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

New Efficient AC

  1. Mar 26, 2005 #1
    Dear Folks

    This is the biggest break through in AC I've seen. My only question is if there is a maintainence issue with the build up of deposits of any sorts in the channels due to poor water quality.

    http://www.coolerado.com/NewsAndNotable/DelphiIdalexCoolerado.htm [Broken]

    Last edited by a moderator: May 2, 2017
  2. jcsd
  3. May 3, 2005 #2
    re:AC breakthrough

    Just another application/refinement of some pretty old stuff. We've been using water for air conditioning for a long time. Everything from a pool of water or sprays on the roof to highly automated temp/humidity systems. See what you can find at ASHRAE or any of the HVAC sites.
  4. May 4, 2005 #3
    Evaporative cooling+dehumidification is what they are talking about(I don't see any savings in power, if I have been to the correct location of their website).
  5. May 4, 2005 #4
    Here's a description of the "NEW" cycle.

    It looks like a modified two pass cooler (similir to the one used on the last power plant I worked at).

    The site above makes it sound like scientists the world around gasped in horror because this system should not be possible. Silly. I looks possible because it doesn't use the subcooling to perpetuate the process. The process still uses an external blower and external power source so it should not be an "impossible" device. Ok I take that as a marketing ploy. "We make the impossible possible" type of gimmic.

    All that aside, it looks like an ingenious little work around. In the first pass the cooling air is cooled by the second pass. The second pass lowers the temperature more by using evaporation. The outside air is cooled by the second pass air. Not too hard to understand.

    Ground breaking? Probably not. If it works the way they claim will it see service? Yes.

    As for power savings, running a single blower (I can see a way to run the cooling air and the conditioned air from the same blower unit) would probably use less power than a compressor and seperate blower like current automotive AC systems use. Moreover, use the new 42V system (don't include that in the discussion directly---compare the new design and implementation to current designs without mentioning things like new designs run at different voltages than they used to) thus reducing I^2R losses and you will end up with a more efficient system.
  6. May 4, 2005 #5
    Power wise they are not bad. The EER for the unit described is 60000/1200 = 50 (their site rates the SEER at 40). This is quite good. However, it is at the expense of water use. The machine uses 12 gallons per hour. This is rather high. Evaporative coolers with drift eliminators and desuperheaters use less water with similar results. I recently designed a project with a chiller with an evaporative condenser and it will use less water and deliver 110 tons (using conventional compressors) at a 40% savings in energy over an plain aircooled chiller.
  7. May 4, 2005 #6
    Eric thanks for the heads up on this variant heat transfer system. Seems pitiful that all that "value added" air leaving the building as 'working fluid'could be renamed 'product air' by simply reducing the vapor pressure over the 'waterside' heatexchanger using a 'penberthy' pump. the mass flow of air driving the 'pump to reduce the pressure would be insignificant related to the amount needed in the above which is 50% of the total air flow. A 'fast responce' vapor pressure controller on a conventional freon heatexchanger adapted to circulate water through the freon side and then discharging into a flash tank would do as good or better than any of these other alternatives . I'd probably use a tandem 'savery engine' for producing the vacuum using water from the water heater or solar concentrator.etc.etc.

    frank MR. P
  8. May 4, 2005 #7


    User Avatar

    Staff: Mentor

    They throw around numbers like...
    ...but what does that really mean? It doesn't make any sense to me.

    Performance of evaporative cooling is so heavily dependent on climate, I can't imagine it being viable anywhere besides a desert. In Philadelphia, the design conditions are 92F dry bulb, 75F wet bulb. To get adequate cooling capacity and dehumidification, you need to supply at about 55/55. I can't see that happening.

    In Vegas, design conditions are: 109/69. If you look on a psych chart, the enthalpy of the air is actually lower than in PA because of the lower DB. I assume the conditions are probably about the same in Chihuahua, Mexico, where I just was - in Chihuahua, people already use evaporative cooling almost exclusively. And yes, the biggest problem is the water usage.

