I’m not sure that is a logical conclusion. Whilst your reactor managed. 1.5 drop for your tank, you can’t conclude that that a tank with twice the surface area would require a reactor with twice the surface area because we do not conclusively know whether it is tank surface area or tank volume that that determines the drop in relation reactor surface area.
We can be pretty confident that my assumptions are good enough, and we don't need more than an rough estimation as to what would be the minimum dimensions for a given tank to achieve 1-1.5 pH drop. The logic is that we discuss steady state, i.e the CO2 in the water column is constant. This means that injected CO2 will either be consumed by plants, or outgassed at the surface. Now the plant consumption is just a minor part, typically 5-10%, so for the purpose of this estimation we may assume that in the steady state the CO2 injection balances with the outgassing at the surface. Take then for example a twice larger tank surface, means twice more outgassing, requires twice more injection and thus twice the surface area of the reactor. I hope this clarifies it.
But those assumptions do not hold when we consider that a certain volume of water can hold a certain amount of CO2 and we therefore need to be able to provide that to even get our 1-1.5 ph drop in the first place.
Check
this should work. If you can, try to be on the safe side and take a bit extra reactor length for a stronger ramp up (don't know your tank volume, and don't know your surface agitation).If I understand your argument correctly - assume that under steady state the total water column will hold a constant mass of CO2, there is no CO2 added or taken out (otherwise pH would still change and we have no steady state). What we inject, in fact an equivalent volume of CO2, will therefore not be added to the water, but go straight to the plants and to outgassing. Just think of it as a CO2 molecule accountant
It is an estimation with a practical objective - design a working reactor with roughly correct dimensions. Indeed we can apply more subtle math, but it would not really be helpful and be merely confusing.
The outgassing at the surface is a function of the CO2 ppm in the tank, the surface area and the surface agitation. It is NOT dependent on the tank volume/depth. Therefore for steady state operation (same CO2 ppm, same surface area and same surface agitation in both tanks) we will need exactly the same performance from our reactor, with exactly the same injection rate to compensate for the outgassing CO2 losses.
I've been questioning myself what the appropriate CO2 shut off time I should be. I'm currently having it turn off 1 hour before lights off. Based on your post, I'm wondering if I should reduce the CO2 off time to say 30 minutes to light off. Thoughts?
This is a recommendable post @Hanuman , as usual your great building skills and well documented. I admit, too often when I work on my tank I consider how @Hanuman would do it
We’ve seen so many posts on many variation of bubble reactors in the past 20-30 years or so, I guess in the thousands, and yours is probably not the last. It’s not that dissolving CO2 in water is so difficult, or that it is a real technical challenge…
I think it’s probably bad forum etiquette to use your post only to promote an alternative, so I’ll make sure that I give some detailed replies on your post, even though I would never again consider building or optimising a bubble reactor. With lots of data and arguments I have had limited success gaining traction for the Horizontal Reactor on this forum, so it is probably foolish to expect that more scientific arguments and calculations will make any change … so YESSSS that’s exactly what I’ll do
Let’s start with a thought experiment. We take your bubble reactor, turn it 90 degrees into a horizontal position. This is then a Horizontal CO2 Reactor , no bubbles anymore - will that make things better or worse?
So let's do some simplified estimations, not exact science but just enough to get a feeling and a sense why Horizontal CO2 Reactor works.
View attachment 205743
When the reactor is horizontal, upper half filled with CO2 and lower half with a flow of water, the surface area for CO2 absorption is 25.806 mm2. If we assume for a moment, if we're to have a vertical reactor, that our bubbles are 3 mm diameter, each bubble represents a 28 mm2 surface area for CO2 absorption to the water. So we calculate that we would need 913 bubbles to get to the same CO2 absorbing area as the horizontal reactor would do. When I took slow motion videos of my bubble reactor, it did not even come close to this number, as anyone may confirm in their reactors.
The objection to this simplified calculation may be that in reality we do not have all same size 3 mm bubbles, but some smaller and some bigger. If we would redo the above calculations, we find:
So for any mix if above bubbles sizes (if we feel that that’s closer to reality), we would need many hundreds of bubbles – much more than we can observe in our real life reactors. Therefore it is safe to assume that for any given pipe size, the horizontal reactor variant packs much more punch than the vertical bubble variant. I won’t duplicate the other advantages of this simplified reactor method, they can be found in the thread, but will mention some below.
Please do not take my remarks as addressed to you personally, because indeed you do not claim you ‘own’ the ideas from what you have posted, and just post your implementation for the benefit of others.
Copying your reactor makes sense for a similar size tank, water flow from pump, etcetera. I have never seen posts on any forum that prescribes what are the correct dimensions for any bubble reactor, as a function of the user’s setup This is probably also not possible, as there are too many unpredictable variables at play in a bubble reactor, so that it is really hard to make a reliable model for its operation and performance. The horizontal reactor is much simpler to understand and model, and that’s why I’ve posted an estimation how to calculate the correct dimensions for any tank size.
When we focus on what we want to achieve (no noise, 100% efficiency, no bubbles in tank), and follow the Optimisation procedure that I used in my AquaMedic reactor, we see why in my personal view a by pass is highly desirable. I would personally never build a reactor without by pass, because you miss the optimization option and you get either bubbles escaping or noise if and when CO2 / water flows are not matched.
You said it
CO2 pressure will be at equilibrium with local water pressure, and therefore always very close to atmospheric pressure.
CO2 build up in the chamber is the main cause of water splashing and making noise. I would slowly increase the water flow at any given CO2 injection, using bypass valve, until this CO2 pocket disappears and the reactor is perfectly silent. If it appears that we then have bubbles escaping from the bottom, we learn that we have exceeded that maximum reactor capacity and it is better to reduce CO2 flow.
You are correct that a CO2 pocket in the top of the reactor also gives a reduction of flow in the system. It is worth mentioning (I did not before as I found the physics too confusing to mention) that the bubbles in a reactor will give a back pressure to the pump and reduce the flow to the tank. This is the same physics as airlift pump, and also explained in the youtube that I attach below. The Horizontal Reactor will not give any backpressure, as there are no bubbles pushing water against the flow and no CO2 pocket where we could create a noisy and energy wasting waterfall.
I believe the Niloc reactor is the best commercially available, but pricey, but still some room for improvement IMO, especially in the top where I worry a bit about noise.
For a Horizontal CO2 Reactor any plastic pipe from the local plumbing store will be good, transparent is just nice to have, as there is really no need to observe what’s going on inside.
I modified my AquaMedic reactor so that I could have automatic purging of any trapped air. I am mean enough to leave that as a little innovation puzzle, and am happy to share my solution in a PM
Horizontal CO2 Reactor also purges itself of trapped air, the valve is just an optional nice to have.
---------------------------------------------------------- // -----------------------------------------------
Youtube videos to illustrate physics of backpressure in bubble reactor:
I think this bit I can answer. If I go this right, you don't actually need an additional pump since this is a passive system and only relies on your spray bar or out-flow flow which is positionned on the opposite side of the CO2 horizontal reactor.
I was about to comment. Indeed you are right.
I don't prescribe any pipe length, or diameter. I even don't prescribe that a pipe should be used. When you follow the design calculations in the first post (I think I should repost a more concise and eloquent version that I wrote for another forum), the only prescription is the reactors CO2/water surface. Once you know the surface area, you could do that in a circular box (like the FX4 filter I posted earlier), a rectangular kitchen Tupperware box (for small tanks), a pipe where the product length*diameter gives the correct surface, or perhaps a combination of pipes as also @Unexpected has done.
The reason for vertical reactor is that nobody ever thought that we can make decent reactors without bubbles. As my calculations demonstrate, a horizontal reactor will be more compact than a vertical reactor, and besides that have many advantages that the bubble reactor does not have.
You just include a horizontal large diameter pipe in your circuit, with a gentle river of water. The pressure drop will be negligible, certainly compared to bubble reactors. It is in fact just some extra length of tubing.
When we know the necessary surface area, we don't care about complexity of design. Just use a Tupperware box if your like, and forget the internal structure.
In conclusion, I don't see any form factor problem at all. Happy to further discuss this if my remarks were not clear, or you don't buy it. Happy to do calculations for your tank, and this should be really straightforward. For demonstration of the concept, it could be nice if you post what bubble reactor is going to give you guaranteed success for you larger tank, and then we can then compare that to the Horizontal Reactor in a 'beauty contest' of designs.
You are correct, but there is little support required, added value if you like, from commercial solutions as we are using the simplest of the simplest.
Opening and cleaning of the reactor seems not necessary to me, as I wrote you could fill it with bioballs and get free of charge a bio filter capacity. Why would you want to clean it? If you insist, then buy end pieces with thread so that you can screw.
Correct
Check.
Correct. We don't need strictly laminar flow (that is a confusion that goes back to my imperfect description of CO2 Spray Bar a year ago). Just a gentle river of water, nearly everything will do from that perspective. You just don't want a too powerful pump on it that turns the pipe into a water splashing, noisy, bubble ejecting washing machine. So I do recommend a bypass, as a gentle flow will make operation easiest.
The short answer is perhaps Unexpected tank picture
When you start with the situation that the closed pipe or box is 100% filled with water, the situation is quite clear. Water is pushed out from the exit, and if the box or pipe is big there will be very little pressure drop or flow reduction.
There will be a flow rate reduction, anything that is added will have that effect. But it is going to be negligible. Probably not measurable.
