Results 1 to 48 of 48
  1. #1
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Desuperheating by liquid injection



    Hi Refrigeration Engineers,

    Is it a correct statement to say that: compressor discharge desuperheating by means of liquid injection into the discharge line between the compressor and the condenser actually does not remove any heat but rather spreads the heat already there through a larger volume of refrigerant and that volume of refrigerant is now closer to the saturation temperature.

    In other words, discharge desuperheating by liquid injection it is not heat removal but rather heat dilution ?



  2. #2
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    Yes. You are right.

  3. #3
    TRASH101's Avatar
    TRASH101 is offline Veteran Poster I am starting to push the Mods: of RE
    Join Date
    Mar 2007
    Location
    england
    Posts
    397
    Rep Power
    11

    Re: Desuperheating by liquid injection

    Yup heat dilution I suppose, as the enthalpy for the mass flow will be lower than without it, but why do we do it, when in reality the efficiency of the system is less than it could be?

    Just thinking out loud ........ not had much to do with desuperheating hot gas
    Quidvis Recte Factum Quamvis Humile Praeclarum.

    Latine dictum, sit altum videtur.

    Si hoc comprehendere potes, gratias age magistro Latinae.

  4. #4
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Given that desuperheating the discharge gas is heat dilution, how much dilution would be the upper limit of liquid injection before any further benefit is negated ?

    Enough liquid to reach the saturation point and begin condensation ? or could it be pushed further ?

  5. #5
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by TRASH101 View Post
    Yup heat dilution I suppose, as the enthalpy for the mass flow will be lower than without it, but why do we do it, when in reality the efficiency of the system is less than it could be?

    Just thinking out loud ........ not had much to do with desuperheating hot gas
    Gonna jump on the train, this is interesting, and this post intrigues me

    Your post sparked a discussion between me and a co-worker.

    The basis was LRI on a screw compressor, liquid refrigerant injection into the discharge.

    Two points:

    1. LRI into the discharge port itself requires a HP liquid pump to feed the TEAT valve with liquid at a high enough pressure to overcome the discharge pressure of the compressor.

    2. using only the condenser pressure to feed the TEAT means you have to inject "liquid" refrigerant earlier in the compression cycle of the compressor.

    Now... the reason I put it this way, is because you say that the use of LRI is reducing the capacity of the system.

    Me, I say, why?

    We mostly use LRI on Howden XRV compressors, on RSW systems.
    We take the liquid from the pilot receiver, and since we don't have a HP liquid pump, we have to feed the LRI into the economizer port of the compressor, instead of using the designated LRI port that is at the end of the compression. For me, in my mind, this serves as both cooling of the discharge gas/oil, and as a "superfeed/economizer", since the liquid that is evaporated across the TEAT valve also acts as more refrigerant introduced into the compression at the economizer port.

    I have noticed that we need the condenser pressure to be at 19C before the LRI starts working, so in a system working at -6.5C and +30C I see that as a bonus.

    ----------------------------------
    In a system with a HP pump feeding liquid to the TEAT valve to inject directly in the discharge, or at the end of the compression cycle I see it a loss...


    All I can say really is "discuss"
    -Cheers-

    Tycho

  6. #6
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Tyco,
    Are you trying to say that liquid injection into the rotors of a screw compressor, economizer port is good & does not reduce efficiency?

  7. #7
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    http://torbolab.tamu.edu/proc/turbop...8/Vol28015.pdf.

    Page 153 gives information of negative effects on liquid injection into rotors.

    Higher mass flow, more power
    Liquid leakage past rotor tips back to suction
    when compressor unloaded
    Last edited by RANGER1; 15-03-2013 at 10:28 PM.

  8. #8
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    Tyco,
    Are you trying to say that liquid injection into the rotors of a screw compressor, economizer port is good & does not reduce efficiency?
    I'm not saying anything, I was hoping to start an argument

    Right now, I'm arguing for the point that liquid refrigerant that has been passed through a TEAT valve and has been flashed to gas state that is injected at the same place in the compression that gas from an economizer how are they different from each other?

    sure, in an economizer, the refrigerant is already in a gaseous state when it leaves the economizer tank, where the liquid the gas is created from helps cool down the liquid that is about to enter the liquid separator, so that's a win

    But what I am saying is... You take liquid from the condenser, and you try to introduce it in the compressor, to cool down the discharge and thereby the oil... if you introduce it to late in the compression cycle, the pressure in the compressor is higher than the condensation pressure, and thus, refrigerant from the condenser can not enter the compressor, and then can not be part of "dissipating" the heat over a larger amount of refrigerant, because the pressure at the discharge port of the compressor is higher than the condensing pressure.

    However, if you connect the LRI in the middle of the compression cycle, at the economizer port, you have a pressure in the compression cycle, lower than the condensing pressure, and by introducing more refrigerant at this stage, you not only have more refrigerant to dissipate the heat, but the compressor also has more gas to compress, similar to an economizer, but without the need for a huge tank.

    This is all sparked to make a discussion...

    Because, picture you have a economizer tank, and you cool down the liquid going from the condenser into the LP liquid separator, by passing it through a closed economizer, and the gas you boil off on the other side, you introduce into the superfeed/economizer port... this is a gain, yay, we have sub cooled the liquid going into the LP liquid separator, and yay, we have created superfluous gas that we can introduce into the middle of the compression cycle of the compressor, giving us a gain...

    However, don't we at the same time need a larger oil cooler?

    Where is the gain?

