Reactor assembled. Pic Now running.

[stephane]

Tank DJ for the diagram it will help me to build mine

did you have a small piece of tubing that you have take to built your buble counter

I want to built one but I dont want to buy a hole lenght for a couple of inch

Email me if you have one want it zapdesign@sympatico.ca or anyone else!!!

[Delphinus]

I'm resurrecting this ancient thread, because a couple of us have built reactors recently using this design.

Hindsight being 20/20 I have a few suggestions for anyone comtemplating this design for their own DIY reactors.

1) Go with 3/4" PVC for the piping. I went with 1/2" because the mag drive inlet and outlet are both 1/2". But 1/2" does reduce pump output due to friction. With 3/4" and just bushing'ed down to 1/2" at the pump, you would be able get better flow from your pump.

One drawback with this design is that the piping is in place permanently. Once it's in and glued you're never going to be able to say, swap up from 1/2" to 3/4" without first hacksawing the reactor to pieces and rebuilding from scratch. So go with the 3/4" from the start.


2) The pump sits on a localized high point of the piping. Ie., it sits higher than it's intake, and the piping on the outlet goes DOWN before goes back up. This means that any air in the line will stay trapped in the pump, and it makes for priming the pump to be a chore. The pump basically cannot prime itself whatsoever. To prime it, you have to turn on your pump, undo one of your unions, and let the air out until the pump is pushing water. This lets out a lot of water onto your floor.

One potential way around this, is take the output of the pump and aim it UPWARDS rather than sideways. Have your reactor effluent on the high point of this pipe. This way the pump can push out the air by itself. Yes, one drawback is that there is now yet another 90-degree bend in the piping and this means reduced pump efficiency. BUT, if you go with 3/4" PVC instead of 1/2" maybe that will work to somewhat reduce the losses due to friction.

3) I'm not sure what the solution is just yet, but, I'm having a heck of a time clearing the lower chamber of bubbles. I was going to not worry about it, but, as I leave the reactor running for long periods of time, the bubbles in the lower chamber grow. And eventually it becomes a level drop in the lower chamber because the air bubbles just grow, and grow, and grow. I think that what is happening is that the holes joining the two chambers don't allow enough flow, thus, the pressure in the lower chamber just drops and drops. This suggests to me, that left unchecked and without some kind of corrective measure, eventually the pump will end up cavitating because it will not be able to draw from the lower chamber any longer (because it's pulling against negative pressure).

Hopefully if there is no air in the lower chamber, the risk of this might be reduced. But since I am having trouble even clearing the chamber of "that last little air bubble that refuses to leave" .... I think maybe I don't have enough holes or my holes are too small. So I'm going to empty my reactor and see what kind of action I can take (it's going to be difficult to get into that chamber and enlargen them holes ... maybe with a dremel snake thing .. there's no way I'm going to be able to get a drill down that tube .... :?

[DJ88]

To prime it, you have to turn on your pump, undo one of your unions, and let the air out until the pump is pushing water

You dont need to turn the pump on to prime it. Open a union or else place a guest fitting on the PVC before the pump. YOu may get a small leak of water(I never had more than what a face cloth could catch). There were mods I have done that corrects all of these problems. I moved both the CO2 in and Recirc lines onto the PCV prior to the pump intake. To prime you take the tubing out of a fitting an let the air bleed out due to the pressure created by the water in the chamber above. NO muss no fuss.

I'm not sure what the solution is just yet, but, I'm having a heck of a time clearing the lower chamber of bubbles.

I don't know why this is happeneing.. Both mine and Bruce's reactors built in this fashion do not have air accumulate in the lower chamber. Where are you injecting the CO2? Try and have in injected so it goes into the pump to be chopped then into the TOP of the reactor. if any collection occurs it will then be at the top of the reacotr and removed by the recirc device.


A trick of mine that I will use on any tank I may set up now is to have the effluent outlet directly into the overflow box up in the tank. This allows any excess CO2 to have a long drip down through a chaotic bubble area then into the sump and through the sump. Much better. As well it eliminates the possibility of a possible draining of the reactor due to it's own water volume pressing down on the outlet. With a check valve after the pump injecting water in the unit will then not lose water volume due to internal pressures.

