Welcome to ScapeCrunch

We are ScapeCrunch, the place where planted aquarium hobbyists come to build relationships and support each other. When you're tired of doom scrolling, you've found your home here.

Yugang Reactors for Smaller Tanks

  • Thread starter Thread starter xNitroPT
  • Start date Start date
  • Tagged users Tagged users None
Joined
Mar 11, 2026
Messages
19
Reaction score
38
Location
MI, United States
Hi all, first post. Thought it'd be fine to start a new thread specifically about yugang reactors for smaller tanks, under surface area 3000cm^2 (for the most part). Guess this could be a journal to some degree, but I think this thread is more for info than sharing my tank. I know people are generally using pvc pipes and fittings to DIY their reactors and to keep the cost down. I wanted something that looked super nice and was already "prebuilt". After scouring the internet for the longest time, I found that these PC water cooling reservoirs should serve that purpose.

1775177466368.webp

Maybe someone's tested this already and found out that it doesn't work, but I wanted to put this out there since I haven't seen anything on it. I don't want to gatekeep this and I hope that this gives people another option since not everyone may have tools readily available. Since this is considered a PC component, there are a number of options available for fittings that I'll get into in a bit that make this great.

There are a few downsides though:

1. Limited diameters. I've only seen options for 50 or 60mm diameters. The 50mm ones have 40mm inner diameter, and 60mm ones I assume have 50mm inner.
2. Limited lengths. I've seen them as small as 6cm long all the way to 40cm long. The caps/lids range from 6-10mm thick.

Largest I've seen is 50mm and 400mm long. Useable diameter and length, let's say are 40mm and 380mm respectively. If you want a pH drop of 1.5, largest your surface area can be is 2690cm^2. Obviously you can stretch this if you don't need that large of a pH drop.
1775177969542.webp 1775178015255.webp
3. Standardized G 1/4" thread. This isn't as common a fitting (in the US at least). This thread size can be good or bad.
4. Some others mentioned later.

As for the upsides:
1. Clean look, no DIY necessary, no PVC glue, certain listings come with brackets and screws.
2. Inexpensive, I got a 210mm variant for ~$15USD
3. Outlet is offset to the side. You can rotate as necessary to control in overflow mode.
4. Others mentioned later.

Again, this is for PC water cooling. 95% of the parts I've seen are made of brass. They are however either chrome or nickel plated.
I have not seen any adverse affects at all, and I do have low KH, low pH water. There are a few amanos living happily in the tank too.

Its a 150g tank with weekly 50% water changes though, so dissolved metals from the connectors must be minimal. You have got be thinking about changing them for nylon ones though.
Not sure if this will be an issue since brass is ~60% copper. Hamfist's tank is also huge compared to my 60p and it may be dependent on water changes. Having the plating on the fittings will lower the corrosion rate since we are in acidic lower pH water. I honestly don't think this should be a problem at all. There are acrylic and stainless steel options out there if you look for them. I've heard nickel plating is best.

As of right now, I have my ADA 60P freshly cycling on a dark start. I was able to test the co2 functionality (not measure co2 levels). I am able to get a co2 bubble to develop inside and work with overflow. My "reactor" is 210mm long and has 50mm diameter. Calculating with the inner diameter (40mm on mine) and the useable length (190mm if you take away cap length), I should theoretically be able to get a ~1.4ph drop.

The filter I'm using is the Biomaster 350 thermo. Tubing is 5/8" for this, and this is one of the downsides I want to get into. This is for PC cooling like I mentioned and they generally don't use tubing that large. If you're running a filter that uses 12/16mm tubing instead of 5/8" (16/22mm) then you might have a much easier time finding fittings to work. G 1/4" thread is roughly a 1/2" opening. I'm not actually getting less flow with this because I realized the 16/22 quick tap valves I use actually have a 1/2" opening on the inside as well. This G 1/4" thread shouldn't be an issue then. FYI, G 1/4" is the same as 1/4" BSPP. (british standard pipe parallel)

1775179432023.webp
I tried making my setup as NEATLY and good looking as possible. Above is my final work (as of now). I have some tube trimming to do though to make it nicer. Another thing I want to mention, the first time I tried this, I connected my outlet straight to the yugang reactor . I saw this thread reply (below) saying that their 2" pvc (~50mm) had no issue with the flow speed after adjusting their tap valve. That was not true in my experience. My reactor should be about the same size as this one, but the water rushes through very turbently:
@Yugang has been so generous and helpful to me, held my hand through the calculations and parts questions, and opened to for to the world is CO2 reactors. I wanted to say thank you and that I appreciate your help.

