440]

How critical is it to maintain an exact temperature? If you can increase the dead band a bit you might reduce the short cycling issue.

 

[zlatkodo]

Most of motors of refrigerant compressors are required to start in very specific manner, either with a part of winding or with increased differential leakage or with simple YD starter, depending on manufacturer's specification and any change is not recommended to be done.

Winding Design and Repair

 

[itsmoked]

Bill; Suction restrictions are not real helpful as it causes the suction pressure to drop faster yet which causes an off cycle which rebounds quickly after the compressor shuts off.

Good point Wayne. I'll have to look into that.

zlatkodo; These compressors are just dumped across-the-line. But you're right, the makers will blame any issues on not doing everything the same old way as from eons ago.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

My thought was a moderate restriction. You don't want the crankcase pressure going negative. Less crankcase pressure means less volume per piston stroke. The pressure switch should be connected ahead of the restriction.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[MikeHalloran]

Around 1972, Ford vehicle A/C systems got a 'suction throttling valve' that allowed/forced the compressor to run nearly all of the time. I'm not sure if the valve was pressure or temperature operated, but it was definitely not electrical, and it was definitely not reliable.

The same compressor ran fine, cycling its electric clutch, using a transistor to sense the suction line temperature, and a low-pass filter to prevent rapid cycling (which started at ~10 Hz, before I put the filter in to bring it down to ~0.1 Hz. I don't remember removing or modifying the STV, so it must have failed in a position that allowed cycling the compressor clutch to work.

I think the STV was intended to improve fuel mileage, but the consumption was still egregious using a cycling clutch (which still had to be present to disconnect the compressor drive under some circumstances.
I don't think they use STVs any more.





Mike Halloran
Pembroke Pines, FL, USA