Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

Pump head calculation comparison

hot_rod
hot_rod Member Posts: 22,019
Assume a circuit with a developed length of 100 feet of 3/4" copper tube, AWT 140F, flow rate 5 gpm.

Using the formula in Idronics 16, we come up with 4.93 feet of head.

Using the Sizer wheel it shows just over 8 feet.

The Taco calc formula shows 3.7, they solve the 1.75 factor and show that number in table 5, making the math simpler without a scientific calculator, nice!

Even with a temperature change to 180F in the formula, still quite a bit of spread?
Am I missing something, or is there that much "slack" in the System Sizer number?
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream

Comments

  • bob_46
    bob_46 Member Posts: 813
    Crane Paper 410 Flow of Fluids was/is kind of a bible for ME's and ChemE's. It starts with the statement that " All practical formulas for the flow of fluids are derived from Bernoulli's theorem" in that equation Bernoulli squares the velocity thus it shows up in Darcy's formula. My engineering manuals from B&G are from back in the 60's and they square the velocity. So , I believe the difference is squaring the flow instead of 1.75 and some differences in fluid factors and C. I noticed Caleffi follows Siggy to the letter ,Taco combines a and C and comes up with a new factor K. If you do the Taco calc and square you come up with 7.375 feet, close to the System Syzer .
    bob
  • hot_rod
    hot_rod Member Posts: 22,019
    Thanks Bob, that sounds like the correct explanation.

    My question would be who's numbers would you use to size a circ. This is a fairly low flow rate example, of course, not a huge differance..

    The next step in pump selection is plotting the system curve, this is what those 3 examples graph out to look like. Now the pump is selected based on the above data.

    The load calculations being discussed on the site got me thinking about this.

    Chose 3 different Load Calc programs and you could indeed find a wide discrepancy in numbers. Then you size the pump to a load that is 25% too high, add the pump size error of up to 50%!

    Blinding me with science :)
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
    Tinman
  • Paul48
    Paul48 Member Posts: 4,469
    Hmmm....Too bad we don't know someone that has access to the equipment to do the testing. :wink:
  • Harvey Ramer
    Harvey Ramer Member Posts: 2,239
    I often thought it would be nice to have a system curve plotting pump for existing systems. Basically, the pump would have to be infinitely variable. Have pressure gauges on the inlet and outlet and a flow meter that is accurate across a wide range.

    But it probably isn't worth it on most residential systems.

    Be great for commercial though.

    I never realized there is such a discrepancy in the software or sizing charts.

    Really though, most design/build systems, we all take an educated guess on the number of fittings. Or multiply by a factor.
  • bob_46
    bob_46 Member Posts: 813
    Just to add , the Copper Development Association uses the Hazen-Williams formula and they come up with 5.31' for the same example.
    bob
  • Zman
    Zman Member Posts: 7,561
    Siggy's software comes up with 4.9 feet...
    "If you can't explain it simply, you don't understand it well enough"
    Albert Einstein
  • bob_46
    bob_46 Member Posts: 813
    Hot Rod, did you use the iDronics formula for constructing the three system curves ? If you use scale No 5 on the Syzer wheel
    to construct the system curves the head losses get even greater,it compounds the the excess head from scale No2.
    So I guess my right hand helper from the last 50 years goes in the round file .
    bob
  • hot_rod
    hot_rod Member Posts: 22,019
    I used the Idronics formula for the middle curve, System Syzer, scale 2 for then upper, and the Taco formula for the lower curve.

    I see what you mean, the further you go around the "wheel" the larger the discrepancy.

    Makes me wonder how a series 100 circulator ever got specced for anything other than very low flow, large diameter pipe jobs, it is such a flat curve, low head circ?
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • bob_46
    bob_46 Member Posts: 813
    Hot Rod I went through your math on your original post and I think you made an error on the Taco calc. I come up with the same answer as ironics 4.9321'. I have a couple of questions, can you tell me how the numbers for the factor "C" are derived and where the (f)^1.75 comes from ?

    Since B&G use Darcy-Weisbach and Copper Development uses
    Hazen-Williams I think Caleffi should publish a paper with ASHRAE and call theirs the HotRod-Siggy equation.
    bob
  • hot_rod
    hot_rod Member Posts: 22,019
    C is referred to as the pipe size coefficient, I'm not sure which formula we used to develop our table. Taco calls it k in their formula, and their table is different when compared to the one we used.
    As I understand it, "c" is based on the id of the tube and the roughness factor. The roughness varies on the type of tube, and even from old to new copper in some formulas.

    The exponent, again as I understand it, is a method to crunch all the units into the same measurement.

    I asked Siggy about exponents that many years ago and got this explanation. It is in Nov 2007 PM mag, if this link doesn't work.

    I'll try and get more info on which formulas were used to develop the c tables.

    http://www.pmmag.com/articles/91363-where-do-the-constants-come-from-br-john-siegenthaler-pe
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
  • bob_46
    bob_46 Member Posts: 813
    Thanks HR , your comment about the B&G 100 tells how things have changed. On residential construction we always picked the pump first (B&G 100) and then designed the system around it. That was also the pump that came on the boiler.
    bob
  • Tinman
    Tinman Member Posts: 2,808
    edited July 2015
    This is a very interesting thread, maybe even more so because I've been using the wheel for so long. In this specific example maybe a moot point too because I'm guessing not too many of us use anything smaller than the smallest 3 speeds, in which case we'd be covered in all 3 results.
    Makes you think though about some of the larger systems and all the wasted energy.
    I like Harvey's idea of the pressure gauges at the inlet and outlet and using a flow meter but I'd like to see it in the smaller systems too. Why not dial it in tighter if we can? There's so much talk about parasitic waste and right now we're just saying it's not worth it on residential. Maybe not one residence but if you multiply that waste by the millions of homes...
    Steve Minnich
  • hot_rod
    hot_rod Member Posts: 22,019
    Agree, on this small load, not a huge deal. But it may be over pumping AND consuming more electrical energy than needed.

    On a good day a wet rotor PSC circ runs mid 20% efficiency. If it's running off it's curve? Who knows.

    Wonder what a 50 gpm example would look like?
    Bob "hot rod" Rohr
    trainer for Caleffi NA
    Living the hydronic dream
    Tinman
  • Tinman
    Tinman Member Posts: 2,808
    I'd run the numbers but I'm on my way to a Boiler Room. : )
    Steve Minnich
  • bob_46
    bob_46 Member Posts: 813
    Iran da numbers for 2"M 100' 140º 50GPM water.

    Syzer 5.4' of head
    Copper Dev. Assoc. 4.62' of head
    Ironics 3.72' of head
    Taco 3.73' of head

    2"PEX 100' 140º 50GPM water.

    Uponor 10.62' of head
    Ironics 9.39' of head
    Taco 9.4' of head

    If you develop a system curve using scale 5 on the Syzer
    you come up with a much steeper curve than using the
    equation in Ironics 16.
    bob
  • Tinman
    Tinman Member Posts: 2,808
    So the moral of the story seems to be that it's either crunching the numbers via idronics formula or using Siggy's software.
    Steve Minnich