    I need to see some real performance numbers before I can say I buy that this is that big of a deal.
  9. May 4, 2005 #8


    User Avatar
    Gold Member

    Evaporative cooling is a technology that has been around since ancient Egypt. It is a great idea and huge power saver in Colorado, where the company the makes the things is based (as well as the rest of the arid West), and using it for cars as well as homes only makes sense.

    In Florida or North Carolina, I can't imagine that the technology would be beneficial in any signfiicant way.
  10. May 4, 2005 #9
    Re: humidity Issue

    Dear Folks:

    They claim to get "wet bulb" temps, and I dought Delphi (a 33 billion $ corp) would agree to produce it if it did not work.

    It's not a "swamp cooler", there is no transfer of moisture to product air. Althought in high humidity applications , to be most effective, they must use an additional component for dehumidification, IMHO, like Honeywell's " Perfect window" desicant wheel ERV. I've have sent them e-mails to this effect, am waiting for their reply.

    They claim to be able to purge all mineral build up, but I do not believe the "all" part.

    I also suggested to them to include a turbulent flow percipitator type air filter in their design. Like the Nutech ACS 2000R, that has a 5 year maintainance cycle,and removes 97% of particles down to 0.1 microns to reduce their units maintainance.

    Yes, water usage is a concern, in general applications , however in my nursery business plants love 60 to 70F water.

    I've posted them your coments and will report back here any answers to your questions.

    Thanks for the feedback, and please e-mail me if you have any other concerns

  11. May 4, 2005 #10


    User Avatar

    Staff: Mentor

    Yeah, a large corporation would never overhype a new product. :uhh:
  12. May 4, 2005 #11


    User Avatar
    Gold Member

    Two words: New Coke.
  13. May 5, 2005 #12
    I have been through the theory given in the above link but still unable to understand how the exit stream can go up on the saturated line by the addition of moisture. Theoretically it is possible to cool down the air to its wet bulb temperature. But if DBT = WBT then DBT = WBT = DPT and further moisture addition shouldn't be possible.

    Secondly, when we precool the wet channel air, its moisture carrying capacity reduces. If we plot two process lines starting from point A, say, the line AB which with only evaporating cooling follows consant enthalpy line. The line ACD first follows the constant humidity ration line(as long as precooling exists) and then follows the constant enthalpy line. Thus point D is always below point C thus showing low moisture content.

    Further the process of precooling and then moisture addition show a drop in enthalpy. I am still not clear where this heat will flow to.

    Finally, dehumidification by desiccant wheel is also isenthalpic(infact it increases a bit due to thermal inertia of regenerated desiccant wheel. To bring the process air to the starting point DB, it requires cooling and this may not be a good combination.

    I will go through it tonight leisurely and will let you know if I come up with anything better.

  14. May 6, 2005 #13

    Can you get actual observations viz., flowrates, inlet, oulet wbt and dbt for various streams, spray water temperature and flowrate from their case studies?
  15. May 6, 2005 #14
    I would like a little more information on the heat exchanger. Are the two air streams totally isolated, or is there possibility for cross contamination with the humid side? Where is the water introduced, The drawings at the site describing the process are ambiquious at best.

    I believe that they are marketing this mainly in the areas that would benefit from swamp coolers, but without the humidity rise within the space. I think in this application, these could be a useful tool, but as Russ was saying, in my area, NorthEast USA, I also see very little practical use for this type of equipment. Possibly in a warehouse or factory environment with a desire to add some cooling, which has access to a cheap or free water source.
  16. May 6, 2005 #15
    Erich - Great comment on mineral buildup - Figuring out how to control minerals has been almost as challanging from an R&D point of view as indirect evaporative cooling below wet-bulb. We have designed the heat and mass exchangers (HMXs) so the minerals are carried to the edge of the working air plates and either drop off in liquid form, or collect on a 'mineral lip' in solid form. The minerals that solidify on the lip do not impeed the performance of the HMX and eventually flake off and are carried down the drain.

    There is no scaling on the inside of the HMX. This based off 3 years field experience on an early HMX design, 1 1/2 years field experience on the current production HMX, and accelerated testing using high TDS (total dissolved solids) water in our test chamber.

    Rick Gillan - Idalex and Coolerado
    Last edited by a moderator: May 2, 2017
  17. May 6, 2005 #16
    kleinjahr - Thank you for your comments. It is true that direct evaporative cooling has been around for thousands of years, and it works! Indirect evaporative cooling has also been around a long time. All the indirect evaporative coolers on the market before the Maisotsenko Cycle (M-Cycle) have been able to achieve about half the temperature drop as directive evaporative coolers - which makes sense because you have to go across a heat exchanger.

    What's new and different is that Coolerado Coolers can cool below the wet bulb temperature of the incoming air stream. At first glance this would seem phisically impossible because a direct evaporative cooler's limitation in the wet bulb temperature, and you still have to go across a heat exchanger where there will be more losses.

    Coolerado Coolers have the ability to cool below wet bulb, and approach dew point. The Department of Energy's (DOE) National Renewable Energy Laboratory (NREL)tested a unit and reported that we had a 120% wet bulb approach (meaning we were 20% below wet bulb), which was an 86% dew point approach. All without addind any moisture to the product air stream (indirect evaporation).

    There are current text books and articles from industry experts that state indirect evaporative cooling's theoretical limitation is wet bulb. See HPAC Engineering's July and December issues for two such examples.

    We discuss direct and indirect evaporative cooling in more detail on the second video that is available on the Coolerado website.

    Rick Gillan - Idalex and Coolerado
  18. May 6, 2005 #17
    Quark - Thank you for your interest in Coolerado. The system is not cooling plus dehumidification. Rather, it is indirect evaporative cooling. The best way to get a basic understanding of those concepts is to watch the two videos on the Coolerado website in the Cool School section.

    Rick Gillan - Idalex and Coolerado
  19. May 6, 2005 #18
    faust9 - Thanks for your interest and the time you spent going to both websites to understand the cycle. We'd like to see a schematic of the two pass system you mention if you could possilby send it to us.

    We did not try to create a marketing gimick; rather, our approach has came about through the years as a responce to the comments we get. Sounds like it is coming off different than we intend and we need to revisit our website approach. Thank you for pointing this out.

    From your short comment, it does sound like you have a basic understanding of the cycle. The key to getting below wet bulb is incrementally cooling both the product and the working air streams.

    The power savings we report are based on DOE testing where they report our Energy Efficency Ratio (EER) is 40 and higher. Compare that with vapor compression systems that have an EER of about 10 (Seasonal EER of 19).

    Rick Gillan - Idalex and Coolerado
    Last edited by a moderator: Apr 21, 2017
  20. May 6, 2005 #19
    Artman - Thank you for your interest in Coolerado and your comments. Actually, our water use is about the same as an evaporative cooler. The numbers we report are generally very conservative, as is the 12 GPH. That number is for the Coolerado R600 during the hottest, driest part of the day in the western USA with high mineral content water. Average GPH is likely to be less than half of that, which is about the same as evaporative coolers.

    DOE's NREL is conducting a long term test regarding this issue, and we hope to able to report thier independent results at the end of this season.

    Rick Gillan - Idalex and Coolerado
  21. May 6, 2005 #20
    Mr. P - Thank you for your suggestions on how to improve a building system with a Coolerado Cooler. Several of our clients have calculated the costs and benefits of adding more equipment, but eventually they come back to a simpler system. A three to four fold improvement in energy savings over traditional systems is typically good enough for now. Further improvements as you suggest may be cost effective in the future.

    Rick Gillan - Idalex and Coolerado
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook