Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Air conditioner compressor capacity control in VRF systems

  1. Jan 14, 2016 #1
    hi guys.... am working on a variable refrigerant flow (VRF) type air conditioner.
    my aim is to change the capacity of the compressor based on load change. i can vary compressor capacity by using inverter....now my doubt is on what basis i should change compressor speed for muti evaporator system....i have one method. it is varying compressor speed based on suction superheat....
    assume that am using single compressor with 1 evaporator system (single evaporator)....now first thing i will do is 1)sense suction pressure and convert that to corresponding saturated suction temperature (Ts) 2)next i will sense evaporator temperature(Te) 3) am finding superheat by SH=Ts-Te 4)i will be comparing SH with target superheat SHS...if SH>SHS i will increase compressor speed else i will decrease compressor speed........NOW CONSIDER SINGLE COMPRESSOR WITH 3 EVAPORATOR SYSTEM...now i will have three superheat value..BIGGEST DOUBT IS SHOULD I ADD SH1+SH2+SH3 and then compare with target value to change speed of compressor??
     
  2. jcsd
  3. Jan 19, 2016 #2
    Thanks for the post! This is an automated courtesy bump. Sorry you aren't generating responses at the moment. Do you have any further information, come to any new conclusions or is it possible to reword the post?
     
  4. Jan 21, 2016 #3
    I don't have a whole lot of refrigeration experience, but you're going about controlling compressor output the same way I would...

    If I understand everything correctly, you're looking to make sure each evaporator always has liquid refrigerant available to it, and let the flow be controlled by evaporator valves, one on each evaporator.. right?

    Could you sense the post-condenser side for pressure and temperature to make sure you always have liquid there?.. full compressor speed at the point where you could have any bubbles of gas, and lets say minimum speed at 20* or 40 PSI above this? Perhaps sensing the accumulator fluid level would be another way?

    Just throwing some thoughts out!
     
  5. Jan 21, 2016 #4
    I see a problem. What if SH1=1, SH2=1, SH3=20 with 1 and 2 having 90% of the cooling load. You would calculate total superheat of 22 degrees, yet you are close to putting liquid into the compressor!
     
  6. Jan 21, 2016 #5
    I guess you'd have to go with the highest superheat value in that case?
     
  7. Jan 21, 2016 #6
    Seems to me you'd need the temperature of the mixed vapor from the three evaporators and calculate the superheat from that.
     
  8. Jan 21, 2016 #7
    I'm wondering how that would work... If all 3 cooling units are near a similar temperature I think that could work well, but I think if 1 cooling unit is a freezer and 2 are refrigerators, the outlet gas of the two refrigeration units could flow back and condense in the freezer unit's evaporator.. this could cause problems when the freezer unit's valve opens up and dumps more liquid in and you get it in the compressor in a real hurry.

    As a safety, you may need to have a gas/liquid separator before the compressor.

    the OP will have some things to ponder when he checks back anyhow :)
     
  9. Jan 22, 2016 #8
    I'd think all three evaporators would be running at the same pressure (common discharge line to compressor) which would obviate any back-flow.
     
  10. Jan 22, 2016 #9
    Pressure, yes, it would have to be, but if one is much colder than the others, wouldn't it condense what evaporated in the others? Pressures would be the same, thus the boiling point of the refrigerant, but one unit could be below that temperature.
     
  11. Jan 22, 2016 #10
    But, if you're controlling total superheat at, say, 10oF, wouldn't the cold and warm streams mix to the final superheated condition with no condensate?
     
  12. Jan 22, 2016 #11
    I don't really know... . I like solving things by extremes...

    Lets say you have 3 flasks and they're connected and evacuated.. (no atmospheric air in them), One has water that you're boiling in it, the other is at room temperature, and the other is in an ice bath... What is going to happen?

    Well, the pressure in all 3 flasks must be the same, but what is it? It's the vapor pressure above the water in the coldest flask... Since there's no atmospheric air, the temperatures in all 3 (at equilibrium) must be the same, but to get there is a different story... The flask you're heating will create steam that will travel and condense in the coldest flask until it has no water left in it.. now your cold flask is full.

    What happens if you add water to the already full cold flask? As in the evaporator valve on the coldest evaporator opens because that unit is above set temperature?
    I think you'll get a sudden rush of liquid into your compressor.

    I think one of the keys here is to have a large enough liquid accumulator to be able to keep the entire evaporator side significantly below the condensing temperature of the coldest one, perhaps then you can just sense the low side pressure and maintain it at whatever setpoint you like, and you cannot get any buildup of liquid.

    Just thoughts I get to by my own logic... Can't say I'm right, but it makes sense to me
     
  13. Jan 22, 2016 #12
    Your experiment is very different from a flowing refrigeration system where all vapor streams are continually carried away to a compressor inlet.
     
  14. Jan 22, 2016 #13
    how about check valves to prevent any condensation in a colder unit?

    I think you are right in post 6 that the vapors will need to be mixed and then the superheat calculated from that.
     
  15. Jan 23, 2016 #14
    Check out Fujitsu Mini-Split heat pumps. They have up to 8 inside coils running off one outdoor variable-speed compressor unit. They've obviously solved all the problems, at least for HVAC-type applications where inside coil temperatures don't vary by a whole lot.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Air conditioner compressor capacity control in VRF systems
  1. Air Compressors (Replies: 2)

  2. Unsteady Air Conditioner (Replies: 19)

Loading...