Here's what we came up with. This is for a 60U (24" x 14") 20 gallon tank. 2" clear schedule 40 conduit 6" in length. At first, I mounted the reactor above the canister filter with both 90° elbows pointed down like so.View attachment 5115
This worked ok, but ( I'll add that I built another reactor for my smaller tank) the other reactor has a slightly different design due to space limitations and it worked better. Both of those 90° elbows faced up. So I moved the reactor down and turned both elbows upward like soView attachment 5116
The flow is more gentle and less bubbles. I control the CO2 flow with a regulator. I did not install a bypass or overflow. I'm using double tap disconnects for maintenance and flow adjustment purposes. I'm running an Oase 350 on this tank and need to throttle it down a bit anyways. This quiets everything down and stops the gurgling and sucking in air at the skimmer. I easily hit 1 ph drop in 40 minutes. I'm using a John Guest fitting for CO2 connection, which is threaded into reactor body using 1/8 x 27 NPT tap and 21/64 drill bit.

Such a huge improvement over the inline diffuser I was using. No more micro bubbles!
And so I had to resort to making a simple bypass. One lesson I learned, don't get a cheap valve on amazon for $5. I placed the valve on the bypass initially:
1775180012945.webp
It slowed the flow so much that almost all water was going through bypass even though valve was fully open. (Don't get cheap Y or T splitters either...)
Here's a video of the flow WITHOUT a bypass. Video starts with tap valve already slightly shut 25% of the way and slowly getting shut more as you can see:
View attachment IMG_7536.mov
It doesn't matter where you put the valve but I think it makes more sense to put it on the reactor side. You can control how slow the flow is that way. Ff you place it on bypass side, you can never slow down water through reactor that way - only more going to it.

You might ask, where's the co2 supposed to go if you don't have the inline accessory that I have? Well there's a threaded acrylic pipe that comes with this reactor if you look at listings. All you need to do is get a 4/6mm push connect to g1/4" male thread fitting for a few bucks and your issue is solved. The threaded acrylic pipe can also be screwed off and there are more extender fittings and such to replace that if you search them online. Options like the 60mm length reservoir I posted above may only have 2 outlets, but the larger 400mm example I gave has 4 holes with different hole positions too. You can place a purge valve that slightly works (there's literally a fitting for that too). The main outlet hole is still offset as well. (another example shown below) Any hole you don't want to use? Just buy a g1/4 plug. The benefit of using this PC cooling reservoir is the amount of fittings they have that can work for your needs.
1775180675889.webp

For my setup, I ended up using a $3 dollar plastic garden hose splitter that has 3/4" GHT (garden hose thread) that was two valves on it (best of both worlds). I also bought the appropriate barbs to fit into it (so hosing can attach). Because I wanted to put this is my AquaForestAquarium Archaea cabinet, everything is packed really tight. Would be nicer if tubing had more room to move and didn't kink as easily. Just a tradeoff so you see I had to use elbow barbs and keep some tubing real short.

My fittings right now are brass but I found one seller on ebay (shipped from China) that has G 1/4" male thread to 5/8" barb in STAINLESS STEEL. However, it is BSPT (british standard pipe thread) so this one tapers inwards and will probably need teflon/very tight twist to seal with the o-ring. BSPT fitting can go into a BSPP threaded hole, but BSPP cannot go into BSPT. Someone correct me if I'm wrong about that, but I'm fairly certain that is the case.

If you've read to this point, I think this should be enough to understand what you can do with this PC cooling reservoir. It has limitations size wise and a need for a bypass if you don't want to slow the flow on your tank too much (depending on your filter), but it really is a nice cheap option that we can use for our hobby.

