Horizontal CO2 Reactor - Yugang 鱼缸 Reactor

[JohnLaPaz]

Now at home, I checked my notes and double checked my reactor, I am now at a ratio of 40 to be more precise.

Drop checker lime green, but until I decide to buy a new pH probe I can't tell precisely what my pH drop is. As far as CO2 is concerned, I have gone minimalistic, as I don't see the point (at least for my own tank) of doing more measurements and my tank is more stable than ever. Using overflow mode, I am happy with my pH drop, whatever it exactly may be, and my motto for CO2 is now "launch and forget" - no need to measure, monitor or adjust anymore and just I focus on the other things that are important for the tank.

My advice would be, when building the reactor for overflow mode, anticipate that you may want to disassemble to adjust the size a bit until you're perfectly happy with your pH drop. Obviously the ratio's that I estimated (like the original 17.7 and now my current 40) are all based on my tank and other tanks may have some different surface agitation and outgassing and the ratio's may be different from mine. The reason that nobody complained about the 17.7 so far is that it gives a reactor that is for nearly every tank overpowered and when CO2 is regulated with needle valve or controller always good enough for any targeted pH drop. I use overflow mode, and then the reactor geometry will be the sole driver of CO2 injection.

In summary @JohnLaPaz if you use precision CO2 regulator or pH controller you may want a really strong reactor for the fastest ramp up and you may want to stick with the 17.7 ratio. In overflow mode, you may want to have a less powerful reactor so that fish are less stressed at pH drops exceeding 1.5.

My KH is around 2 to 3, depending on the season in HK.

Thanks for all the details !

 

[Unexpected]

Sorry it took so long to post this and I figure posting here would get other's opinions. My question to you was, will there be enough pressure inside the reactor for this to work? For those who have already built a reactor off the 17.7, and want to reduce the power of the reactor without shortening or rebuilding, having an adjustable overflow mode would be a nice feature like the CO2 Spray Bar.

I've seen these rubber, donut like, gaskets (no idea what they are called) where tubes can be inserted into pipes and create a tight seal. The pipe could then be adjusted up or down to change the water level inside the reactor. This probably is more for those with sumps or AIO tanks so water could be returned back to the system without tinkering too much.

So, would there be enough pressure? @Yugang

 

[Yugang]

Sorry it took so long to post this and I figure posting here would get other's opinions. My question to you was, will there be enough pressure inside the reactor for this to work? For those who have already built a reactor off the 17.7, and want to reduce the power of the reactor without shortening or rebuilding, having an adjustable overflow mode would be a nice feature like the CO2 Spray Bar.

I've seen these rubber, donut like, gaskets (no idea what they are called) where tubes can be inserted into pipes and create a tight seal. The pipe could then be adjusted up or down to change the water level inside the reactor. This probably is more for those with sumps or AIO tanks so water could be returned back to the system without tinkering too much.

So, would there be enough pressure? @Yugang
View attachment 3680

It's an interesting idea, and if you do it exactly as in your diagram I believe it could work.

The pressure in the reactor is very low, and depends how low the reactor is below the tanks water surface. It is this hydrostatic pressure that counts -- as the reactor has open inflows and outflows the water pump and CO2 injection will add very little pressure to this hydrostatic pressure.

With the tube as in your diagram (it ends below the reactor) it will leak some CO2 out if there is too much CO2 in the reactor, or it will drip water if the reactor water surface is above the top of the tube. So your logic works, as long as you're willing to find a solution for the water leak - perhaps the sump as you mention.

When the tube is connected to a flexible tube that goes up to the water surface of the tank, in an attempt to avoid water dripping from the system, obviously it does not work anymore. The hydrostatic water pressure in that return tube (when filled with water) will make it hard or impossible to the CO2 to escape -- there is not enough overpressure to blow the water out for CO2 to escape.

An alternative idea would be to use an overflow as you propose in your diagram, but then leading up instead of down. Perhaps with the tube ending just above the tank's water surface. This may work a little better (and any water drip ends up in the tank), but I would not expect it works well as the overpressure (real pressure minus hydrostatic pressure) is probably too limited for this to work reliably.

