DIY-ing a external reactor

**|squee|** · 22nd Mar 2005, 10:28

First and foremost, the design of this reactor is based on this by Ghazanfar Khan Ghori. You'll find that the article is almost exactly the same; I'm just adding in more pictures and step-by-step instructions.

Injecting C02 into a planted tank is one of the most troublesome and expensive things to do and solve. There are so many ways of getting 30ppm of C02 into your tank: Diffuser, reactor, flipper, bubbling into an impeller chamber. I’ve always had problems with my C02 levels. I mostly stuck to diffusion through my trusty Kare’s ceramic plate diffuser. But the bubbles irritated me; they were all over the tank as my water current had to push the bubbles around for maximum diffusion.

I decided I had enough (the bubbles were disturbing my view of the tank) and I decided to try to do an inline external reactor using PVC piping. There are numerous articles on the net, but all the measurements and names of US stores got me fuddled (not to mention the lack of step-by-step instructions in “For Dummies” style). I decided to document my experience. Hopefully it serves as a better guide.

Firstly, I went over to this shop:

Hai Chew Hardware & Co.
Block 153
Serangoon North Ave 1
#01-502
Tel: 62844445

That’s where I got the PVC parts, PVC glue and hose clamps for my reactor, along with a tube of Selley’s silicone glue at HomeFix. I also got the 5 bioballs from my old UP internal reactor. The parts are shown below:

Parts list:

1) 10 inches long of 2 inch diameter PVC pipe
2) 2 compatible end caps for the pipe
3) 2 brass hose connectors
4) (Not included in the picture and thus optional) 2 fitting nuts for the hose connectors
5) PVC glue
6) Epoxy
7) Bioballs
8 ) Hose clamps
9) Silicon glue
10) Airline connector ( You can buy this at your LFS. I reused the green part of my UP C02 diffuser (simple type) that I bought from NA )

Total price: Under SGD$25

Try to bring along a length of hose from your filter so that you can be sure the connector fits. The owner of that shop speaks English too, so it should be no problem describing the various parts you want.

The first step to do is to drill a hole into the end caps to fit the hose connector in. The brass piece I got was 10.5mm approx, so I drilled a 10mm hole in.

Next, I took a pair of pliers and screwed in the hose connector into the hole, stopping when there’s a few mm left to screw.

I applied silicon around that remaining edge, (really hate water leaks, wanted to be safe here) and screwed in the rest afterwards.

**|squee|** · 22nd Mar 2005, 10:29

I turned the thing around, and did the same thing for the nut.

That was it for the end caps. I left them to dry for 24hours.

Meanwhile, I drilled a hole halfway up the PVC pipe to insert in the airline connector. I used a 5mm drill bit. I coated the area with a layer of Epoxy afterwards (like I said, I hate leaks =p).

**|squee|** · 22nd Mar 2005, 10:29

Basically it was done. After the epoxy had cured (another 24h here) I dropped in the bioballs, glued in the end caps, gave the whole thing a good wash to clear any chemical residue, and put it inline at the inlet of my Atman 3335.

**|squee|** · 22nd Mar 2005, 10:30

Results:

Tank interior looks much neater, and less wastage of C02.

A possible improvement will be to use clear PVC instead of opaque, as you can see whether or not the bioballs are moving, and whether there is clogging. Also, innovative people out there could try gluing the ceramic plate from the UP diffuser I mentioned earlier to help aid even better diffusion. The cost of the whole project can be reduced if plastic barbs instead of the expensive brass ones are used.

For reference, I’m using this reactor with an Atman 3335 with a flowrate of 600L/hour, connected to the inlet of the filter, bubbling C02 at a rate of 1 bubble per 3 seconds.

**Simon** · 22nd Mar 2005, 10:48

nice job terence

**PeterGwee** · 22nd Mar 2005, 12:32

Well done...

I'll suggest using an individual powerhead to run the reactor rather than the filter (more flow..better mixing..better response time.).

Regards
Peter Gwee

**Cacatuoides** · 22nd Mar 2005, 14:55

did you get e result you wanted as in:
1) less bubbles in your tank and
2) maximum diffusion of CO2 into your tank?

great detailed n step by step pictures!!

cheers!
Eman

**|squee|** · 22nd Mar 2005, 15:22

Cacatuoides,

Yeap, less bubbles. Now the only bubbles I see are from the bubbling

As for the maximum diffusion part... no idea. I'm happy with it in the sense that it works... since then plants have been great and algae hardly bothers me.

PeterGwee,

You asked, I noticed bubbling 2 hours after lights come on. Hopefully this is due to the reactor actually working, and not excess C02 stored in the reactor contributing.

**neon** · 24th Mar 2005, 13:09

Originally Posted by |squee|

[
Results:

Tank interior looks much neater, and less wastage of C02.

A possible improvement will be to use clear PVC instead of opaque, as you can see whether or not the bioballs are moving, and whether there is clogging. Also, innovative people out there could try gluing the ceramic plate from the UP diffuser I mentioned earlier to help aid even better diffusion. The cost of the whole project can be reduced if plastic barbs instead of the expensive brass ones are used.