    Discuss
    -Cheers-

    Tycho

  9. #9
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    First. Liquid pump for LPI into discharge side. It can be done and it can be helpful to reduce condensing pressure and a lot of energy can be saved.
    Second. Use economize port for the suction of vapor from economizer. This improve energy efficiency as well, because in this case vapor will be compressed from intermediate pressure to the condensing pressure instead of from suction pressure to the condensing pressure.
    Third. Liquid injection into economize port is not efficient, because cooling refrigerant will evaporate and additional compressor work should be done to compress this vapor to condensing pressure. Better solutions. Inject refrigerant in LRI port. Every compressor manufacturer located this ports at later stage of compression to minimize work of the compressor that required to compress vapor to the condensing pressure.
    Larger oil cooler. Usually, thermosyphon oil cooler have overfeed 3:1 and it will be able to handle load from superheated refrigerant.

  10. #10
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Great discussion guys,

    i agree with Segei, liquid injection into the economizer port would be inefficient as the liquid would vaporize and have to be immediately recompressed, adding further load to the compressor.

    But i am wondering if enough liquid is injected into the LRI port by means of a pump, could it be possible to lower the compressor discharge temp low enough to not only desuperheat the hot gas but to actually condense it ?? (Precondensing)

    This would supply the condenser with a flow of hot liquid instead of hot gas, flooding the condenser, wetting all of the condenser tubes with liquid refrigerant much the same as what takes place in a flooded evaporator.

    Flooded evaporators are more capable of transfering heat due to a larger internal surface area in contact with liquid, is this not the case for condensers too ??

    To think a little outside the square,
    Could it make the condenser into a kind of massive subcooler, with greater heat exchange ability than it previously had when trying to remove heat from gas ?

    By the use of a liquid pump set up a liquid cycle that draws liquid from the receiver and pump it to the condenser inlet and then the compressor discharge gas can be added to this liquid flow, use the liquid flow to "precondense" the discharge gas, remove the heat from liquid only in the "condenser", subcooled liquid flows to the receiver then to the rest of the system. Use a LPA pump to raise liquid to higher pressures as required for expansion devices.


    What are your thoughts ?

  11. #11
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    This is heat dilution. It means that portion of liquid will evaporate to eliminate superheat and gas temperature will be SDT and not lower. If you oversupply liquid, it can go to oil separator and this can be problem. Oversupplying liquid to discharge line will not have benefits because it will dilute the heat. Condenser removes the heat and condense refrigerant.

  12. #12
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Thankyou Segei,

    So only the amount of liquid actually vaporizing is beneficial, any extra only dilutes the heat through a larger volume of refrigerant at saturation temperature ?

    Heat spread through a larger volume of refrigerant entering the condenser may have a lower overall temp that may be closer to ambient therefore the lower temperature difference will decrease the condensers ability to transfer heat ??

    Thus negating any benefit of excess liquid injected ??

  13. #13
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    R717,
    The draw back with liquid in oil seperator is oil dilution.
    Usually discharge temp has to be 15 deg C above SCT to keep oil dry.

    I have seen a large system where economizer port is used for liquid injection. It was recommended by Howden for cooling oil & gas flow in compressor at earlier stage of compression.
    I guess a lot more liquid may boil off sooner but you still use more power.

    Injecting liquid directly into discharge line has to be more efficient capacity & power wise.

    Also to much liquid entering condenser inlet could also possibly cause liquid logging reducing its capacity.
    You need heat tanfer to take place in condenser.

  14. #14
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    Sometimes liquid injection in side port can be beneficial, because it can be done at low condensing pressure. Definitely more work should be done to compress evaporated liquid, but lower condensing pressure saves more energy. Every energy savings is trade in. If you spend 1 kWh of energy to save 2 kWh, this is step in the right direction. Dedicated LI port more efficient at high condensing pressures, but it can not be used at low condensing pressures.

  15. #15
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Segei View Post
    Dedicated LI port more efficient at high condensing pressures, but it can not be used at low condensing pressures.
    What prevents injection into the LI port at low condensing pressures ?

  16. #16
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by R717 View Post
    What prevents injection into the LI port at low condensing pressures ?

    I don't know why either as it can be used on booster compressors!

    The only thing I see would be if injection valve is to small with lower inlet pressure.
    Surely these days with all the Danfoss gear etc it could be versatile enough to handle all conditions.

  17. #17
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    I mentioned about high stage compressor with TEV. At lower condensing pressure, pressure diffidence between condensing pressure and port will be reduced and liquid refrigerant will be undersuppllied. EEV is one of the solutions to this issue.

  18. #18
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Given the parameters of the discussion so far, what do you make of this automotive A/C system i have been experimenting with ?
    F500 AC system 2.jpg

  19. #19
    Join Date
    Apr 2001
    Location
    The Netherlands
    Age
    57
    Posts
    51
    Rep Power
    17

    Re: Desuperheating by liquid injection

    Injecting liquid refrigerant in the discharge causes that both gas and oil are cooled down.

    Too much injection results in a mix of oil, gas and liquid whereas the oil separator (if installed) will separate the oil and liquid from the gas........
    >> liquid refrigerant goes with the oil back to your compressor: not good.

    If the injection is just right than you remain with a minimum superheat of the gas and relative cold oil.

    Effect on oil:
    Cold oil will be easyer to separate in the oil separator and cold oil means no more oil cooling required (e.g. for screw compressor).