The simplest way that I have thought of to reduce the reactor power (apart for cutting it shorter or the styrofoam that I mentioned earlier) is to tilt the reactor a bit from its horizontal levelling, so that CO2 bubbles escape earlier from the water outflow tube than they would in the perfect horizontal position.

 

[Unexpected]

It's an interesting idea, and if you do it exactly as in your diagram I believe it could work.

The pressure in the reactor is very low, and depends how low the reactor is below the tanks water surface. It is this hydrostatic pressure that counts -- as the reactor has open inflows and outflows the water pump and CO2 injection will add very little pressure to this hydrostatic pressure.

With the tube as in your diagram (it ends below the reactor) it will leak some CO2 out if there is too much CO2 in the reactor, or it will drip water if the reactor water surface is above the top of the tube. So your logic works, as long as you're willing to find a solution for the water leak - perhaps the sump as you mention.

When the tube is connected to a flexible tube that goes up to the water surface of the tank, in an attempt to avoid water dripping from the system, obviously it does not work anymore. The hydrostatic water pressure in that return tube (when filled with water) will make it hard or impossible to the CO2 to escape -- there is not enough overpressure to blow the water out for CO2 to escape.

An alternative idea would be to use an overflow as you propose in your diagram, but then leading up instead of down. Perhaps with the tube ending just above the tank's water surface. This may work a little better (and any water drip ends up in the tank), but I would not expect it works well as the overpressure (real pressure minus hydrostatic pressure) is probably too limited for this to work reliably.

The simplest way that I have thought of to reduce the reactor power (apart for cutting it shorter or the styrofoam that I mentioned earlier) is to tilt the reactor a bit from its horizontal levelling, so that CO2 bubbles escape earlier from the water outflow tube than they would in the perfect horizontal position.

Ugh, I completely forgot about tilting it. I’m an idiot .

 

[Yugang]

Ugh, I completely forgot about tilting it. I’m an idiot .

No, quite the opposite. Kicking around ideas always leads to something good, and appreciate it.

I mostly used CO2 Spray Bar in overflow mode, and with dialling the overflow end piece I had an easy way to adjust the power. For Horizontal Reactor, none of the solutions so far I consider elegant, I am not really happy with tilting or the styrofoam idea. As you do, I am also playing with ideas, but haven't come up yet with a nice practical solution to reduce power in overflow mode. Of course the question remains, why would I want to change the power, as CO2 stability is my priority?

 

[SingAlongWithTsing]

I've seen these rubber, donut like, gaskets (no idea what they are called) where tubes can be inserted into pipes and create a tight seal. The pipe could then be adjusted up or down to change the water level inside the reactor. This probably is more for those with sumps or AIO tanks so water could be returned back to the system without tinkering too much.[/ATTACH]

Think youre thinking of uniseals

 

[Unexpected]

why would I want to change the power, as CO2 stability is my priority?

Cost savings (not rebuilding the reactor smaller) and for me, I want the reactor to control pH drop so my tank controller doesn’t have to cycle so much..

 

[Unexpected]

Think youre thinking of uniseals

I believe you’re correct. Thank you.

 

[raj]

I would like to understand the math of how you think there is more surface area given same about of gas in a horizontal reactor vs vertical.

In a vertical suspended bubble reactor, the bubbles entire surface area (depending on bubble size) is in contact with water.

In horizontal reactor, only the portion of gas in contact with water ( the under side of the gas pocket is in contact.

So while I saw some of your calculations you are already assuming you have filled up a pocket of gas where your surface area now execeeds the total surface area of bubbles in suspension. But to reach that point you have to have a large amount of undiaaolved gas sitting at top of the reactor. Until such time the reactors surface area is lower that vertical.

Is my understanding correct of why you think there is more surface area.

Also would be interesting to to a table of how many bubbles it will take for horizontal reactor to break even and exceed.

