1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Calculating radiator/fan/flow required when have a low Delta T

  1. Mar 26, 2017 #1
    Trying to narrow down what type and size radiator and flow of both the water and air I should be looking at that would give me my best heat exchange efficiencies, if I want to input 50F water into a radiator that will have 100F air (60% RH) blowing through it and my goal here is the cooling down of that outgoing radiator air as much as possible.

    I want to see if I can do a little better than just fooling with any random old car radiator, fan and pump, if any can point me to what I might should be looking at instead. Thank you!!
     
    Last edited: Mar 26, 2017
  2. jcsd
  3. Mar 27, 2017 #2

    russ_watters

    User Avatar

    Staff: Mentor

    Generally the most important consideration for designing a heat exchanger is how much heat you need to exchange. Do you have any idea how much you need? Next, the working temperatures and flow rates may or may not be important, depending on the application; can you tell us the application?
     
  4. Mar 27, 2017 #3

    anorlunda

    Staff: Mentor

    Your question can't be answered without specifying size and weight. For example, 10 auto radiators connected in series will transfer more heat than 1 radiator.
     
  5. Mar 27, 2017 #4
    Russ, I'm trying to reduce hottest air temp at top of greenhouse that, with CO2 injection in there,
    I want to slow rate of heat gain in there, as long as I can each day, before having to surrender and
    open up greenhouse to replace hot air with cooler outside air.

    I have 50F water, could rig up to pump through car radiator with fan drawing hottest air from top to
    go through it, but likely better matches of radiator, pump flow and fan cfm's for the low delta T temps,
    so as to achieve better efficiencies than random car radiator/fan/pump, that's what I'm inquiring about.
     
  6. Mar 27, 2017 #5
    anorlunda,
    I thought about multiple radiators, too. I should have been more specific initially.

    Like to see what can be done with one, radiator/pump/fan setup sized right for this application,
    best as can be acquired off the shelf without spending tons of money on something too exotic.
     
  7. Mar 27, 2017 #6

    anorlunda

    Staff: Mentor

    Thanks for being specific. :frown:

    A truck radiator & fan from a junkyard.
     
  8. Mar 27, 2017 #7
    anorlunda,
    That's always been my fallback, but wanted to see if anybody here had any better ideas, as those are designed specifically for much higher Delta T's.

    I'm sure there's some 'rules of thumb' out there for dealing with smaller Delta T's that would be telling me then things like to look to use the thickest finned radiator available and/or move more or less water and air through it faster or slower, etc., than what a truck radiator is designed to optimally do.
     
  9. Mar 27, 2017 #8

    anorlunda

    Staff: Mentor

    @shane2, you won't make friends here on PF by posting questions while holding back pertinent information. That just wastes our time such as by telling about things you already thought of.

    You get the highest heat exchanger efficiency with the biggest surface area, the thinnest walls and the most rapid air flow.
     
  10. Mar 27, 2017 #9
    anorlundra, If I told you everything I've already thought about, I'd never get around to asking about what I don't know and hadn't thought about yet.

    But, yes, I should have initially specified preference for one radiator, and my earlier rejection of utilizing multiple units in series.

    You wrote; "You get the highest heat exchanger efficiency with the biggest surface area, the thinnest walls and the most rapid air flow."

    I get that, but let me ask this, for a high Delta T heat exchanger design, which of those components above, if any, would you want to most emphasize if it were to now be used, instead, in a low Delta T application?
     
  11. Mar 27, 2017 #10

    anorlunda

    Staff: Mentor

  12. Mar 27, 2017 #11
    Thank you!
     
  13. Mar 27, 2017 #12

    russ_watters

    User Avatar

    Staff: Mentor

    Ok, so that's a little to go on -- you didn't say how big, but to make a dent in the temperature I'm going to guess you need at least 2 tons of cooling per 1,000 square feet of greenhouse.

    With a 30F water delta-T and air delta-T you would need 1.6 gpm of water and 740 CFM of air. If the radiator is 24" on a side (24" fan) that seems like a reasonably slow airflow to get that from a car radiator. You might look to see if you can find the specs for a real car radiator though and see how they translate....keeping in mind that a car radiator dissipates about a third of the power of a car's engine as heat....
     
  14. Mar 27, 2017 #13
    Russ,
    That's very interesting, that "30F water delta-T and air delta-T you would need 1.6 gpm of water and 740 CFM of air."