    Effect on condenser:
    Superheated gas has a poor heat transfer to the pipe wall, but when the wall is wetted than it is much better.
    When the refrigerant has to cool down just little, liquid drops are formed after just a small part of the condenser has been passed and that has a huge effect on the heat transfer coefficient.
    So it could be that injecting refrigerant increases the condenser capacity and thus reduces the condenser pressure and energy consumption.

    I have heard about that being used in Ammonia recip systems where discharge temperature is often very high.
    I have heard about that being used with success on screws, although the mixing is not easy and the injection pump is a pain on the ass (cavitation).
    Must admit though; I have never seen / measured one.

    I don believe in liquid injection in the LI port of the screw.
    Quickly 6% enegy loss or more, only to safe an oil cooler.
    I believe that is mostly done in USA where energy has been cheaper than here in Europe.

  20. #20
    Join Date
    Apr 2001
    Location
    The Netherlands
    Age
    57
    Posts
    51
    Rep Power
    17

    Re: Desuperheating by liquid injection

    The above is typical for ammonia systems with e.g. 20mm steel condenser pipes.
    Other condensers have modified internal pipe (e.g. grooves) and that makes a difference.
    Injection could increase the condenser capacity.

  21. #21
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    http://torbolab.tamu.edu/proc/turbop...8/Vol28015.pdf.

    Page 153 gives information of negative effects on liquid injection into rotors.

    Higher mass flow, more power
    Liquid leakage past rotor tips back to suction
    when compressor unloaded
    The link to the pdf does not work, would very much like to read it
    -Cheers-

    Tycho

  22. #22
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Segei View Post
    First. Liquid pump for LPI into discharge side. It can be done and it can be helpful to reduce condensing pressure and a lot of energy can be saved.
    Second. Use economize port for the suction of vapor from economizer. This improve energy efficiency as well, because in this case vapor will be compressed from intermediate pressure to the condensing pressure instead of from suction pressure to the condensing pressure.
    Third. Liquid injection into economize port is not efficient, because cooling refrigerant will evaporate and additional compressor work should be done to compress this vapor to condensing pressure. Better solutions. Inject refrigerant in LRI port. Every compressor manufacturer located this ports at later stage of compression to minimize work of the compressor that required to compress vapor to the condensing pressure.
    Larger oil cooler. Usually, thermosyphon oil cooler have overfeed 3:1 and it will be able to handle load from superheated refrigerant.
    First: if you want LRi on the LRI port of the compressor, wich is at the end of the compression, you NEED TO HAVE A PUMP, to supply the refrigerant to the TEAT valve, because the pressure at the end of the compressor compression is higher than the condenser pressure if you have a "normal system with a PM valve between the compressor and the condenser... you can have a discharge pressure of 30C, and a condensing pressure of 20C on a system with a VFD controlled condenser pump...
    Second: I don't get this... if I am supplying "evaporated liquid" from the condenser as compared to gas from the economizer tank, what is the difference? there will be extra work on the compressor no matter what... So normally the economizer supplies the economizer port with gas at around -10C... There is extra load on the compressor in either case, how is the load from the LRI introduced in the economizer port different from the gas introduced to the LRI port?
    I have seen a howden 255 with an economizer fitted, where the economizer valve opened when the compressor was at 100% and it was injected with -10C gas where the compressor almost took a bow and called it a night... so we had to introduce the economizer at -25 and slowly raise it to -10.
    Third: The LRI port only works if you have a pump feeding refrigerant to the TEAT
    if you want an economically working LRI, you have to use the economizer port... because the condensing pressure is lower than the pressure at the compressors "LRI stage"
    -Cheers-

    Tycho

  23. #23
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by R717 View Post
    Great discussion guys,

    i agree with Segei, liquid injection into the economizer port would be inefficient as the liquid would vaporize and have to be immediately recompressed, adding further load to the compressor.

    But i am wondering if enough liquid is injected into the LRI port by means of a pump, could it be possible to lower the compressor discharge temp low enough to not only desuperheat the hot gas but to actually condense it ?? (Precondensing)

    This would supply the condenser with a flow of hot liquid instead of hot gas, flooding the condenser, wetting all of the condenser tubes with liquid refrigerant much the same as what takes place in a flooded evaporator.

    Flooded evaporators are more capable of transfering heat due to a larger internal surface area in contact with liquid, is this not the case for condensers too ??

    To think a little outside the square,
    Could it make the condenser into a kind of massive subcooler, with greater heat exchange ability than it previously had when trying to remove heat from gas ?

    By the use of a liquid pump set up a liquid cycle that draws liquid from the receiver and pump it to the condenser inlet and then the compressor discharge gas can be added to this liquid flow, use the liquid flow to "precondense" the discharge gas, remove the heat from liquid only in the "condenser", subcooled liquid flows to the receiver then to the rest of the system. Use a LPA pump to raise liquid to higher pressures as required for expansion devices.


    What are your thoughts ?
    To address one of the points you make here
    if enough liquid is injected into the LRI port by means of a pump, could it be possible to lower the compressor discharge temp low enough to not only desuperheat the hot gas but to actually condense it ??
    The discharge from a screw compressor goes into an oil separator, so if you already lower the temperature of the discharge gas to actually condense it, it would cause the oil in the oil separator to flush into the system.
    In a system with an LRI, we want to cool the discharge gas Juuust enough to, as you said in the beginning "dissipate the heat" in the discharge line, but not so much as to get to much refrigerant in the oil, as this will cause a large oil carryover into the system.
    -Cheers-

    Tycho

  24. #24
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    R717,
    The draw back with liquid in oil seperator is oil dilution.
    Usually discharge temp has to be 15 deg C above SCT to keep oil dry.

