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.

Hypothetical

bill_105
bill_105 Member Posts: 429
So I'm describing the benefits of placing an expansion tank at the suction side of a circulator to some guy.

I then get hit with a scenario which I was without a decent answer. Here it is.

What if you were to put an expasion tank at the discharge side of a circulator, then put a second tank at the suction side.

First, the tank is a hydtonic bookmark. Pump operation cannot affect system pressure at tank location.

Second, a circulator creates pressure differential. High pressure goes to low pressure.

Now with two facts it would seem a circulator wouldn't work.

So what would happen if you did this?

Now don't ask why you would do this. Just wondering.

Comments

  • icesailor
    icesailor Member Posts: 7,265
    Hypothetical:

    The circulator will always work. It doesn't matter or care where the tanks are, just to some theoretical technocrats.

    If you had the system at rest at 12#, and you started the circulator, and had gauges on both sides of the circulator/tanks, and the outlet said 14# and the suction/inlet side said 10#, you would have a 4# differential that is the head pressure and the flow rate in GPM's once you checked the chart.  That equals the resistance of the system.
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    As a learned technocrat...

    Thanks for the handle Chris. I see your attempt to increase your knowledge and understanding of the PONPC hasn't changed much :-)



    To Bill, having two PONPCs, one before and one after the pump, will cause the pump to split it pressure differential potential. So if the pump is capable of generating 10 PSI differential, it would subtract 5 psi from its inet, and add 5 psi to its differential. I know, the math would tend to make you believe that it would produce ZERO differential, but the OVERALL differential across the pump would still be 10 PSI.



    You can get the same effect by having the pump located halfway in the loop with one expansion tank. Split pressure differential.



    I have seen this on older systems that had a functional non cap tank in the attic, and a bladder or diaphragm tank in the basement.



    Kind of reminds me of the theoretical scenario of driving in a super hyper drive space car, going backwards at the speed of light, and you turn your headlights on. What happens to the beam of light? CLUNK? Brilliant darkness?



    I also had a demonstrator that we were going to use to demonstrate the effect of the PONPC to jurors in a major case that the defendants.lawyers kept claiming that the PONPC was bogus. After they saw the video of the working unit, they dropped that portion of their claim. It was REAL graphic :-)



    HTH



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
  • Gordy
    Gordy Member Posts: 9,546
    What if ?

    With the same scenerio they shared the same tank, instead of 2 seperate tanks?



    Gordy
  • Paul48
    Paul48 Member Posts: 4,469
    Mark

    "So if the pump is capable of generating 10 PSI differential, it would subtract 5 psi from its inet, and add 5 psi to its differential".

      Isn't that a characteristic of all running pumps in a closed system, regardless of the position, or even presence of an expansion tank? My understanding of pumping away was that it basically smoothes out momentary pressure drops on pump starts?
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    Same thing...

    Split head. The exact distance between connections and the eye of the impeller would have to be exactly the same. Otherwise there could be enough of a difference to actually have flow in the common connection.



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
  • TonyS
    TonyS Member Posts: 849
    A super hyper drive space car

    hmmm. I'm pretty sure that comes under the laws of relativity Mark. LOL.

    Kinda Lika fly in your car with the windows up doing a hundred miles an hour or throwing a baseball back and forth on a moving non accelerating train or why the machine guns don't shoot the wing off on a plane going 3 times the speed of sound.

    You cant go jogging at 3 miles an hour and then add in the rotational speed of the earth plus the expansion of the universe. I think your headlights will still work Mark but I cant prove this because my hyper drive space care is broken right now.:)
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    Paul,

    If there is no expansion tank in the system, then yes, it is split half and half (think GSHP loop pumps for example). If there is an expansion tank connection as close to the pumps inlet as possible, then the pump will ADD its whole pressure differential potential to the static fill pressure of the system. If the connection is further up stream, there will be a droop in pressure between the PONPC and the eye of the impeller. In certain cases, that droop can actually cause the pressure to go below the static fill, but it requires a BUNCH of conditions to actually cause it to happen.



    If the expansion tank connection is down stream of the pump, then it has no option but to subtract its differential from the static fill, which is where problems begin. If it is a high head pump, it can cause cavitation, and in certain cases, it can actually suck air IN to the system through an open auto vent on an upper floor.



    The only disadvantage of having the pump pumping away from the PONPC, is the possibility of having a high static fill requirement, and having a pump pumping towards the boiler, and the boiler having a high pressure drop, then there is a possibility of causing a relief valve to lift. But if properly sized, adjusted and applied, it can be controlled.



    In all honesty, in all of the years I have been teaching, explaining the PONPC is probably THE hardest subject for people to comprehend. It's not something that is real clear in the minds eye.



    Thanks for asking and learning.



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
  • SpeyFitter
    SpeyFitter Member Posts: 422
    edited July 2012
    Interesting Question

    What if we took it further.

    We if we installed 2 expansion tanks - 1 at the suction side of the circulator, the other at the discharge side. We have a static fill pressure of say 12 PSI just for giggles, we set the SUCTION expansion tank to an air pressure of 24 PSI and the DISCHARGE expansion tank to an air pressure of 12 PSI. Then perhaps add some boiler heat to make things interesting as things expand, etc. . Would flow occur?Not that this would ever happen but.....I guess we could ask hypothetical questions all day. Perhaps someone if they had the time could try it?
    Class 'A' Gas Fitter - Certified Hydronic Systems Designer - Journeyman Plumber
  • DanHolohan
    DanHolohan Member, Moderator, Administrator Posts: 16,513
    What Gil Carlson taught me.

    With two tanks, the circulator will establish a point of no pressure change somewhere between the two. I've been on jobs were an old-style radiator fills with air and acts as a second tank. You can see this on the pump gauges. It threw for a loop at first.



    And it doesn't matter what pressures are inside the tanks. It's the point of no pressure change because the circulator can't add or remove water from the tank.



    This stuff is endlessly interesting.
    Retired and loving it.
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    Scott...

    In my minds eye, the tank that is charged to 24 PSI will not be "Seen" by the system fluid, and hence will have no effect. The diaphragm of THAT tank would not be interfaced to the water, so its atmospheric influence will not be effective.



    The PONPC, is essentially THAT point where the atmosphere comes into contact with the fluid.



    In the case of an old gravity heating system with a tank in the attic, THAT is the PONPC.



    In the case of a diaphragm or bladder tank, where the fluid comes into contact with rubber bladder is the PONPC.



    In the case of a deep well, initially, when charging the pressure tank, deep down in the hole is the PONPC, until the pressure tank is charged, in which case it switches to the pressure tank.



    In the case of a city reservoir, the PONPC would be where the atmosphere interfaces to the fluid of the lake.



    My epiphany as it pertains to the PONPC came during a class I was teaching on hydronics. I made the claim that hydronics made more sense than forced air because the Earths surface was covered 75% with water. From the back of the class, a voice countered "Yeah, but the Earth is 100% surrounded by air!"



    He eventually became an instructor at the same college, teaching forced air piping.



    So, back to the epiphany, air pressure under static conditions is so powerful that in a closed loop system, the pump can not have any influence over it. If it weren't for air pressure, all the water would leave the face of the Earth (Gravity? Fugedabowdit...)



    So, now you know more than you wanted to about the PONPC. As Dan said, the late great Gil Carlson grasped this concept LONG before anyone else did, and we owe him a great deal of gratitude. Thanks Gil!



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
This discussion has been closed.