another reactor design conundrum

[mulm]

The intent is to use a "mixing pump" to help disolve the CO2 fully. The question is... How will this affect the fluid dynamics of the system? Pressure, flow, etc... Any takers?

CO2 H2O
|o| |~|
|v| |v|
|0| |~|
_|0|_|~|______________
|~|~~~~~~~~~~~~|
|~|~~~~~~~~~~~~|
|~|~~~~~~~~~~~~|______
|~|~~~~~~~~~~~~ < ~~~|
|~|~~~~~~~~~~~~____ ~|
|~|~~~~~~~~~~~~| | ~ |
|~|~~~~~~~~~~~~| | ~ |
|~|~~o~~~~~~~~~| | ^ |
|~|~~~o~~~~~~~~| | ^ |
|~|~~o~~~~~~~~~| | ^ |
|~|~~~O~~~~~~~~| | ^ |
|~|~~o~~~~~~~~~| | ^ |
|~|~O~~~~~~~~~~| | ^ |
|~O~~~~~~~~~~~~| | ^ |
|~~~~~~~~~~~~~~| | ^ |
|~~~~~~~~~~~~~~| | ~ |
|~~~~~~~~~~~~~~| | ~ |
|~~~~~~~~~~~~~~|__| ~ |______
|~~~~~~~~~~~~~~|***********|
|~~~~~~~~~~~~~~|***********|
|~~~~~~~~~~~~~~~~ >*PUMP***|
|~~~~~~~~~~~~~~~~ *********|
|~~~~~~~~~~~~~~|_____________|
|~~~~~~~~~~~~~~|
_~~_________________
|~|
|V|
H20

 

[scottward]

What's a 'mixing pump' ? ;-)

 

[mulm]

In my case it would likely be an Eheim 1048... but any external pump could work.

The idea is to augment the flow within the reactor to help mix the CO2 so to not rely solely on flow thru the system for dissolution of the CO2. The hope is that this will decouple CO2 injection efficiency (distribution within the tank asside) from filtration efficiency while keeping them inline and using the same "plumbing".

 

[scottward]

Ok, so what you mean is a pump *inside* the reactor itself (whilst there is still a seperate pump actually pushing water through the reactor).

 

[csmith]

Seems like overkill unless you're running a huge amount of bubbles into the reactor. Let physics do the work, keep it simple.

 

[mulm]

scottward-
Sorry...the ascii "drawing" is pretty bad. Unfortunately the forum strips out whitespace. The pump is outside of the reaction chamber and pumps water within from the bottom to the top. The inflow to the chamber would come from the return line from the filter/sump (in my case a 2215) thru the chamber to the tank.

csmith-
Suffice it to say that I am very familliar with the traditional reactor designs and other methods for injection. "Simple" or otherwise. Suppose the amount of bubbles is larger than the system's flow can dissolve alone and increased flow thru the system is not desired.

 

[nipat]

I think putting the pump at the bottom is undesirable because it risks pulling
bubbles down to near the exit. Some bubbles will prematurely escape the reactor.

I believe increasing flow through the system (but keep the current inside the system
at optimal level) is a key to better dissolving.

 

[mulm]

nipat-
Exiting bubbles is an issue with any increased flow and can be address with chamber length.


What other adverse effects might this have on the system? Will it increase the pressure within the system? Over pressurization? Decreased flow thru the system?

 

[nipat]

nipat-
Exiting bubbles is an issue with any increased flow and can be address with chamber length.

I don't think chamber length affects flow-rate in a significant way. Try imagining longer hose
and shorter hose (I'm not talking about 1 mile vs 1 foot hose, so let's say 1 foot vs 2 feet hose),
the difference is mainly caused by surface resistivity of the hose's surface.

But increasing the diameter does slow down the current in the chamber in a significant way:
http://www.1728.com/flowrate.htm

Flow rate = cross sectional area x velocity

 

[mulm]

nipat-
I think you are misunderstanding my reply. My statement about chamber length was in response to the concern about bubbles exiting before being disolved... That is...the longer the chamber the less likely this is to occur. This is a digression...

My interest is in what affect the "mixing pump" will have on the fluid dynamics of the system.

 

[nipat]

nipat-
I think you are misunderstanding my reply. My statement about chamber length was in response to the concern about bubbles exiting before being disolved... That is...the longer the chamber the less likely this is to occur. This is a digression...

My interest is in what affect the "mixing pump" will have on the fluid dynamics of the system.

Then I think it's the same efficiency as placing it (the mixing pump) on top as Tom's.
But of course, it's just a speculation.

 

[mulm]

I am fairly new to this site.... but isn't Tom's reactor internal and driven by a powerhead and venturi? Not following the connection.

 

[nipat]

http://www.barrreport.com/showthread.php/3444-Dual-venturi-DIY-External-CO2-reactor

His design doesn't increase flow but is about bubble recycling like yours.

 

[mulm]

Maybe my "drawing" is the problem here...

My design is an external inline reactor (very similar to the DYIs found all over the web; eg: Rex Grigg) with the addition of a pump that will be used to "mix" the water in the reaction chamber --NOT for pumping water into or thru the system. Placement of the pump is secondary in that its purpose is not to move water to and from the tank but to add hydraulic turbulence to the reaction chamber.

 

[pat w]

Your drawing was clear enough.

I've had similar thoughts on reactor design. By recirculation the CO2 rich water within the reactor itself you could increase the overall flow within the reactor without increasing the conventional flow through it. I didn't see any reason to be concerned with any increase in hydrostatic pressure as any increase would simply be relieved through the exit back to the tank. An addition I would suggest incorporating a tube to scavenge any and all gas collecting at the top and recycling it back through the recirc pump to thoroughly smash the CO2 into the smallest bubbles possible to maximize the effectiveness. If the inlet and outlet of the recirc pump were kept as close to the top as possible you could reduce the amount of mist entering the effluent back to the tank.

Pat

 

[shoggoth43]

This really isn't much different that the calcium or other chemical reactors. You run lots of water through the reactor, but the amount that goes in/out of the reactor is considerably less. I don't think you'll have a problem so long as you don't have really fine bubbles blasting out of it.

-
S

 

[mulm]

Your drawing was clear enough.

I've had similar thoughts on reactor design. By recirculation the CO2 rich water within the reactor itself you could increase the overall flow within the reactor without increasing the conventional flow through it. I didn't see any reason to be concerned with any increase in hydrostatic pressure as any increase would simply be relieved through the exit back to the tank. An addition I would suggest incorporating a tube to scavenge any and all gas collecting at the top and recycling it back through the recirc pump to thoroughly smash the CO2 into the smallest bubbles possible to maximize the effectiveness. If the inlet and outlet of the recirc pump were kept as close to the top as possible you could reduce the amount of mist entering the effluent back to the tank.

Pat

Yes...I have the same thoughts about the pressure and the use of accumulated gas at the top of the chamber. As for the placement, I am have concerns about running the pump dry in the event of a major dump of gas... though still considering.

This really isn't much different that the calcium or other chemical reactors. You run lots of water through the reactor, but the amount that goes in/out of the reactor is considerably less. I don't think you'll have a problem so long as you don't have really fine bubbles blasting out of it.

-
S

That's right...except that in this design the flow in and out of the reactor is much greater than that of a Ca reactor...and is in fact the main flow from the filter. The hope is that the CO2 dissolution and filtration (flow) can happen independantly yet within the same plumbing.

Thank you both for your input.

-Kevin

 

[pat w]

As for the placement, I am have concerns about running the pump dry in the event of a major dump of gas... though still considering.

-Kevin

Based on my experience with a similar design of about 2' tall I'd say you'd be safe as long as the inlet to the pump were at least 4" down from the top of the reactor. I never saw a build up that extended nearly that deep, usually about 1/2" - 1" tops.

Pat