Please correct or question me anywhere!!!

--------------------------------------------

Just links to what I bought for anyone curious. I hope links are allowed. I don't have any commission to these. Just purely for info on what I have.

You can buy these for cheaper on AliExpress. I just didn't want to wait and paid a little more for the items. They're from the same supplier, but I also recommend finding a higher quality brand. The one I have is 50mm outer diameter. I think companies like Freezemod have higher quality ones, and they might be 60mm. Do your own research on this, I didn't do enough on this side.
1775181367745.webp
Some other hard to find accessories:
- If you're trying to replace inlet cap, make sure the cap you find works. Here's one I found for 50mm diameter:
- I believe the acrylic reservoir itself is this (please check yourself!!):
- This one is super nice, it's a rotatable cap replacement you can get. Set your in/outlets exactly where you want. I'm not sure if 50 or 60mm diameter though:

I also mentioned not wanting to have to drill anything. You might wonder how to attach those brackets without doing so. I bought adhesive stickers with screw ports from AliExpress, and used screws/washers that came with the reservoir:
1775183051318.webp
Can probably get on Amazon as well. The adhesive first, then stack 3 washers next, bracket next, thread the screw through the bracket hole -> washers -> adhesive screw port.
1775185140171.webp


-------------------------------
Random thought:

I found this one singular option that I've not tested. It looks so similar to the internal bypass option that yugang himself devised. (a prebuilt option at least) It's slotted but to be honest, I don't know if it would work well. If you search for aluminum pc water cooling reservoir, you will find this. Only comes in one size unfortunately. Again, I have NOT tested this.
1775181001142.webp1775181034575.webp
 

Attachments

  • 1775181294690.webp
    1775181294690.webp
    16.3 KB · Views: 8
  • 1775180666087.webp
    1775180666087.webp
    9.7 KB · Views: 5
I opted for glass bells with suction cups to wall of tank for my 20 high. More of a spray bar concept as I have a small 3 watt powerhead pushing about 50 gp hr under the bells. I have three bells on a 20 high giving me about 12 ish ppm per ph drop measurements and somewhat corroborated by drop checker.

IMG_4500.webp

Co2 generated by diy sugar, gelatin, yeast fermenters.

IMG_4480.webp

I run three bottles of sugar gelatine yeast. The fourth bottle is just water acting as a wash chamber for generated gas and providing a bubble counter. I add a new bottle and swap out an old one every 2-3 weeks so the bottles are all in staggered production giving me more even generation of co2. You can tell its time to swap a bottle when the water levels start rising in the bells indicating gas generation is not keeping pace with absorption.
 
Thought it'd be fine to start a new thread specifically about yugang reactors for smaller tanks, under surface area 3000cm^2 (for the most part).. I wanted something that looked super nice and was already "prebuilt". After scouring the internet for the longest time, I found that these PC water cooling reservoirs should serve that purpose.

How has this build been working out for you?
 
How has this build been working out for you?
Extremely well. I turn on the co2 about 2.5hrs before lights come on and it will be yellow on the drop checker. I don’t have a Hanna test kit, but my rodi with gh remineralizer should be <1kh. Measured ph is pure yellow on api liquid test. I’ll post a picture maybe tomorrow or the day after since I took out the drop checker already, given that I saw it was fine.

Plants are doing fine but it seems my rotala blood red has patches of transparency on lower leaves. New leaves are healthy. Not usually a sign of co2 so I’m thinking it’s a phosphorous/nitrogen/calcium issue (things I’m probably low on even if though I’m using fresh brand new APT feast) Rotala Ceylon doing perfectly fine though.. 1780415341932.webp
 
That’s funny. I’ve been thinking about making a co2 reactor from one of those cpu cooling reservoirs since I first saw them years ago. Thanks for actually putting in the legwork!

Given the recent innovation with the yugang reactor , and more specifically the methodology of co2 dissolution, I’ve also been looking at these rectangular acrylic reservoirs as well.

