[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CO2 reactors

Dennis wrote:

> I'm not sure if the correct defination is reactor but what I'm looking for
> is a better pay of getting the CO2 disolved into the water.  My present
> system is to bubble the CO2 into the inlet of my Canaster filters.


> What are some of your ideas for creating something that works and is
> not expensive? 

There are a few DIY reactor designs in the Krib
(http://www.cco.caltech.edu/~aquaria/Krib).  You might read through that. 

Heck, the kids are still asleep and I've got time on my hands, so I'll 

You don't say why you're unhappy with the method you're using now, so 
it's hard to say how to improve it.  I think lots of other people on this 
list use the same method.

A reactor needs to do at least one of two things:  it needs to create a
surface between the gas and water and it needs to maintain contact between
the gas and the water for some period of time.  These are somewhat
offsetting needs.  If you can get a large surface area, then you need less
contact time.  If you can get a long contact time, then you need less
surface area. 

We usually create the large surface area by getting small CO2 bubbles. 
For instance, you can run the CO2 outlet into a power head or
over-the-back filter and let the pump disperse the gas into a lot of very
small bubbles.  I've "run the numbers" on this method and found that with
small enough bubbles the CO2 should be dissolved within a few seconds. 
The bubbles may not disappear in that amount of time, but the gas left in
the bubble after a short time contains very little CO2.  That's because
gases in the water are entering the bubble at the same time that the CO2
in the bubble is dissolving into the water. 

The fact that the bubbles don't always disappear can be annoying.  In the
tank where I use a power head for dispersing the CO2 there is a problem
late in the day when the plants are really bubbling.  It seems that oxygen
dissolves into the bubbles at a higher rate when the oxygen concentrations
are high and the bubbles stay noticable longer.  The remnant bubbles
(along with tiny oxygen bubbles coming off the plants) make the tank look
cloudy.  I don't have as much of a problem when I use an over-the-back
filter the same way, but I get less CO2 in solution, probably because the
surface turbulence at the outlet flushes some of the CO2 back out of the

The other extreme is the "bell" reactor; gas is held in an open-mouth
container inverted under water, usually with a continuous feed of CO2 into
the container.  The water and CO2 are in contact across the open end of
the container and the contact is maintained continuously.  This is a very
simple system that seems to work for some people.  The problem is that
dissolved gases from the water enter the gas in the bell and dilutes the
CO2.  The diluted CO2 dissolves into the water more slowly than pure CO2
and the gas mix builds up in the bell until it starts bubbling out. 

I experimented with this and ran some numbers.  I found that without using
a larger bell than I wanted I would always lose a lot of my CO2. 
Other reactors combine these approaches.  One design releases CO2 bubbles
into a downward-moving stream of water.  The bubbles provide a lot of
surface area and the downward moving stream keeps the bubbles in contact
with the water. 

Bubbling CO2 into the inlet of a canister filter combines both approaches
and seems to work for a lot of people.  I've read some complaints about
noise from the filters and a need to "burp" the canisters.  I think the
problem might be that gases from the water enter the bubbles, the
mixed-gas bubbles won't dissolve and a gas pocket builds up in the

Has anyone with that problem ever tried using much smaller bubbles?  It
seems like smaller bubbles would dissolve faster, with the remnants
flushing through the canister instead of building up in it. 

Just a thought.

Roger Miller