    I have seen a large system where economizer port is used for liquid injection. It was recommended by Howden for cooling oil & gas flow in compressor at earlier stage of compression.
    I guess a lot more liquid may boil off sooner but you still use more power.

    Injecting liquid directly into discharge line has to be more efficient capacity & power wise.

    Also to much liquid entering condenser inlet could also possibly cause liquid logging reducing its capacity.
    You need heat tanfer to take place in condenser.
    Injecting liquid directly into discharge line has to be more efficient capacity & power wise.
    Bot or this to happen, you need a HP liquid pump to feed the LRI? where is the win where is the loss? to inject liquid in the LRI port you need a pump to create the differential pressure to feed it, believe me, I have had to rebuild at least 10 LRI systems because the pipe was connected at the LRI port, where the pressure was to high for the condenser to feed it...
    -Cheers-

    Tycho

  25. #25
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by R717 View Post
    What prevents injection into the LI port at low condensing pressures ?
    In a normal system, where the compressor is separated from the condenser by a PM valve, and the condenser is cooled by a VFD pump, you may have 20C on the condenser side, but in a ammonia system you would like to keep 30C as the discharge on the compressor.

    So:30C discharge, 20C condencer... How are you going to get 20C liquid through a TEAT valve, and into a 30C discharge on a compressor?
    You don't... you find an earlier stage in the compression, where it is around 10-15C instead

    When I use Celsuis here, convert to to bar, I am used to using the "temperature" to relate to pressure
    -Cheers-

    Tycho

  26. #26
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    I don't know why either as it can be used on booster compressors!

    The only thing I see would be if injection valve is to small with lower inlet pressure.
    Surely these days with all the Danfoss gear etc it could be versatile enough to handle all conditions.
    Nope, danfoss may be good, but they are not god
    -Cheers-

    Tycho

  27. #27
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    To revert to the original discussion:

    Why is an LRI introduced into the economizer port less efficient than a pure economizer?

    1.Economizer: the gas introduced into the economizer port is created by subcooling the liquid going into the LP receiver, so the extra work the compressor does actually has a benefit.

    2.LRI in economizer port: the gas (depending on the distance from the TEAT to the compressor port) is taken from the condenser and causes the compressor to do more work, and again causes the condenser to do more work to create more liquid for the LRI and is an endless cycle of no gain

    3. LRI in LRI port with use of a HP liquid pump: the gas (depending on the distance from the TEAT to the compressor port) is taken from the condenser and causes the compressor to do more work, and again causes the condenser to do more work, and add the work of the pump...

    So wich one is beneficial for oil cooling?

    LRI or watercooled oil cooler?
    -Cheers-

    Tycho

  28. #28
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    R717 the subcooling shell & tube suction, which goes back to compressor suction is achieving no real gain.
    You are still loading up suction with extra load to achieve subcooling in liquid line.
    I would cut it out as well as LPA pump.
    Get you suction superheat right & no high discharge temps.
    Make sure you use propane or butane to keep it interesting!

  29. #29
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Tycho View Post
    The link to the pdf does not work, would very much like to read it
    Tyco,If you can give me your email address I will send no problem

  30. #30
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Tycho View Post
    First: if you want LRi on the LRI port of the compressor, wich is at the end of the compression, you NEED TO HAVE A PUMP, to supply the refrigerant to the TEAT valve, because the pressure at the end of the compressor compression is higher than the condenser pressure if you have a "normal system with a PM valve between the compressor and the condenser... you can have a discharge pressure of 30C, and a condensing pressure of 20C on a system with a VFD controlled condenser pump...
    Second: I don't get this... if I am supplying "evaporated liquid" from the condenser as compared to gas from the economizer tank, what is the difference? there will be extra work on the compressor no matter what... So normally the economizer supplies the economizer port with gas at around -10C... There is extra load on the compressor in either case, how is the load from the LRI introduced in the economizer port different from the gas introduced to the LRI port?
    I have seen a howden 255 with an economizer fitted, where the economizer valve opened when the compressor was at 100% and it was injected with -10C gas where the compressor almost took a bow and called it a night... so we had to introduce the economizer at -25 and slowly raise it to -10.
    Third: The LRI port only works if you have a pump feeding refrigerant to the TEAT
    if you want an economically working LRI, you have to use the economizer port... because the condensing pressure is lower than the pressure at the compressors "LRI stage"
    Tyco can you remind me of what LRi & LRI is?

    I feel a lot of your statements are a bit of a set up but here goes.

    Liquid refrigerant injection into liquid injection port does not need a pump, it goes inbecause pressure is lower.
    To the best of my knowledge economizer port is positioned to open to rotor thread almost immeidiately
    after male rotor closes of female rotor vane. At this point as rotor continues to rotate it starts to reduce volume in female rotor thread(compress).
    At this point economizer port is open to this vane in female rotor, which is slightly higher than suction pressure(asssuming compressor is over80% loaded).So this allows economizer suction to flow into rotor vane.
    After it passes that port, then liquid injection port is open to metered liquid refrigerant to control discharge/oil temp.
    So not sure why you need a pump.
    If wrong selection of Vi is selected maybe, but over compression is possibly the last thing you want(draws more power consumption.
    You also mention "normal system" having PM valve in discharge line raising pressure upstream of valve.
    Why?
    To me its another waste of power. Everything points to having as lower condensing pressure as possible.

