Aquatic Plant Forum banner

Is a power CO2 reactor better than a simple bell diffuser?

14640 Views 13 Replies 7 Participants Last post by  plantbrain
We've all seen them- a cylinder with a powerhead hooked up to it that generates turbulance and "theoretically" increases the diffusion of CO2 into aquarium water. On the flip side is the good, old bell diffuser- a simple bell-shaped object into which CO2 is bubbled into.

Is one truly better than the other?

Does turbulance really increase diffusion?
1 - 14 of 14 Posts
it is all about surface area. With a bell diffuser, you have a MUCH MUCH smaller surface area of CO2 than with a bunch of tiny bubbles. Also, with a reactor, you constantly have fresh water to saturate with CO2. With a bell diffuser, the water across the CO2 is an order of magnitude or more less.

I think it is a no contest comparison. The reactor will kill the bell diffuser in efficiency.
In my 15g I'm using a used Mini-Vortex I got from a certain someone here ;). It does a pretty good job of diffusing the CO2 into the water, with absolutely no visible bubbles coming out of the bottom and floating to the surface.

I actually have a pretty interesting idea on paper for a little somethin'... ;)

A bell difusser sitting in the aquarium would have aquarium water passing by it constantly. I guess you can view it as one large bubble.

I understand that you are saying that the more surface area created with CO2 (which many tiny CO2 bubbles would have more than one single large CO2 bubble) would allow more CO2 to diffuse into the water column or at least at a faster rate.

Certainly nitrogen and oxygen would diffuse out of the water and into the large bubble in the bell diffuser thereby decreasing the CO2 content. Perhaps this also creates inefficiency.

However, why would a power reactor be any more efficient then a scintered glass diffuser sitting at the bottom of the aquarium creating tiny bubbles then?

Also, as for no bubbles escaping from reactors, how can that be if nitrogen and oxygen (barring other gases) would replace the CO2 in the bubbles? Wouldn't there have to be bubbles remaining but just bubbles filled with nitrogen and oxygen?
I agree with Tony that a bell diffuser is very inefficient.

However I don't understand why so many people trash the sintered glass diffusers as compared to a little 12 long reactor. They claim that since the reactor is in a closed system 100% of the co2 goes into the water, while with a sintered glass diffuser the co2 bubbles just get released into the air.

I have compared sintered glass diffusers against reactors and the former come out very favorable. Two Eheim diffusers are much more effective than my 3 inch diameter 3 foot long reactor for my 125 gallon tank. What people don't understand is that the bubbles coming out of a diffuser are smaller and have a total surface area as much as 10 times greater than a large bubble fed to a reactor with a simple nozzle fitting for its CO2 feed. Also as the bubbles rise I would think they would get bigger due to less pressure. They might also get bigger due to absorption of O and N.

Right know I estimate that one Eheim diffuser is about 75% as effective as the large reactor referenced above. Two of the Eheims would blow the doors off of the reactor.

One thing I must try is Tom Barr's recommendation of feeding the CO2 directly into the input of a filter, where the pressure is less and the co2 should be absorbed easier.

Steve Pituch
See less See more

Why would there be lower pressure in a filter? I would imagine that a pretty good differential in pressure would be necessary to result in any measurable difference in CO2 diffusion.

Also, would decreased pressure increase CO2 diffusion or would an increase be needed?
Well I am talking about putting the CO2 into the "suction" line of the filter. Not into the filter itself. In fact I believe that if you have DIY CO2, you can probably feed the co2 into the inlet line because of the suction. Imagine trying to feed DIY CO2 into the outlet line which is under additional pressure (it would probably be a disaster). I think this is what T.B. recommends for pressurized co2 systems (suction side). Most newer canisters have been redesigned (improved) to continue to function with gas bubbles in the water. They used to lock up with air in the line. He mentions easier absoption of co2 into the water this way.

Maybe he will chime in.

There was just a post on this subject at:
Perhaps it should be moved to here.


Thanks. I still don't understand the pressure in filter thing. Perhaps Tom will chime in. I can certainly understand bubbling CO2 into the intake where the bubble will get held up inside the filter thereby providing time for CO2 to diffuse into the water.

As for the difference between reactors and diffusers, the only difference I can see is contact time. I am assuming that a tiny bubble created by a diffuser will not have enough time to diffuse all of the CO2 out if it by the time it reaches the surface and degasses into the atmosphere.

I am also assuming that the CO2 bubble in a reactor is trapped until all CO2 diffused. However, if no bubbles exit the reactor, I don't know where the N and O bubbles would go?
It is not the turbulance per say, it is the length of time the gas remains in contact with the water. In a chamber with a jet stream, the bubles remain trapped until they are dissolved. In a bell, the bubbles may sit there a while, but can be pushed out from the bell when you try and raise the amount of gas...know what I mean? The bell can only handle a very low flow rate. A reactor can handle a very high flow rate. Other types of diffusors like the Eheim are glorified air stones and a total waste of money. They become clogged and dirty easily and quickly and no matter how tiny the bubbles may be, most simply rise to the waer surface and are disapateded.

I think the bells date back to Europe when it was thought all you needed was a very small amount of C02 in your aquarium, like 10ppm max.
Diffusers create tiny bubbles that, as Steve mentioned, have MUCH more surface area than bubbles typically created by reactors. These 100% CO2 bubbles will immediately begin diffusing CO2 because of the lower CO2 content in the water column. Likewise, N and O2 will diffuse into the bubbles because of their lack in the bubble. When the bubble reaches the water surface, it will be mostly, if not completely, N, O2 and other gases. The CO2 will have diffused into the water.

In my experience, I have not found power reactors to be more efficient than a good diffuser. Yes, they trap the bubbles until all CO2 is diffused, but is that really necessary? Are we so sure that a tiny CO2 bubble created by the diffuser doesn't release 100% of its CO2 also? Perhaps even quicker?

Power Reactors are bulky, unsightly and require a powerhead to work. The addition of another electrical gadget to me is not worth it.

Diffusers can be well hidden in the aquarium and are easy to clean. I simply dip it for a few minutes in a water/bleach solution for a few minutes every month. Not much to it.
See less See more
I have the end all statement :-D

Bell diffusers take up space in your tank while inline reactors don't

End of story *LOL*

Bell diffuser yes, but a small diffuser like the one I have on auction is virtually invisible.
I have that reactor 500 in my tank. I know its rated for 500 gallons and my tank is only 50. But i can get my tank to over 100ppm. EASY. I add a bubble per second to get to 35-40ppm. I have heard people adding way way more bps in a similar tank. My tank is extremely densly planted. So i would have to say a bell diffuser is not as effective. none of the bubbles i put into my tank ever go into the atmosphere. my 20 has a bell diffuser and the bubbles reach the surface. I dont understand how something where bubbles are wasted is more effecient?
The glass diffusers are orginally filters design to filter chemicals.
The filter pads are placed on top and the vacuum is applied.

These a very fine sizes, they will clog and hence have some variation in their ability to diffuse if not cared for.

To improve this: place in a deep, dark corn and with good current that blows into a stand of plants at a slight downward angle. Clean at least monthly.

Diffuser disc and Reactors can be the same efficiency, a disc cannot be placed in line. If you place a reactor in line then you no longer require a power head etc or anything that is not already in the tank, I fail to see the arguement with that issue.

It can be avoided if the hobbyist choses to.
If you want to throttle the CO2 on/off, a solenoid can be had for about the same cost as a powerhead.

The easiest types of CO2 reactors are similar to the Krib's design with the gravel vac. I modeled mine after a Dupla Reactor.

You should be able to make a decent working one for 10$ or so and the cost of a powerhead.

You can make them self leveling also rather easily as well.
Simply drill a hole 3/16" at the preset level inside the tube you want the reactor to sart wasting the Gas or the "burp" level when the reactor is shut up and you still have gas entering(say if you use DIY yeast or have no solenoid).

I use a the powerhead's venturi to suck the gas after it builds up to a preset set level and then the reactor starts to grind the gas up and spits out a very fine mist similar to that of the Disc diffusers.

I'll typically use a 6"L x 2"OD tube and put the venturi hole at about 1.5" down from the top and the burp hole about 2" down.

As far as what Steve was referring to:
The negative pressure on the suction side of a pump will pull the bubbles in. When the powerhead is "off", these bubbles will escape out to the surface rather than go into the impeller and gas lock the powerhead.

The suction side also atomizes the bubble very effectively by pulling it/stretching it out and then it hits to the impeller blades and shreds it.

The Turbofloator venturi Skimmer does the same thing.

Speaking of venturis, they work great for gas diffusion also.

Tom Barr
See less See more
1 - 14 of 14 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.