For reference, I’m using this reactor with an Atman 3335 with a flowrate of 600L/hour, connected to the inlet of the filter, bubbling C02 at a rate of 1 bubble per 3 seconds.

Hi |squee|,

If both end of the pvc is sealed with the PVC glue , and when there is clogging , what can you do ?

Has anyone come up with a screw cap and transparent version ?

Cheers

**schema** · 24th Mar 2005, 18:49

Originally Posted by neon

Hi |squee|,

If both end of the pvc is sealed with the PVC glue , and when there is clogging , what can you do ?

Has anyone come up with a screw cap and transparent version ?

Cheers

Neon,

Clogging can be removing by flushing method,
transparency is a + bcos u can see the clogging/cleanliness
Screw cap type will increase inventory cost & potential leaks.

**schema** · 24th Mar 2005, 19:16

This might interest u further....................Mine was derived from this design

Pics below was lifted off this webby www.hoftiezer.net

**|squee|** · 24th Mar 2005, 20:25

And you can see if those bioballs are actually spinning or not; up till now I've no freaking idea if the reactor is even working

**PeterGwee** · 25th Mar 2005, 08:37

Originally Posted by |squee|

And you can see if those bioballs are actually spinning or not; up till now I've no freaking idea if the reactor is even working

A simple check on the movement of the pH should tell you if its working or not.

Regards
Peter Gwee

**schema** · 26th Mar 2005, 17:13

Originally Posted by |squee|

And you can see if those bioballs are actually spinning or not; up till now I've no freaking idea if the reactor is even working

The bioballs will NEVER spin bcos there is no vortex created in the process, in my case it only gets push down to the bottom by the water flow, the other function of the bioballs is to "block" undissolve CO2 bubbles from exiting the reactor quickly.

Bro Squee, I notice your inlet (to the reactor) is at the bottom, IMO is technically wrong cos CO2 bubble will rise upward (naturally) thus the water flow will simply "push" them out.
Invert you reactor and I believe the CO2 dissolving rate will be higher.

**PeterGwee** · 26th Mar 2005, 18:00

Originally Posted by schema

The bioballs will NEVER spin bcos there is no vortex created in the process, in my case it only gets push down to the bottom by the water flow, the other function of the bioballs is to "block" undissolve CO2 bubbles from exiting the reactor quickly.

Its more for promotion of gas exchange and turbulence. They don't need to spin for things to work.

Originally Posted by schema

Bro Squee, I notice your inlet (to the reactor) is at the bottom, IMO is technically wrong cos CO2 bubble will rise upward (naturally) thus the water flow will simply "push" them out.
Invert you reactor and I believe the CO2 dissolving rate will be higher

Not really unless the outlet of the CO2 is very near the water outlet. The aquamedic reactor 1000 has its CO2 outlet at the bottom corner of the reactor with its connector at the top though (A rigid tube from the top of the reactor stretch to the bottom corner where CO2 is release. More contact time for the CO2 bubble..helps a bit in dissolution of CO2.). Folks might consider the Tom Barr method using the CO2 fed at the powerhead suction side(that is if you use a powerhead to drive the reactor). The gas mist from the impeller "chopping" helps in the dissolution of CO2 as thousands of finer bubbles have more surface area of contact than one big bubble. No need for drilling of CO2 inlets at the reactor side (no more potential leak issues nor check-valves needed as the CO2 inlet has no backpressure.).

Regards
Peter Gwee

**|squee|** · 26th Mar 2005, 20:42

Originally Posted by schema

The bioballs will NEVER spin bcos there is no vortex created in the process, in my case it only gets push down to the bottom by the water flow, the other function of the bioballs is to "block" undissolve CO2 bubbles from exiting the reactor quickly.

Is that so? Ok, something learnt today.

Originally Posted by schema

Bro Squee, I notice your inlet (to the reactor) is at the bottom, IMO is technically wrong cos CO2 bubble will rise upward (naturally) thus the water flow will simply "push" them out.
Invert you reactor and I believe the CO2 dissolving rate will be higher.

Nono, I think you misread it haha. The water flows from top to bottom of the reactor, "sucked" in by the filter. So there's a downwards pushing force against the rising C02 bubbles.

**schema** · 26th Mar 2005, 22:34

Originally Posted by |squee|

Nono, I think you misread it haha. The water flows from top to bottom of the reactor, "sucked" in by the filter. So there's a downwards pushing force against the rising C02 bubbles.

OIC, my conclusion is after scrutinising pic(tell a thousand word) la, then flow is correct!

IMHO but won't that create new problems like..............
1. Unfiltered water enters reactor = dirty internals....mine so far is clean, so no maintenence called for.
2. CO2 bubble enters canister, tech. not so good bcos bubble cos blade wear and introduce air trap.

**|squee|** · 26th Mar 2005, 23:23

Yeap, I'm expecting that. So with this design I have to make sure I regularly clean the reactor.

For the C02 bubble part, I think it's highly unlikely in my case, since I listened very carefully at intervals for any sound of chugging from my cannistor. Also, my Atman 3335 ( 2ft tank ) is quite weak, I don't think the cannistor has the power to actually "suck" the bubbles into the compartment!

This design is darn flawed