[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE Dual stage regulators and CO 2 Setups
Julie E. Martin wrote concerning the equipment needed for CO2.
James Purchase wrote an excellent reply. The following supplemental
info might prove helpful.
If you haven't already, check out Erik's excellent and comprehensive
site at www.thekrib.com.
Also, George Booth's site is worth reading at
Summary: You need a tank with a valve, a pressure regulator, some
fittings and tubing, a metering valve, a backflow checkvalve if you're
the nervous type, and a device to disperse or inject the gas. Good
places to look are local welding parts suppliers (yellow pages),
McMaster-Carr, and Harbor Freight Tools.
CO2 Tank: The tank should be a CO2 tank. This will be a steel tank
that meets DOT 3AA2015. This number will be stamped on the tank along
with a boatload of other numbers that tell you the manufacturer, the
date the tank was first tested to meet DOT standards, and other things.
(Just for your info, the "3AA" means "steel seamless tank" and the
"2015" means the tank meets DOT requirements for holding Industrial gas
up to 2015 psi service pressure.) You can get one from a local
supplier of Welding Accessories (check the yellow pages) for about $80
to $120 for a 12 pound tank. A twelve pound tank (that is, 12 pounds
weight of an empty steel tank) can be filled with 5 pounds of CO2,
which brings the tank pressure up to about 850 to 950 pounds at room
temperature. Although a 3AA2015 tank can hold much more gas, DOT
regulations do not allow more than 5 pounds in a 12-pound tank. A
source for a decent tank is www.harborfreight.com but it will tank
about 3 - 7 weeks for free delivery
. This is a good tank at a good price. Some home beer brewery parts
suppliers also sell tanks. Some scuba shops also sell tanks but these
usually have an oxygen valve on them which will need to be replaced or
adapted with fittings to connect to a CO2 type regulator.
Tanks almost always come with the tank valve on the tank. The
different kinds of valves have different numbers. CGA580 is a female
thread fitting intended for Argon gas. GCA320 is a male threaded valve
which is a standard for CO2. (Oxygen tank valves have reverse threads
to prevent the patient ever getting the wrong gas!) Which valve is on
the tank determines what fitting might be need to attach a regulator to
the valve. Expect a tank called a CO2 tank to have a CGA320 valve on
The regulator should have a high pressure and a low pressure gauge if
you want to know what's going on in the CO2 tank and in the aquarium.
The low pressure side should be able to read in the neighborhood of
0-30 or 0-60 psi. The high side needs to be able to read up to about
1,000-2,000 psi. Valves are not hard to find; inexpensive ones are
hard to find. Any welding parts supplier will sell a valve capable of
handling CO2, which welders would call a noncorrosive gas. Victors are
probably the most expensive. You can get a really nice two-stage one
(Victor model Vts253b-320) for about $180 or a really nice single stage
(Victor model Sr253b-320) for about $120 by submitting a request
through www.mygeek.com or check a local welding parts
supplier that carries Victor. Other brands are usually cheaper. OR
you can get a really cheap one from www.harborfreight.com for about
$30. I have used these and they are kind of sloppy -- some don't hold
a setting very well -- some better than others -- for the price
difference they might be worth a try for you -- I am using them on two
tanks right now.
Fittings: You need to get your regulator to find out what fittings you
need. The regulator will thread onto the CGA320 valve. The regulator
will have an output opening for a 1/8" or 3/16" or 1/4" fitting. IF
you are going from the regulator directly to the tubing to the tank,
then you need a fitting that is threaded on one side and barbed on the
other for a tube to be slipped onto it.
The harborfreight regulator have this fitting on it
). If you are going to go from the regulator to a metering valve and
then to the tubing then you need fittings that will connect the
metering valve to the regulator. Which fittings depends on the
regulator and the metering valve. If they are both 1/4", for example,
and match up as male vs female threads, no extra fitting is needed.
The most comprehensive source for hunting down fittings is
http://www.mcmaster.com/ . Look for brass fittings at their
site-it has excellent search functions.
Metering valve: If you want to control the flow of CO2, the regulator
might not be precise enough. If so, then you'll want a metering valve.
www.grainger.com sells Nupro but I recommend against them. I found
them to be too cheaply constructed and too sloppy to be reliable.
Swagelok has a number of different valves that are very well made. To
find out about these valves, Select Metering Valves from the left menu
on this page (
then Select M series or S series from the middle menu. Then select M
Series or S series from the rightmost menu.
Also, you can find a dealer by checking out that site. A decent
metering valve is expensive and might not be necessary with your
Tubing: I haven't heard of anyone with experience that thinks regular
airline tubing isn't adequate for CO2 injection except people selling
"special" tubing. Use mini or regular tubing, what fits your injector
or barbed fitting or vice versa.
Injector device: You can use an air stone, but it probably won't be
efficient. You can use a special airstone, called a CO2 diffuser,
which is more efficient but expensive. Eheim makes one that comes with
a backflow check valve and minitubing and also serves as a bubble
counter. www.thatpetplace.com sells the eheim for about $30, a good
deal on that particular item. I find that the diffusers get gunked up
quickly and need to be cleaned often. Worse, I found them to be
inefficient, although others with more CO2 experience are happy with
Here's what I found:
Experience with A Glass-disk CO2 Diffuser and Tube-type Mixing Chamber
Summary of CO2 Injection Rates for Equal Absorption Levels
Shown by Method
Method 1 Eheim Diffuser at Bottom of Tank Bubble rate=180/minute
Method 2 Eheim Diffuser at Bottom of Tank &
Directly in line with Powerhead Outflow Bubble rate=80/min
Method 3 Gravel Tube Mixing Chamber Bubble rate=45/min
The table above summarizes the absorption results with three different
methods of injection.
This tank had been established for about 6 months. It was not
intended as a planted tank and had two biowheels, the space-saving
alternative to trickle filters. The biowheels promote gas exchange,
very good for oxygenating the water and encouraging biofiltration but
probably contraindicated for CO2 injection (in comes the bad air, out
goes the good).
But I decided to leave them on to see if I could get reasonably good
results. I had read
( http://www.thekrib.com/Plants/CO2/co2-loss.html#0 ) that CO2
injection can work on tanks with trickle filters if the drip chamber is
reasonably well sealed – the atmosphere in the chamber becomes
CO2-enriched which limits the mount of CO2 escaping from the water. So,
as a temporary measure, I wrapped the biowheel boxes with Reynolds
Plastic Wrap. If that worked, I could devise a more permanent “wrapper”
later. This is essentially an enclosed tank except for the openings
necessary for the biowheels returns, and heater and powerhead cords.
I first installed an Eheim CO2 diffuser, placing it against one side
panel at the top of the substrate. The Eheim is really just an
expensive, very fine grain glass air-stone, with a bubble(counter)
chamber on the bottom and it comes with a back-flow check valve. For
about twenty bucks, it’s a relatively inexpensive package. And Eheim
canister filters work so reliably, I figured this product was a safe
The water in my area is very soft and near neutral – GH & KH about
1-2, pH about 7.2. I pushed the KH to 5 with sodium bicarbonate and
started the CO2 injection, aiming for a level of 15ppm. I measured KH
and pH, every few hours the first day, and once per day thereafter for
about a week. After that, I measured about once or twice a week. Since
the resolution of pH tests leaves a wide margin of “error” when
computing CO2 levels, I also used a Hach CO2 test for the once per week
testing to confirm the KH & pH measurement. Each week I changed 20% of
I ended up having to set the CO2 injection rate to 180 bubbles/minute
to achieve a CO2 level of about 15ppm. Awful results! I relocated the
diffuser at the substrate directly under a biowheel powerhead outlet –
this outlet was at a t-joint that diverts some of the powerhead output
directly back to the tank water column, bypassing the biowheel. I
directed the powerhead flow downward at about 60 degrees so that the
bubbles from the diffuser would be forced downward to circulate around
the tank before reaching the surface.
The powerhead is rated at 170 gallons/hour. Powerhead ratings generally
assume 0" of head (i.e., no static pressure), so the actual flow rate
in use probably half or about 80-90 gph. Assuming the t-joint splits
the flow roughly in half, the flow that bypasses the biowheel is about
40-45 gph. I repeated the original testing sequence: every few hours;
every day; then every week.
Performance was substantially improved. I was able to turn the CO2
back to about 80 bubbles/minute to get a CO2 level of about 15ppm. Not
bad but not very good either. Also, it seemed that biofilm would build
up on the diffuser quickly, causing it emit larger, less efficient
bubbles. It needed cleaning every couple of weeks.
I thought about scrapping the biowheels to improve things, but I
want to change over the filtration system if I didn’t have to. I had an
old Python gravel tube which many folks have suggested as DIY reactor
(http://www.frii.com/~booth/AquaticConcepts/). With a small flame I
curled a piece of rigid airline tubing into an elongated J-shape --
about 10" long on one side and about 5" long on the other. I attached
it to the 10" gravel tube with plastic wire-ties so that it would feed
CO2 into the gravel tube from the bottom, reaching about halfway up the
gravel tube. Then I attached the unit to the powerhead outlet with a
piece of 1/2" tubing. I hoped the roughly 40 gph powerhead output would
be enough to absorb, and carry into the water column, sufficient
amounts of CO2. In fact, as the top 1/2" or so of the tube filled with
gas, and the water turbulence created a swirl of bubbles in the tube,
the smallest pinpoint bubbles were washed out of the tube, so even less
water flow would be acceptable. Even with an estimated water flow of
only 40-45 gph, I was able to get a 20ppm CO2 level with a setting of
45 bubbles/min. Although the gravel tube unit also gets coated with
biofilm and some algae, as did the Eheim diffuser, it happens slowly
and does not affect the performance of the reactor.
If I replaced the biowheels with a canister filter, and placed the
canister outlet well below the tank water surface, I could probably get
substantially better results. For now, I am happy to have gotten the
CO2 rate below 1 bubble/sec.
Obviously, I recommend getting a python water changer/siphon tube. I
bought two Eheims for my tanks and ended up using the ckeckvalves and
tubing and tossing the rest. But you probably can get check valves and
tubing for less than $28 a set.
CO2 gas was not as easy for me to find as I thought it would be. A
search through the yellow pages and a few calls should find one or more
of the following that will fill a tank:
--a beverage or beverage-dispenser parts supplier;
--a local welding shop;
--a fire extinguisher refill store.
5 pounds of CO2 might cost anywhere for about $10 to $20, so shop
This is probably more than you wanted to know but not quite what you
were hoping for. But I hope it helps
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger