Turbulent flow

Discussion in 'General Discussions' started by docjenser, Feb 10, 2011.

  1. Looby

    Looby Member Forum Leader

    WF and CM (and everyone else's) spec sheets say otherwise.

    Lower flow = reduced HC, and reduced HE, and reduced COP.

    Compressor amps also fall -- but not as much as HE or HC,
    (as evidenced by the lower COPs). Of course, delta-T rises,
    but not as much as HE falls.

    Absolutely nothing surprising here -- it's all 100% predictable
    from basic (125+ year old) physics and thermodynamics.

    ...not just a good idea, IT'S THE LAW!

    Last edited: Feb 21, 2011
  2. AMI Contracting

    AMI Contracting A nice Van Morrison song Industry Professional Forum Leader

    hmm but we did boast higher EWT.
    have we discovered the momentum heater?
  3. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    Wasn't it the LAW a few hundred year ago as well that the sun goes around the earth?

    You have been wrong here before with many of your predictions in this thread, you should ask yourself if you could be wrong again once in a while.

    Tranquility 27 data sheets for example for lets say a 6 ton in full load at 2100 cfm airflow show at 30 F source a HE of 35.7 kbtu/h at 12.8 gpm source flow, and 36.8 kbtu/h at 17 gpm, an increase of roughly 3%. COP remains the same 3.2, may be a rounding issue. No question you can increase HE slightly with higher flow, but there is a limitation how much heat can be extracted with an increase in flow. In the example above a over 30% increase in flow increased the HE only by 3%. I always attributed the slight increase in HE to the lowering of the average water temperature in the coil more towards the EWT.

    I was unable to measure any difference, but I admitted the limitations of my instruments, which were specifically the 0.1F accuracy for the thermometer.

    I am not trying to beat the physics here, but I was originally asking if we can slow down the flow and save pumping power without paying too much of a penalty for the heat exchange. The reason for this was that I could not find any measurable penalty for loop performance or HP performance, but was able to save 245 watts of pumping power.

    So are we advised well to stick to the "above 2500" reynolds number rule. Or is something else going on where which we missed with our explanations. Is the water mixing good enough in the slinkies, or is the loop too long that the reynolds number becomes less important? We might be missing something here.

    Is it different with vertical, straight pipes being much shorter? More importance to turbulent flow with shorter pipe? I always thought that one way to find out is to try it and produce some data to find an answer, or to question the answer we had before.

    But may be we should not do this anymore. We could just have Looby predict everything from his 125 year old basic physics knowledge. And don't report an observation which might challenge this knowledge. Then he will tell you that you have too much of an open mind...that your brain now will fall out......;)
    Last edited: Feb 21, 2011
  4. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    I'm inclined to agree. Circ pumps are relatively inefficient, so increases in GPM accompanied by increased head loss come at a fairly high price.

    I agree that turbulent flow yields much greater heat transfer than laminar flow...from the loop fluid to the pipe walls

    However, if the limiting factor turns out to be conductivity / diffusivity of the soil surrounding the carefully-turbulently-heated / cooled tube wall, what good does increasing heat transfer from loop fluid to HPDE pipe wall by 3x actually accomplish?

    My present car can go 100 mph, but roads I use regularly top out at 50 mph. Would buying a 150 mph car do me any good?
  5. Looby

    Looby Member Forum Leader

    Yep, exactly! It was prescriptive law proclaimed (and enforced!) by
    the science-denying forces of superstition and invincible ignorance.

    Although the flat-earthers haven't disappeared entirely, the proponents
    of descriptive law (a.k.a. science and engineering) have earned some
    street cred over the past few centuries -- largely by building upon the
    discoveries of earlier generations, rather than hitting the reset button
    every 20 years or so.

    OTOH, if you think Gibbs, Carnot, Reynolds, and all them other dirty
    furriners got it wrong, have fun retrying all possible permutations and
    combinations of design elements that might be important ...or not.

    ...there's a healthy list of interesting (re)experiments in your post,

  6. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    I don't question that they got it right, but based on experiment/observation, I question what else is going on and trying to explain it. In our application with a loop, there might be other limitations, as Curt gave one example. It is the difference between theory and practice.
  7. Palace GeoThermal

    Palace GeoThermal Well-Known Member Industry Professional Forum Leader

    doc... keep the data coming. I am interested in the results.
  8. geome

    geome Member Forum Leader

    For our system, 1 pump running draws 0.80 amps at 241volts. 2 pumps running draws 1.85 amps at 240.7 volts. I did not read pump draw individually when 2 pumps were running (just combined) sorry.

    If I understand the following correctly, I'm getting similar loop results to Doc. That is, roughly the same EWT with 1 pump vs 2 pumps running despite a greater delta-T on brine temperature taken at PT ports with 1 pump running.

    You may recall I have no slinky, six 500 foot circuits in parallel (in a 4-pipe layout with 3 trenches), loop and one flow center with two pumps shared in common between a 3-ton packaged unit and a 2-ton split unit (both Envision 2-stage compressors.)

    I took the single pump readings first after 15 minutes of run time. Units were then shut off for about 10 minutes so I could reconnect the second pump. I then took the 2 pump readings after 15 more minutes of run time.

    1 pump, 3-ton unit, 2nd stage
    EWT 38.1F
    LWT 29.3-32.9F (avg 31.1F) Delta-T 7.0 F
    EP 55.2 psi
    LP 53.2 psi Delta-P 2 psi

    1 pump, 2-ton unit, 2nd stage
    EWT 37.8F
    LWT 30.3-33.0F (avg 31.65F) Delta-T 6.15 F
    EP 54.8 psi
    LP 52.4 psi Delta-P 2.4 psi

    2 pumps, 3-ton unit, 2nd stage
    EWT 38.1F
    LWT 32.7F Delta-T 5.4 F
    EP 60.0 psi
    LP 56.2 psi Delta-P 3.8 psi

    2 pumps, 2-ton unit, 2nd stage
    EWT 37.9F
    LWT 32.5-34.5F (avg 33.5F) Delta-T 4.4 F
    EP 60.9 psi
    LP 56.5 psi Delta-P 4.4 psi

    It should be noted that our systems have not been running much recently due to more mild weather, and our loop is probably over sized by 20% (600'/ton instead of 500'/ton.)

    Sorry if this was suggested before, but since loop flow is lower with only 1 pump running, could the additional time that the brine is in the loop be making up for less turbulent flow (etc.) under some conditions?

    I'm off to figure out GPM for my (somewhat related) thread:
  9. geome

    geome Member Forum Leader

    It looks like our flow is:

    One pump
    3-ton unit 6.5 gpm
    2-ton unit 5.4 gpm
    Total 11.9 gpm

    Two pumps
    3-ton unit 9.5 gpm
    2-ton unit 7.5 gpm
    Total 17.0 gpm

    Looby did previously note in the other thread: "BTW, you're fine on Reynolds number/turbulence as long as you keep the total flow a little over 12 gpm." So maybe at 11.9 gpm we are just close enough to turbulent flow?

    I'll leave this to you guys to figure this out so my brains (what's left of them) will stay in. :)
  10. Mark Custis

    Mark Custis Not soon. Industry Professional Forum Leader


    Keep going. I for one belive that things are not always as they appear.

    Warm regards from the wingnut in cleveland with a passion for understanding flow,

  11. Mark Custis

    Mark Custis Not soon. Industry Professional Forum Leader


    I have learned a great deal from you. Thanks.

    This time I am with Joe and Doc.

    Keep teaching, I will try to keep learning.

  12. engineer

    engineer Well-Known Member Industry Professional Forum Leader

    Scatter a few wingnuts randomly in the water lines and the water will have to go around them, creating eddies that will act similarly to turbulent flow.

    I don't doubt any of Looby's science - it is consistent with everything I've forgotten from college since graduation (1988)

    That said, there is nothing quite so good as accurate data...Geome's, DJ's and my own experience suggest that maintaining a minimum Re # may not always be de rigueur in every situation.

    Dumping an inefficient,unneeded pump for several thousand hours per year represents a clear and present conservation opportunity if carefully justified.
  13. Mark Custis

    Mark Custis Not soon. Industry Professional Forum Leader


    OU '71. Journalism.
  14. One should note that if this work leads to a correct understanding of turbulence, the financial gain could be more significant than the $100 in annual pumping costs:

    It is widely understood that the first step in understanding turbulence is a proof of the Navier-Stokes equation--a feat worth $1 Million as #6 on the original list of 7 (6 remaining) Millenium Prize Problems in Mathematics: Millennium Prize Problems - Wikipedia, the free encyclopedia

  15. Mark Custis

    Mark Custis Not soon. Industry Professional Forum Leader

    Thanks Adam

    Will it heat my home?
  16. Looby

    Looby Member Forum Leader

    "Carefully justified" ...uhh, based on what design critria?

    Are you really going to design and sell a loop field that ignores
    manufacturer's recommendations and your geo-design software?
    As a prospective customer, I'd RUN away from such a proposal.

    BTW, regarding geome's data, how does the 15 minute run-in
    period compare to the brine's transit time through the system?
    What level of confidence do you have that the measurements
    reflect something reasonably close to steady-state conditions?

    BTW2: what does IGSHPA say about Reynolds numbers?

    BTW3: why are there so few old, bold pilots?

    Last edited: Feb 23, 2011
  17. docjenser

    docjenser Well-Known Member Industry Professional Forum Leader

    "One measurement is worth a thousand expert opinions."

    I start to like your credo.

    Genome, your data suggests that you would benefit from a setup where one heatpump turns on one circulation pump each, so they are staged. I will take some measurements with even shorter vertical loops, and post them here, we will see what happens, even if it means to be lectured by Looby again, telling you it is all not possible.
    I am really glad now that he is going after GENOME now, and not me anymore;)

    PS: The measurements I posted were after 8 hours run time each, pretty close to steady-state.

    It appears to me that the lower flow causes a 3% performance penalty at the heat pumps at the flow rates we are talking about for the lower flow, according the manufactures' specs, so a 6 ton HP drawing 5 KW would pay a 150 W penalty, but you would gain 245 watts for lesser circulation power. Now, with 3 or 4 ton HP, that looks even more favorable. Now, that is 2nd stage running assumption, thinking the pumps only run at 2nd stage 20% of the time, needing only 2/3 of the flow, this looks even more favorable since the 3% performance penalty does not exist anymore.

    So why would you run as a customer? Shouldn't we use the least amount of pumping power which makes the system run most efficient?
    The future will be variable speed pumps where you dial in the sweetspot, like the most efficient delta T.
    Last edited: Feb 23, 2011
  18. Looby

    Looby Member Forum Leader

    When a vendor claims that industry-wide design practices are
    all wrong and that he has the inside track on "a better way,"
    (endorsed by some guy on TheIntertubes) what would you do?

    BTW, could I interest you in a 300 mpg carburetor?

    Once again I ask:

    - what does WaterFurnace recommend? ...why?

    - what does ClimateMaster recommend? ...why?

    - what does IGSHPA recommend? ...why?

    It would seem that all of the above have a major vested interest
    in optimizing efficiency, capital cost, and operating costs. What
    possible motive could they have to recommend over-pumping?

    ...yep, but only ONE thousand,

    Last edited: Feb 23, 2011
  19. geome

    geome Member Forum Leader

    I'd be happy participate in a test that everyone would agree to in advance.

    For the next 10 days, our average high is expected to be about 55F, and our average low is expected to be about 40F. If our units need to be running many hours per day for many consecutive weeks, our window of opportunity may be over for this winter.
  20. Designer_Mike

    Designer_Mike Member

    I'll toss in my opinion

    "Why would the different manufacturers recommend high turbulent flow rates in the loop water?">>

    My opinion would be because in improves performance of the heat pump. BUT how much does it cost?...ah, that's right....the loop pump requirements are not figured in on the heat pump efficiency numbers are they? (so they don't care)

    SO lets get that loop water flying!....it's free;)

    At least that is my understanding, but I'm not a expert.

    I'm 99.9% positive that turbulent flow improves heat transfer over laminar flow, I'm also 100% sure that turbulent flow costs more to pump.
    There is a break even point where it doesn't make sense to produce turbulent flow....what point that is will take some experimenting AND will probably be different for every system.

    This is why I intend to use a sizable buffer tank and only circulate the loop water when the buffer tank temp is not where I want it.

Share This Page