Also co2 degassing assumption of yours is not accurate. co2aq doesn’t act like suspended gas and rise immediately. And capacity of co2 dissolution in pppm within water is far higher than what we inject. A simple example will be you can pour soda in a flat dish or leave it in bottle and see how long it takes for all co2 to escape. Equilibrium with atmosphere in a still surface take much longer than how you are assuming it. Obviously our tanks don’t tend to have still water but let’s assume it does.

 

[Yugang]

I would like to understand the math of how you think there is more surface area given same about of gas in a horizontal reactor vs vertical.

This is summarised in post 47

how you think there is more surface area given same about of gas in a horizontal reactor vs vertical.

As you read #47 you will see that this is not what is claimed. The calculations are done for the reactor in "steady state", i.e. with the gas pocket already filled and hence a stable surface area. When you build the horizontal reactor you will see that it fills rather quickly with CO2 gas and is probably in this steady state after typically 20-30 minutes with CO2 on in early morning. In this steady state, which is typical for the rest of the day we will see that

The key point is that it is very plausible that the horizontal reactor gives a higher CO2 absorption surface

It is not a robust scientific proof, but indeed very plausible. The convincing observation in real life is then

When we are interested in the power / capacity of the reactor, we now also have some practical evidence what the horizontal reactor can deliver without much need for finetuning and optimizations as we would have to do with bubbles:

• @RickyV ramps up CO2 on a 1000 gallon tank, achieves 1.0 pH drop in 38 minutes
• @toofewfish ramps up CO2 on a 230 gallon tank, achieves 1.6 pH drop in 30 minutes

Try that with a vertical bubble reactor …

As to your last point:

Also co2 degassing assumption of yours is not accurate

Here I am not sure what you refer to? If you could provide a link to my assumption that would be great.

And capacity of co2 dissolution in pppm within water is far higher than what we inject.

This is true, max CO2 solubility in water is about 1000-1500 ppm, depending on salinity. We are working only in the <40 ppm range. As I mentioned earlier this justifies the linear assumptions that I make for absorption and desorption.

 

[raj]

CO2 from the water column (that is no longer in equilibrium with the atmosphere) will outgas at the surface. The rate of outgassing is a function of the tank surface area, the CO2 ppm in the water, and the surface agitation. The rate of outgassing (gram CO2 per hour) is NOT dependent on the tank depth/volume.

This is what i was talking about. This assumption is wrong, just because co2 is not in equilibrium with atmosphere doesn’t mean it immediately outgases.
The surface of water does want to reach equilibrium but needs some catalyst that breaks surface of water like agitation.
The depth of tank does matter too because dissolved gases rise to the surface thru osmosis displacement. This is why shorter tanks degas faster than taller tanks of same length and width.

Recently someone posted an observation on FB group that it took 18hours to degass his tank and that too with a sump. But you can dissolve co2 into water much faster.

The above is just for clarification and has nothing to do with the design of the reactor. It just helps you right size it.

Moving on to design of the reactor.
Now that you have also listed your assumption that the effectiveness is based on fact that it’s filled up with gas. This indirectly means that you are using a lot more co2 to reach stead state. Your highest surface area will be at center of the reactor because that your diameter.

Next once the chamber if filled, if your rate of injection is higher than dissolution, then gas will start escaping the reactor, just in bigger bubbles not fine, which is why some people have not seen it if it’s happening.

This design would also mean there is also co2 getting dissolved even after co2 was turned off, which could be couple hours depending on how much gas was filled up in the chamber and how quickly it gets dissolved. Now to counter that, if you turn off too early you could be in a situation where your surface area is diminishing and you are no longer at your target PH drop.

I do applaud you on the design. It’s based on similar principle as inverted bell but gives you more surface area and moving water plus depending on you pump could me higher pressure.

The reason I came here to discuss was because, even tho the reactor is effective, it’s not as efficient as a properly built vertical reactor. Vertical reactors do need higher flow and need to be right sized and tall, but just by design they are most efficient in dissolving co2 and maintain steady state. I have done it for years using Rex Griggs and many other have too using either Rex Griggs or Cerges style vertical reactors.
In my Rex Griggs, 24” tall, I never have more than 3-4ml of undissolved gas. I do prefer running it on a dedicated loop but I can reach very high ph drops very easily.

The validity of the design remains and that the simple math of it will tip in favor of the horizontal reactor, especially if you are not able to dissolve 100% of the gas using vertical..

 

[Yugang]

This is what i was talking about. This assumption is wrong, just because co2 is not in equilibrium with atmosphere doesn’t mean it immediately outgases.

For a real life tank, with surface agitation and water circulation the assumption is correct. We're not assuming a stagnant water column.

The surface of water does want to reach equilibrium but needs some catalyst that breaks surface of water like agitation.

Correct, and that's why good surface agitation is generally recommended for good gas exchange.

The depth of tank does matter too because dissolved gases rise to the surface thru osmosis displacement. This is why shorter tanks degas faster than taller tanks of same length and width.

This is only when we assume a stagnant water column, which is not realistic as for our hobby we strongly recommend and assume a tank with circulation and surface agitation.

Now that you have also listed your assumption that the effectiveness is based on fact that it’s filled up with gas. This indirectly means that you are using a lot more co2 to reach stead state. Your highest surface area will be at center of the reactor because that your diameter.

The daily CO2 consumption is many many times the volume of the gas pocket in the top of the reactor. This gas pocket will fill in typically 15-30 minutes. So we are not using a lot more CO2 to reach steady state shortly after CO2 switched on. At the end of the day this gas pocket will also be absorbed in typically 15-30 minutes. Overall, the net effect is that CO2 absorption efficiency is not affected at all.

Next once the chamber if filled, if your rate of injection is higher than dissolution, then gas will start escaping the reactor, just in bigger bubbles not fine, which is why some people have not seen it if it’s happening.

I would recommend to build one, and you will observe otherwise.

This design would also mean there is also co2 getting dissolved even after co2 was turned off, which could be couple hours depending on how much gas was filled up in the chamber and how quickly it gets dissolved. Now to counter that, if you turn off too early you could be in a situation where your surface area is diminishing and you are no longer at your target PH drop.

As per above, this is not correct. The gas pocket will be absorbed very fast, and I have yet to observe any disadvantage of having a gas pocket.

it’s not as efficient as a properly built vertical reactor.

A horizontal reactor of any given size you can pack much more punch than in a bubble reactor, and absorption efficiency is close to 100%. How would you define efficiency to make your statement correct?

Vertical reactors do need higher flow and need to be right sized and tall, but just by design they are most efficient in dissolving co2 and maintain steady state.

What would be your definition of efficiency, and could you explain the physics and/or measurements of vertical reactors being more efficient?

I have done it for years using Rex Griggs and many other have too using either Rex Griggs or Cerges style vertical reactors.
In my Rex Griggs, 24” tall, I never have more than 3-4ml of undissolved gas. I do prefer running it on a dedicated loop but I can reach very high ph drops very easily.

Hobbyists have been using these reactors for nearly 30 years, and obviously they work. If you're happy with it, don't change it.

 

[Yugang]

I have lost count how many now are using this reactor, but believe we have only 2, @FrankZ #3 and myself, using it in overflow mode so far.

Overflow mode is a unique feature, it cannot be done with any diffuser or conventional bubble reactor. I am really a fan and will probably never look back and use other methods again for my tank.

We do not longer rely on the stability of my CO2 regulator or CO2/pH controller, but it is the dimension of the reactor that drives the rate of injection and is therefore very stable and predictable. We get simplicity and stability, without having to spend a lot of $$.

I also did theoretical and experimental work to better understand and optimise what happens with overflow mode, and have found some good checks how to set it up and verify if all is good.

While several hobbyists may prefer to be able to adjust CO2 injection with a precision valve or controller, and spend the $$$ on these, It is probably the unconventional and unknown of the horizontal reactor that holds many back from using overflow mode.

If anyone likes to try overflow mode, please reach out and I will be happy to help as much as I can. When we have more users successfully using it, sharing experience in this thread, it will benefit new users considering what CO2 approach to take.

 

[toofewfish]

I know from the experience I've had with the hortizontal reactor I built in July for my 150 gal tank, I would not hesitate the least bit for my next tank to try an overflow type. I have not made any adjustments in pressure or co2 flow rates since I did some initial testing on my tank with the new reactor. When I had a vertical reactor on my tank, I just couldn't find the sweet spot and was constantly chasing the flow rate. Now I don't have any worries after tracking pH for the last 6 months. Just steady steady as can be.
You all need to try it out!

 

[Unexpected]

More eliminated, little more spread and plants are getting healthier. I’m going to wait and see before removing a few reds but I’m just going to watch which ones do the best.

Also. No KH dosing from here on. Hopefully the weird issue I was having is gone.

 

[Unexpected]

 

[Yugang]

View attachment 4219More eliminated, little more spread and plants are getting healthier. I’m going to wait and see before removing a few reds but I’m just going to watch which ones do the best.

Also. No KH dosing from here on. Hopefully the weird issue I was having is gone.

Nearly a year ago that @Unexpected built the first Horizontal Reactor . I believe that an award winning scaper pioneering it has made all the difference. At the time I was still too happy with my CO2 Spray Bar (the 'mother of horizontal reactor'), building many versions and doing experiments on overflow mode, and I wasn't ready to leave that concept yet.

Only my good friend @Hufsa built a CO2 Spray Bar , which was a partial success as she finally preferred an inline reactor. I guess we are waiting for some unexpected (pun intended) next top scaper to pioneer this concept for it to be more widely adopted. I felt I had to abandon my CO2 Spray Bar and build a horizontal reactor, but if I ever own a second small tank, I won't hesitate.

 

[Unexpected]

lol, I think I was suppose to post this on my journal. Think I hadn’t had my coffee yet!
Sorry @Yugang , didn't mean to disrupt this thread, It appears it won't let me delete it at this time.

 

[Yugang]

The reactor below, photo received via a PM, works well with water intake and water exits from below rather than horizontal, but from time to time some gas may build up and the user does a manual purge.



I am copying my reply from a PM here in the thread, as I believe it could be helpful to share with other users as well.

------

I constructed my reactor so that the pocket of gas in the reactor can never grow more than half of the reactor, and starts to purge bubbles if more gas would build up than that.

The reason is that at half filled we have the maximum gas/water surface interface (in a circular pipe), and therefore the maximum CO2 absorption capacity. When we would inject even more CO2 than that, we do not want the gas pocket to grow, but the capacity to decrease (when more than 50% gas filled), leading to a reduced absorption capacity and a faster and faster filling of reactor with gas. This is the feedback loop that I mentioned.

I have centered water intakes and exits (you can see in the thread), but made internal 'elbows' that end at exactly 50% of the reactors' height. So water coming in is led down, to avoid water splashing and noise, and the exit elbow makes sure the gas pocket starts to purge at 50%.

The reactor may be fed some bubbles by the pump and canister, and maybe you have seen that I posted that the reactor also produces 3% ratio of gas itself. The good thing is that usually it purges itself, and most users even don't have or don't use the gas purging valve. I sometimes do, but mostly let the reactor purge itself.

So it seems that the main take away for your tank is that the reactor will work better if it starts purging at 50% (or earlier) gas filled.

 

[Patterson]

Good morning, here is the progress of my reactor. I had a little problem with the diameter between the PVC and the Plexiglas, so I put it everywhere! I'm waiting for the 32/20 reducing caps to finalize. I think I stuck the outlet side slightly too high. We will see with use if I manage to have 10cm of exchange. In the picture behind you can see my automatic water change plant!

IMG-7358 hosted at ImgBB

Image IMG-7358 hosted in ImgBB

ibb.co