    If you inject gas from economizer you are increasing nett refigerant effect byu subcooling liquid refrigerant feeding evaporator. Your compressor basically increase in capacity, using a bit more power though.

    If you inject liquid from condenser you get cooler discharge temp but thats all, not increased refrigeration effect.
    I however agree that injecting liquid for oil cooling & economizer will still work but possibly a little shy of using correct ports.
    Only Howden could tell us that!

  31. #31
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Tycho View Post
    Nope, danfoss may be good, but they are not god
    I would Danfoss ICM will do the job perfectly due to its turndown.

  32. #32
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Tycho View Post
    To revert to the original discussion:

    Why is an LRI introduced into the economizer port less efficient than a pure economizer?

    1.Economizer: the gas introduced into the economizer port is created by subcooling the liquid going into the LP receiver, so the extra work the compressor does actually has a benefit.

    2.LRI in economizer port: the gas (depending on the distance from the TEAT to the compressor port) is taken from the condenser and causes the compressor to do more work, and again causes the condenser to do more work to create more liquid for the LRI and is an endless cycle of no gain

    3. LRI in LRI port with use of a HP liquid pump: the gas (depending on the distance from the TEAT to the compressor port) is taken from the condenser and causes the compressor to do more work, and again causes the condenser to do more work, and add the work of the pump...

    So wich one is beneficial for oil cooling?

    LRI or watercooled oil cooler?

    Tyco, they all are but water cooled oil cooler or thermosiphon re best hands down!
    Note use of economizer will slightly assist cooling oil a bit more through compression cycle.

    If you put scenario's through your Howden programme you may see differences in efficiency,COP, volumetric efficiency.
    Last edited by RANGER1; 23-03-2013 at 12:05 AM.

  33. #33
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Tyco an extract from Frick article, turn computer upside down to readscan0001.jpg

  34. #34
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    R717 the subcooling shell & tube suction, which goes back to compressor suction is achieving no real gain.
    You are still loading up suction with extra load to achieve subcooling in liquid line.
    I would cut it out as well as LPA pump.
    Get you suction superheat right & no high discharge temps.
    Make sure you use propane or butane to keep it interesting!
    Ranger 1 i appreciate your post,

    to clarify this is a old R12 system running on R134a,
    suction superheat i feel is a bit high at 10 c, i would rather 5 c or maybe less but i have a evaporator design fault causing uneven distribution, this causes part of the evap to flood and part to have high superheat,
    if i drop the overall superheat too much i get liquid going back to the compressor. I might be able to get 1 or 2 degrees less, ill try it.

    Do you feel 67.5 c discharge temp before injection for R134a at 210 psig is too high ?

    I agree about the shell and tube, i would have preferred a suction line / liquid line heat exchanger but was impossible to pipe up. (The suction line is a rubber hose.) It does provide some usefull subcooling when the vehicle becomes stationary and condenser air flow is reduced to almost 0 (fan only air flow) this is needed to cool the LPA pump's windings.

    The LPA pump was used to lower the temp and volume of compressor discharge gas to in turn help raise the compressors volumetric efficiency. Also pumping liquid into essentially the condenser inlet in a attemp to "wet" all of the condenser tubes as superheated discharge gas transfers it's heat poorly to the tubes. Providing high pressure liquid to the TXV was a secondary issue but it is interesting to watch the vent air temp lower as the LPA pump speed is increased.

    There are no oil separators in this system so oil dilution in a separator is not a issue and oil is free to flow through out the system, this is why i sought to understand the upper limit of liquid injection as as it is now discharge temp is dropped to 50 c @ 210 psig after injection and looking on a PT chart for R134a that is not just desuperheated but below the saturation temp, in other words condensing.

    A lot of this seems like a dog chasing it's tail but isn't it normal in a refrigeration system to give up some here to get some more over there, system capacity has noticeably increased with these measures.

    But you must keep in mind this is all a bit of a experiment to use some out of the ordinary means to address automotive A/C system capacity issues and it's a bit of a mental challenge.

    Or some might say it's just a bit mental




    Please keep the interesting posts coming !

  35. #35
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by R717 View Post
    Ranger 1 i appreciate your post,

    to clarify this is a old R12 system running on R134a,
    suction superheat i feel is a bit high at 10 c, i would rather 5 c or maybe less but i have a evaporator design fault causing uneven distribution, this causes part of the evap to flood and part to have high superheat,
    if i drop the overall superheat too much i get liquid going back to the compressor. I might be able to get 1 or 2 degrees less, ill try it.

    Do you feel 67.5 c discharge temp before injection for R134a at 210 psig is too high ?

    I agree about the shell and tube, i would have preferred a suction line / liquid line heat exchanger but was impossible to pipe up. (The suction line is a rubber hose.) It does provide some usefull subcooling when the vehicle becomes stationary and condenser air flow is reduced to almost 0 (fan only air flow) this is needed to cool the LPA pump's windings.

    The LPA pump was used to lower the temp and volume of compressor discharge gas to in turn help raise the compressors volumetric efficiency. Also pumping liquid into essentially the condenser inlet in a attemp to "wet" all of the condenser tubes as superheated discharge gas transfers it's heat poorly to the tubes. Providing high pressure liquid to the TXV was a secondary issue but it is interesting to watch the vent air temp lower as the LPA pump speed is increased.

    There are no oil separators in this system so oil dilution in a separator is not a issue and oil is free to flow through out the system, this is why i sought to understand the upper limit of liquid injection as as it is now discharge temp is dropped to 50 c @ 210 psig after injection and looking on a PT chart for R134a that is not just desuperheated but below the saturation temp, in other words condensing.