    What formula or calculation did you use to generate that?

    I ask because if I know or can't change one of those parameters in either my water pump or air fan, that'd then tell me what the other would need to be, yes?
     
  15. Mar 28, 2017 #14

    anorlunda

    Staff: Mentor

    I missed the part about the greenhouse. Perhaps the whole idea of any kind of radiator+fan is not the best. A simple fountain would do much better. In engineering terms, a wet cooling tower.

    The smaller the water droplets, the bigger the heat transfer area. Of course, wall thickness is zero.

    Therefore cooling capacity in BTU/hour is proprtional to water flow rate, and efficiency is inversely proportional to droplet size. Simple as can be.

    You could sprinkle the water from above or below.

    There might be secondary benefits of extra moisture in your greenhouse.
     
  16. Mar 28, 2017 #15

    russ_watters

    User Avatar

    Staff: Mentor

    For water, it is gpm*delta-T*500=heat (in BTU/hr)
    For air, it is cfm*delta-T*1.08=heat (in BTU/hr)

    It strikes me that I neglected a significant issue: you provided the starting temperature, not the ending temperature. It is the ending temperature that drives the design, since that's when your heat flow into the building is highest and ability to remove it is at its lowest. For example, if you are looking for an 80F space temperature, you won't be able to achieve that 30F delta-T anymore...you'll need a higher airflow and lower delta-T.
     
  17. Mar 28, 2017 #16
    anorlunda, don't beat me up for not disclosing more earlier, but yes, evaporative cooling will also be utilized and expected to do most all of the heavy lifting. Being in desert of very low humidity to begin with, it should both be very effective cooling and add much needed RH to internal environment.

    Before that hottest air at ceiling gets that misting treatment and is sent along floor ducting to plants, I wanted to explore knocking that temp down a bit first in front of the mister with radiator set up initially inquired about here.

    And, for more details, 400 sq ft footprint, 10' high, 4000 cubic feet, but roof opaque and most all insolation in summer reflected in through vertical S, W and E facing diffused poly via adjustable draw bridge like horizontal reflector panels. Which also let me start shutting off insolation anytime, closing them up ever more, if/when internal temps still getting too excessive in spite of all my best efforts to cap it. There's a little more to it, but that's primary set up.
     
    Last edited: Mar 28, 2017
  18. Mar 28, 2017 #17
    Russ, just poking around online, and more money than I'd like to spend, but saw 16" radiator pull fan rated at a little over 2,000 cfm and electric radiator water pump rated at 35 gpm. Set up on quality good size car/truck radiator, that combo would probably do all that anyone could hope for extracting heat from hot air with cooler water that's a low Delta T, yes?

    PS - Then, again, had another tell me I want slowest air movement through radiator to maximize it's contact with water cooled fins.
     
    Last edited: Mar 28, 2017
  19. Mar 28, 2017 #18
    Now that it's been brought up, another question I'm going to have to get a fix on, in using evaporative cooling, is with its cooling effectiveness steadily waning the more it's used in that closed 4,000 cubic foot environment, that's also then raising internal RH inside there, how long will it be effective before I either need to shut down heat coming in and/or suspend closed system CO2, air everything out, starting over again with low RH air.

    Until I do it and see, under varying insolation intensities and durations, hard for me to say now how long through the day it'd still be cooling significantly, but some here might have some 'rules of thumb' for me to consider.

    As far as how much RH is too much for the plants, I do know they like RH of around 60% when 70F, 70%RH when 80F and pushing 80%RH when approaching 90F.

    BTW, right now, at 1:15PM, out there, it's 82F, expecting high of 85F, and RH is 10%.
     
  20. Mar 28, 2017 #19

    anorlunda

    Staff: Mentor

    image.jpeg

    You are confusing two things. A droplet doesn't need to evaporate to absorb heat form the air. Spraying cold water will lower air temperature even if R.H. = 100%.
     

    Attached Files:

  21. Mar 28, 2017 #20
    Gotcha, or when 100%RH, also cooling via blowing that hot saturated air through colder water running radiator, yes?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Calculating radiator/fan/flow required when have a low Delta T
  1. Radiator calculations (Replies: 3)

  2. Calculate fan speed (Replies: 6)

Loading...