From what I understand, if we are strictly relying on the principle of surface area, this specific reservoir provides (+ or -) 28.2 square inches, or 3.79 square inches over the required 24.41 square inches needed for a 1/17.7 ratio to achieve a 1.5ph drop on a 36”x12” (432 sq in) 22 gallon bookshelf tank (my display).

If I’m understanding this correctly, wouldn’t this be more efficient than a cylinder since the gas pocket forms at the desired surface area immediately, as opposed to the gradient when a cylinder fills up to the desired area at the midpoint (diameter)?

This is the reservoir in question:

IMG_3779.webp
Acrylic PC Water Cooling Tank Reservoir with G1/4 Thread

Interesting in your thoughts on this @xNitroPT.

Also interested in feedback regarding the maths @Yugang.
 
That’s funny. I’ve been thinking about making a co2 reactor from one of those cpu cooling reservoirs since I first saw them years ago. Thanks for actually putting in the legwork!

Given the recent innovation with the yugang reactor, and more specifically the methodology of co2 dissolution, I’ve also been looking at these rectangular acrylic reservoirs as well.

From what I understand, if we are strictly relying on the principle of surface area, this specific reservoir provides (+ or -) 28.2 square inches, or 3.79 square inches over the required 24.41 square inches needed for a 1/17.7 ratio to achieve a 1.5ph drop on a 36”x12” (432 sq in) 22 gallon bookshelf tank (my display).

If I’m understanding this correctly, wouldn’t this be more efficient than a cylinder since the gas pocket forms at the desired surface area immediately, as opposed to the gradient when a cylinder fills up to the desired area at the midpoint (diameter)?

This is the reservoir in question:

View attachment 17586
Acrylic PC Water Cooling Tank Reservoir with G1/4 Thread

Interesting in your thoughts on this @xNitroPT.

Also interested in feedback regarding the maths @Yugang.
I have used a similar 200 ml rectangular cooling reservoir for my 8-gallon nano tank. Works like a charm with the added advantage that it always works in overflow mode without ever having to purge CO₂, if it's sized correctly. This is because the maximum surface area is limited to the product of the largest dimensions and the CO2 pocket would not be able to grow beyond this size. The only issue is customization; you aren't able to customize it the reactor to a specific pH drop. You just gotta go with what you get. IMO as long as you're getting a 1.2 pH drop, it's good enough
 
That’s awesome info. Have any pictures of the setup in place? I have a Waterbox 6 gallon and I’ve considered doing something similar. The fixed surface area is actually more desirable to me, as most of my tanks are plant-only and stability of co2 dissolution is the primary objective. Having a reliable, fixed pH drop removes a variable when trying to stabilize a tank.
 
That’s awesome info. Have any pictures of the setup in place? I have a Waterbox 6 gallon and I’ve considered doing something similar. The fixed surface area is actually more desirable to me, as most of my tanks are plant-only and stability of co2 dissolution is the primary objective. Having a reliable, fixed pH drop removes a variable when trying to stabilize a tank.
I don't have any photos on hand but it's basically plug and play. You just need to buy 1/4"G threaded fittings for your CO2 inlet and water inlet and outlet and you're good to go. They're widely available on amazon. Make sure it's G thread not the American NPT.
 
wouldn’t this be more efficient than a cylinder since the gas pocket forms at the desired surface area immediately, as opposed to the gradient when a cylinder fills up to the desired area at the midpoint (diameter)?
Correct! Not an insane improvement (the cylindrical ones work just fine) but the immediate CO2 dissolving effect would definitely be greater.

The issues are 1) cleaning (doesn't need much, but gunk will buildup over time). How do you remove and clean?

And the far bigger issue 2): purging. Running a CO2 reactor in regulator mode (doesn't purge on its own) means you get 100% CO2 dissolving and zero waste, but a pocket of O2+N2+Misc gasses will build up over time. On the tanks that I run my reactor in regulator mode, I usually have to manually purge the gas pocket 1-2 times per week or I get splashing and other issues.

Other issue with this idea 3): you can't adjust the power if necessary. If you under size your reactor, you're out of luck! If you oversize your reactor, you're out of luck. Cylindrical reactors with an offset exit are adjustable, so you can oversize it and still never worry about gassing your fish (if run in overflow mode).
https://www.walmart.com/ip/Computer...AP4HJ9QmqUiuShCaY0_aem_FsDi5ctX5BPuMuYAYCNm7A
Interesting in your thoughts on this @xNitroPT.

Also interested in feedback regarding the maths @Yugang.
 
Correct! Not an insane improvement (the cylindrical ones work just fine) but the immediate CO2 dissolving effect would definitely be greater.


The issues are 1) cleaning (doesn't need much, but gunk will buildup over time). How do you remove and clean?

And the far bigger issue 2): purging. Running a CO2 reactor in regulator mode (doesn't purge on its own) means you get 100% CO2 dissolving and zero waste, but a pocket of O2+N2+Misc gasses will build up over time. On the tanks that I run my reactor in regulator mode, I usually have to manually purge the gas pocket 1-2 times per week or I get splashing and other issues.

Other issue with this idea 3): you can't adjust the power if necessary. If you under size your reactor, you're out of luck! If you oversize your reactor, you're out of luck. Cylindrical reactors with an offset exit are adjustable, so you can oversize it and still never worry about gassing your fish (if run in overflow mode).
Computer Water Cooling Reservoir Rectangle Transparent Acrylic PC Water Cooling Tank Reservoir with G1/4 Thread - Walmart.com
Actually, the offset reactor idea reminds me: you can actually buy the 190 mm cylindrical PC reservoir, and this actually has two holes on one end, which means that the holes are offset from the centre on one end. You can rotate that reservoir on its stand to regulate CO₂ for that model of reservoir.
 
That’s funny. I’ve been thinking about making a co2 reactor from one of those cpu cooling reservoirs since I first saw them years ago. Thanks for actually putting in the legwork!

Given the recent innovation with the yugang reactor, and more specifically the methodology of co2 dissolution, I’ve also been looking at these rectangular acrylic reservoirs as well.

From what I understand, if we are strictly relying on the principle of surface area, this specific reservoir provides (+ or -) 28.2 square inches, or 3.79 square inches over the required 24.41 square inches needed for a 1/17.7 ratio to achieve a 1.5ph drop on a 36”x12” (432 sq in) 22 gallon bookshelf tank (my display).

If I’m understanding this correctly, wouldn’t this be more efficient than a cylinder since the gas pocket forms at the desired surface area immediately, as opposed to the gradient when a cylinder fills up to the desired area at the midpoint (diameter)?

This is the reservoir in question:

View attachment 17586
Acrylic PC Water Cooling Tank Reservoir with G1/4 Thread

Interesting in your thoughts on this @xNitroPT.

Also interested in feedback regarding the maths @Yugang.
I think Rocco basically answered it. Will work like a beast and it’s all based on surface area and having some flow for gas exchange.

The cleaning would be the biggest issue with that type of reservoir if you can’t open it up. Sure you could soak the whole thing in a bleach solution if you want though. Just easier with the cylindrical since I can actually take it apart. Gives more control since you can offset it as scaper mentioned.

For your 22 gallon, you’d have to get one of the longer circular reservoirs, so for space concerns the square shape may be better.

One thing I do want to note is, it doesn’t look like there is ANY buildup in my reactor since I’ve been running the tank for 1-2months so far. I’m not sure why that is the case because there was algae/brown scum buildup in my lily pipes and hosing. (This is to my naked eye though, touching it might say something different) saves me the hassle of cleaning it anytime soon.
 
Did you size for a pH drop of 1.5? or were you going for something lower?
I aimed for dividing my surface area by 17.7 to get the 1.5ph drop just because that's all I really read about in this forum for the most part. No specific reason besides having >30ppm CO2 in the water. My reactor options were the 240mm or 210mm long, and it basically required a length in between. So I went for 210mm version and got a ~1.4ph drop (reactor inner dimensions slightly smaller). It should be perfectly fine with my water and personally think it's more than enough to grow most, if not all plants at that co2 saturation.
 

Top 10 Trending Threads

Back
Top