    A lot of this seems like a dog chasing it's tail but isn't it normal in a refrigeration system to give up some here to get some more over there, system capacity has noticeably increased with these measures.

    But you must keep in mind this is all a bit of a experiment to use some out of the ordinary means to address automotive A/C system capacity issues and it's a bit of a mental challenge.

    Or some might say it's just a bit mental




    Please keep the interesting posts coming !
    R717 Discharge temp well within limits
    LPA pump usually only to boost liquid line pressure to prevent flash gas in long liquid lines. Liquid injection usually in suction line of compressor on low temp semi-hermetic application, to keep motor cool Liquid injected into discharge line to give better condenser performance is a new one in general terms (why isn't everyone doing it)?
    Last edited by RANGER1; 23-03-2013 at 10:28 AM.

  36. #36
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    Tyco can you remind me of what LRi & LRI is?

    I feel a lot of your statements are a bit of a set up but here goes.

    Liquid refrigerant injection into liquid injection port does not need a pump, it goes inbecause pressure is lower.
    To the best of my knowledge economizer port is positioned to open to rotor thread almost immeidiately
    after male rotor closes of female rotor vane. At this point as rotor continues to rotate it starts to reduce volume in female rotor thread(compress).
    At this point economizer port is open to this vane in female rotor, which is slightly higher than suction pressure(asssuming compressor is over80% loaded).So this allows economizer suction to flow into rotor vane.
    After it passes that port, then liquid injection port is open to metered liquid refrigerant to control discharge/oil temp.
    So not sure why you need a pump.
    If wrong selection of Vi is selected maybe, but over compression is possibly the last thing you want(draws more power consumption.
    You also mention "normal system" having PM valve in discharge line raising pressure upstream of valve.
    Why?
    To me its another waste of power. Everything points to having as lower condensing pressure as possible.

    If you inject gas from economizer you are increasing nett refigerant effect byu subcooling liquid refrigerant feeding evaporator. Your compressor basically increase in capacity, using a bit more power though.

    If you inject liquid from condenser you get cooler discharge temp but thats all, not increased refrigeration effect.
    I however agree that injecting liquid for oil cooling & economizer will still work but possibly a little shy of using correct ports.
    Only Howden could tell us that!
    Finally, there it is

    With LRI from the condenser, you take liquid directly from the condenser without any sub cooling effect.
    You add load to the compressor and condenser without any gain other than cooling the discharge and oil in the compressor system

    The PM valve is between the compressor and condenser to keep a stable discharge pressure of 30C and prevent oil carryover, so with 30C in the discharge you may end up with 10-20C in the condenser, depending on seawater temp.

    When I talk about needing a pump to feed it into the LRI port, I see that I have forgotten to mention that it is a cooling system, not a freezing system. suction of -8C.
    I am speaking from experience, that when the system starts with a suction pressure of around 5-6C you will need a condenser pressure of minimum 17-19C before it starts to feed into the economizer port.

    on the Howden XRV's we use, I have never been able to feed the LRI into the LRI port, the LRI port is all the way at the end of the compression, right before the rotors opens to the discharge.

    Was fun playing a little devils advocate here
    -Cheers-

    Tycho

  37. #37
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    R717 Discharge temp well within limits
    LPA pump usually only to boost liquid line pressure to prevent flash gas in long liquid lines. Liquid injection usually in suction line of compressor on low temp semi-hermetic application, to keep motor cool Liquid injected into discharge line to give better condenser performance is a new one in general terms (why isn't everyone doing it)?

    (why isn't everyone doing it)?

    Many systems use different means to keep condenser temperature/pressure artificially high by switching fans on/off, fan speed control, liquid control to back up liquid from the condenser outlet reducing condensing space, all in a effort to keep liquid pressure high to move it through long pipe runs and have adequate pressure drop at expansion devices and eliminate flash gas. I imagine a means to increase compressor volumetric efficiency that lowers condenser temp/pressure would be seen by designers as an anathema.


    Also from what i have seen LPA and it's pumping technology is not held in very high regard. I think from what i have read that it is essentially because there have been many ill conceived systems that have been misunderstood and problematic.

    The LPA pump of choice from what i have seen is the centrifugal pump. I think this is a bad choice and has given LPA a bad name.

    The problem with centrifugal pumps is cavitation, because they raise pressure by violent acceleration in the impellor followed by deceleration in the pump's volute. That acceleration can cavitate most liquids if the pump is not set up correctly, a liquid as tenuous as refrigerant (which is chosen for it's ability to vaporize) doesn't stand a chance.

    I think also refrigeration design has been living in a compressor centric universe, where the compressor is looked upon as the source of all energy and movement within the refrigeration system. Some other device moving refrigerant around and doing usefull work is seen as strange. Pumps have generally been used for mundane jobs such as recirculating refrigerant through flooded evaporators but generally not in the main refrigeration cycle.

    It needs a shift in thinking.

  38. #38
    Join Date
    Dec 2007
    Location
    australia
    Posts
    2,461
    Rep Power
    27

    Re: Desuperheating by liquid injection

    Quote Originally Posted by R717 View Post
    (why isn't everyone doing it)?

    Many systems use different means to keep condenser temperature/pressure artificially high by switching fans on/off, fan speed control, liquid control to back up liquid from the condenser outlet reducing condensing space, all in a effort to keep liquid pressure high to move it through long pipe runs and have adequate pressure drop at expansion devices and eliminate flash gas. I imagine a means to increase compressor volumetric efficiency that lowers condenser temp/pressure would be seen by designers as an anathema.


    Also from what i have seen LPA and it's pumping technology is not held in very high regard. I think from what i have read that it is essentially because there have been many ill conceived systems that have been misunderstood and problematic.

    The LPA pump of choice from what i have seen is the centrifugal pump. I think this is a bad choice and has given LPA a bad name.

    The problem with centrifugal pumps is cavitation, because they raise pressure by violent acceleration in the impellor followed by deceleration in the pump's volute. That acceleration can cavitate most liquids if the pump is not set up correctly, a liquid as tenuous as refrigerant (which is chosen for it's ability to vaporize) doesn't stand a chance.

    I think also refrigeration design has been living in a compressor centric universe, where the compressor is looked upon as the source of all energy and movement within the refrigeration system. Some other device moving refrigerant around and doing usefull work is seen as strange. Pumps have generally been used for mundane jobs such as recirculating refrigerant through flooded evaporators but generally not in the main refrigeration cycle.

    It needs a shift in thinking.
    R717 If your system works that's great, but you are treating symptoms not causes.
    Last edited by RANGER1; 24-03-2013 at 10:20 AM.

  39. #39
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    Quote Originally Posted by RANGER1 View Post
    R717 If your system works that's great, but you are treating symptoms not causes.
    Ranger 1 i feel you have missed the point of it a little, it's a experiment not to see what should be done but rather what could be done.

    Many thanks for your very interesting input.

  40. #40
    Join Date
    Jan 2013
    Location
    Australia
    Posts
    35
    Rep Power
    0

    Re: Desuperheating by liquid injection

    A big thankyou to all participants for a most interesting discussion.

  41. #41
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    Tycho.
    I think that you have one misconception. You assumed that plant which has discharge pressure regulating valve between compressor and condenser is normal. This is not normal. Actually this is wasting of energy to have condensing temperature 20C and discharge 30C. Definitely, oil carry over can be issue, but increased size of oil separator will solve the issue.
    Look at operation of normal refrigeration plant that have condensing pressure close to discharge pressure. Assume that suction pressure is 0 bars, economizer port 2 bars, LRI port 5 bars and discharge 10 bars. Condensing pressure slightly lower than discharge due to pressure drop in discharge line. LI in economize port should be compressed from 2 to 10 bars. LI in LRI port should be compressed from 5 to 10 bars. LRI injection is more efficient. However, water cooling of oil is the most efficient, because no additional work should be done by compressor. Thermosyphon cooling don't require compressor work as well just small additional load to the condensers. It is better than LRI as well.
    Another issue with LRI port is that at lower condensing pressure liquid injection will be undersupplied. If condensing pressure will drop to 8 bars liquid supply pressure difference will be 8-5=3 bars instead of 10-5=5 bars that was before.
    Typical refrigerant vapor injected in economized port was created by useful work of cooling. It means by subcooling liquid refrigerant or by high temperature refrigeration load. LI don't do any useful cooling. I mean cooling effect for refrigeration plant. Oil can be cooled by water or TS without work done by compressor.

  42. #42
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Segei View Post
    Tycho.
    I think that you have one misconception. You assumed that plant which has discharge pressure regulating valve between compressor and condenser is normal. This is not normal. Actually this is wasting of energy to have condensing temperature 20C and discharge 30C. Definitely, oil carry over can be issue, but increased size of oil separator will solve the issue.
    Look at operation of normal refrigeration plant that have condensing pressure close to discharge pressure. Assume that suction pressure is 0 bars, economizer port 2 bars, LRI port 5 bars and discharge 10 bars. Condensing pressure slightly lower than discharge due to pressure drop in discharge line. LI in economize port should be compressed from 2 to 10 bars. LI in LRI port should be compressed from 5 to 10 bars. LRI injection is more efficient. However, water cooling of oil is the most efficient, because no additional work should be done by compressor. Thermosyphon cooling don't require compressor work as well just small additional load to the condensers. It is better than LRI as well.
    Another issue with LRI port is that at lower condensing pressure liquid injection will be undersupplied. If condensing pressure will drop to 8 bars liquid supply pressure difference will be 8-5=3 bars instead of 10-5=5 bars that was before.
    Typical refrigerant vapor injected in economized port was created by useful work of cooling. It means by subcooling liquid refrigerant or by high temperature refrigeration load. LI don't do any useful cooling. I mean cooling effect for refrigeration plant. Oil can be cooled by water or TS without work done by compressor.
    With seawater temp at 4-15 degrees, it's necessary with a PM valve to keep the discharge pressure up, even with VFD controlled condenser pumps.

    Maybe a PM valve is a waste to energy, but it save you lot's of headaches... in case of a sudden shutdown of the compressor when it's at full capacity and full speed, you don't loose 3/4 of the oil in the oil separator.
    The oil separator is more than large enough, but it won't stop the oil from going out into the system if it get's a sudden pressuredrop.

    I agree that an oil cooler is the best solution, but with the pressure on prices, it's cheaper to install an LRI than to install an oil cooler, a 3 way valve and an extra sea water pump (need an extra pump because we run VFD on the condenser pump)
    -Cheers-

    Tycho

  43. #43
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    In North America typical ammonia refrigeration plant has minimum condensing temperature is 20C or 110 -115 psig. This is just typical plant operation. I know a few plants that have condensing temperature of 10C. Suction pressure of these plants around 0 bars. For higher suction pressure we need oversized oil separator. Cost of the compressor with this oil separator will increase by 5%. I talked to several compressor manufacturers and everybody claim that their compressors can operate at condensing pressure 15C even with 2 bars suction pressure, but oil separator should have right size.

  44. #44
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Segei View Post
    In North America typical ammonia refrigeration plant has minimum condensing temperature is 20C or 110 -115 psig. This is just typical plant operation. I know a few plants that have condensing temperature of 10C. Suction pressure of these plants around 0 bars. For higher suction pressure we need oversized oil separator. Cost of the compressor with this oil separator will increase by 5%. I talked to several compressor manufacturers and everybody claim that their compressors can operate at condensing pressure 15C even with 2 bars suction pressure, but oil separator should have right size.
    On freezing plants with -0.3 bar in suction, we still run with a PM valve on the discharge to keep the discharge at 30C, if we do this, we can use the same oil separators on -0.3 bar suction, or at 3.5 bar suction to prevent any accidental oil loss in case of a "alarm stop" on the compressor... we have a "fineoil separator" incorporated in our design, so we can run on any pressures you say and still get 99% of the oil loss back into the compressor suction... we design our freezing systems with two PM valves, so when you need a defrost, we shut of all hotgas to the condenser because the condenser PM is on 0.1 bar higher pressure. and lead it all to the plate freezer or evaporator that needs the hotgas... sure we are like the most expensive air--air-split unit
    but after awhile, we will give you more back for your buck...


    Of course you could add a full speed condenser pump, but with a regulating valve at the outlet of the condenser... where is the gain, with the full speed pump with the trothling valve, or with the VFD pump with a full open outlet?
    -Cheers-

    Tycho

  45. #45
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Segei View Post
    In North America typical ammonia refrigeration plant has minimum condensing temperature is 20C or 110 -115 psig. This is just typical plant operation. I know a few plants that have condensing temperature of 10C. Suction pressure of these plants around 0 bars. For higher suction pressure we need oversized oil separator. Cost of the compressor with this oil separator will increase by 5%. I talked to several compressor manufacturers and everybody claim that their compressors can operate at condensing pressure 15C even with 2 bars suction pressure, but oil separator should have right size.
    On freezing plants with -0.3 bar in suction, we still run with a PM valve on the discharge to keep the discharge at 30C, if we do this, we can use the same oil separators on -0.3 bar suction, or at 3.5 bar suction to prevent any accidental oil loss in case of a "alarm stop" on the compressor... we have a "fineoil separator" incorporated in our design, so we can run on any pressures you say and still get 99% of the oil loss back into the compressor suction... we design our freezing systems with two PM valves, so when you need a defrost, we shut of all hotgas to the condenser because the condenser PM is on 0.1 bar higher pressure. and lead it all to the plate freezer or evaporator that needs the hotgas... sure we are like the most expensive air--air-split unit
    but after awhile, we will give you more back for your buck...


    Of course you could add a full speed condenser pump, but with a regulating valve at the outlet of the condenser... where is the gain, with the full speed pump with the trothling valve, or with the VFD pump with a full open outlet??? I ask you???

    This point could be discussed to death....

    I am sure...

    But of course I am right, if we use an oversized condenser, and use a VFD pump, to pump the minimum required through the condenser, is better than a full speed pump blowing crazy amounts of water though the condenser, throttled by a vale at the outlet, causing a higher pressure, and more cavitation on the pipe out of the condenser
    -Cheers-

    Tycho

  46. #46
    Join Date
    Jan 2006
    Location
    Norway
    Age
    39
    Posts
    792
    Blog Entries
    4
    Rep Power
    21

    Re: Desuperheating by liquid injection

    Quote Originally Posted by Segei View Post
    In North America typical ammonia refrigeration plant has minimum condensing temperature is 20C or 110 -115 psig. This is just typical plant operation. I know a few plants that have condensing temperature of 10C. Suction pressure of these plants around 0 bars. For higher suction pressure we need oversized oil separator. Cost of the compressor with this oil separator will increase by 5%. I talked to several compressor manufacturers and everybody claim that their compressors can operate at condensing pressure 15C even with 2 bars suction pressure, but oil separator should have right size.
    Maybe in plants running in some allaskan fjords will be running at +20 C
    I'm talking about seawater cooled plants running in the waters off norway that can be between 4-15 degrees
    Yeah well, with 20C on a freezing system, that is normal operating temperatures... if you have 0 bars of suction pressure, that is all and good, everything should be working well

    thanks to you

    You know a few plants that run on 10C condensing, I can tell you, they are not using LRI for oil coolong... if they do, I would like to see first hand...

    You know... without being hard headed, I thnk that if I had the chance to install one of our systems on a ship fishing in Alaska, I could have imroved that ships catch of 10% in a season
    -Cheers-

    Tycho

  47. #47
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    I have nothing against VFD pump this is good idea.
    About alarm stops of the compressors. I had them several times and no problem with oil. When compressor stop check valves on suction and on discharge will close and pressure in the compressor will something between discharge pressure and suction pressure. For the compressor with suction 0 bars and discharge 8 bars, when compressor stop pressure will be 5-6 bars.

  48. #48
    Join Date
    Sep 2008
    Location
    Canada
    Posts
    751
    Rep Power
    12

    Re: Desuperheating by liquid injection

    Plants with 10C condensing have thermosyphon oil cooling. For LRI cooling several options are available: electronic expansion valves, liquid pump or injection in port side.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •