[Prev][Next][Index]

Re:Kevin Conlin's tank design

>Way back when, Paul Krombholz wrote:
>> Kevin Conlin <kcconlin at cae_ca> described in Digest VI #55:
>>>                           Aeration chamber
>>>                        |                    |
>>>          ---------------      Nutrient      ---------------
>>>          | _____________   -> Water Flow -> _____________ |
>>>          | |           |____________________|           | |
>>>   |      | |                               chamber water| |     |
>>>   |    -------                            level set here| |     |
>>>   |--- |     | ------- Water Line --------------------- | | ----|
>>>   |    |______                                          | |     |
>>>   |      | |   Powerhead                                | |     |
>>>   |----- | | ------------2" Gravel -------------------- | | ----|
>>>   |_ _ _ | |_ _ _ _ _ _ _ Fine Mesh _ _ _ _ _ _ _ _ _ _ | |_ _ _|
>>>   |                    2" Root Growth Area (no gravel)          |
>>>   |_____________________________________________________________|
>> 
>> The pathway of the nutrient water flow above the tank water level will have
>> to be completely air tight.  If air can get in, the weight of the water
>> column above the tank water line will force the nutrient solution up
>> through the fine mesh and through the gravel into the tank water.

> Am I missing something?  My EE brain figures that because the water flow
> up the riser tube on the left is exactly matched by the flow down the
> tube at the right, there won't be any mixing between the plenum and
> the main tank water (to a first approximation).  Water level in the
> aeration chamber (which must remain open for oxygenation) is set by
> the height of the outflow tube.

Oh.  I see now.  Your EE brain is right.  :-)  I thought it needed 
to be air-tight also, but not so:

    In a non-air-sealed system, the water level in the "down" 
    tube will *always* reach the same level as that in the main 
    body of the tank.

Dynamics summary:

(x) You don't need an air-tight system if you have a never-broken
    "siphon" pulling water up from from the plenum (like a
    powerhead in a lift tube with the powerhead below tank level).
    This appears to be your design.  A pump in the aeration
    chamber above water level would require an air-tight seal
    for the substrate circulation system (which does not appear
    to be your design).

(x) The water in the plenum on the LEFT side is under negative
    pressure because of the powerhead, and thus water is pulled
    up the tube.  Water in the main tank may be drawn down
    into the substrate by this localized vacuum, but encounters
    resistence moving through the substrate.

(x) The water can flow across the aeration chamber under
    relatively low net pressure and mix with air/nutrients,
    with the only significant force being the slight differential
    (slight drop caused by gravity) induced by a "high" water 
    level on the left side (caused by the powerhead), and a "low" 
    water level on the right side (caused by the hydraulic sink: 
    the right tube, where water "falls" into the tube to the water 
    level of the tank).

(x) In a non-air-sealed substrate circulation system (your design),
    the water in the "down" tube will *always* reach equilibrium
    at the main tank level.  Thus, net gravitational force should
    be zero, and you should (primarily, other then osmotic 
    gradients) circulate only through the substrate.  Likewise,
    net pressure will be zero for a sealed system where no air
    is exchanged through your substrate circulation.

(x) Of course, if greater evaporation occurs in the main tank,
    water will move from the substrate to compensate.  Likewise,
    if greater evaporation occurs in your aeration chamber,
    water will move from the tank into your plenum.  You will
    never have a water level in your down tube different from
    the water level in the tank, and you can add make-up water
    to the tank or to the aeration chamber to force temporary
    pressure gradients.

> [snip] the amount of mixing [between the tank and the plenum]
> should be negligible at low flow rates.

Ok, you're right.

--charley                            Fort Collins, Colorado USA
charleyb at gr_hp.com	or	charley at agrostis_nrel.colostate.edu