[Delphinus]

Welcome back, Darren...

A few mods have been made to the design since that post Canadian_Man & I (and I should add mostly Canadian_Man's brain power behind this) have been doing a lot of trial & error ...

The main problem with the air bubbles in the lower chamber was the feed arrangement. Basically the reactor was fed off the overflow and since the output was lower than the top of the reactor a slow siphon was taking place.

Out of necessity, I ended up hacksawing off the end as there were leaks. First off, both CO2 injection and reactor intake are on the pump inlet. The reactor effluent is at the top. The top of the chamber is pressured enough to bring back the water several feet higher than the reactor. The pump inlet has enough suction pressure to draw water in by siphon. In my case I have the feed and effluent in my main tank, the effluent goes right into my overflow standpipe via a small piece of airline tubing. In Jon's case the reactor sits lower than his sump so it just goes into the sump. Same effect though.

We found (see the thread "Reactor feed questions") that the reactors were not capable of handling the line pressure in the sump return line. Problems with leaks, bulging edges, impossible to seal flange, etc.

I still cannot completely eliminate the air bubbles in the bottom but I have learned to live with them. At least they no longer grow into a large level drop.

One added benefit of changing the feed arrangement was that since there are now two inlets on the plumbing, one is a "T" facing upwards and another is a "T" instead of a 90-degree elbow at the end, there are now two places in the plumbing that qualify as "higher than the pump" and the pump is now self-priming without undoing any of the unions.

There are still some things that had I to do it all over again, I might still do differently. But that's hindsight. In all honesty finding the right groove has been a three month nightmare (started this in beginning of September). I only just turned on the reactor for the first time for over 24 hours without some kind of "oh, crap" situation, two days ago!! I'll have to tell you about all my mistakes and "do overs" sometime... but now it is fast fading into obscure memory.

[Delphinus]

I still cannot completely eliminate the air bubbles in the bottom but I have learned to live with them. At least they no longer grow into a large level drop.

Hmmmm, OK, I lied. I see this morning that indeed the air bubble in the bottom chamber continues to grow. It's very frustrating. The only thing I can think of is that all the air trapped in the media eventually gets sucked down into the lower chamber and then has nowhere to go. Turning the reactor off, the air bubble doesn't seem to want to trickle back up to the top to be able to exit. It just stays in the lower chamber. Short of drilling a hole in the lower chamber, there's no way to get this air out. I even enlarged my holes separating the chambers when I had this apart to try to make the flow through a little easier.

So what gives? Is this an artifact of the design or am I still doing something wrong? What is the risk of leaving the air in the reactor?

BTW, the CO2 injection is already on the inlet side of the pump (as you have suggested). Thus the CO2 does have nowhere to accumulate but the top of the chamber which then gets sucked into the recirc line. It seems I can't completely eliminate a small air bubble at the top of the reactor either but I imagine that in time that bubble may dissipate. It's the air bubble at the bottom which has me frustrated.

[Canadian Man]

Tony,
I too had an air bubble stuck in the bottom when I reset up the ractor but the air eventually dissapated and was gone in a week or so.
I didnt worry abou it too much but if it dosent shrink over time then I would be somewhat concerned.

[Delphinus]

How big was the bubble? When I first turned it on, there was no bubble. Three days later it now pretty much covers at least half of the footprint. I would say .... 4" (the whole width of the box) by 6" (about half the box in the long direction), and in a triangular shape. If the rate of growth continues it will be turn into a level drop in a couple more days. I can't tell where the air is coming from and it doesn't appear to be a leak as the ground is dry all around the reactor.

[ldzielak]

Tony,

I too still have the bubbles under the media, but if I unplug the pump, they float up through the media. I'm starting to think we are maybe over doing it on the CO2 and maybe like you said the small bubble just join to make bigger ones under the media.

I'm really thinking of changing to a up flow system. Maybe with the pump inside the reactor.

Lee

[Delphinus]

I've seen some "up-flow" reactors and boy is it nice to see the entire media bed charged up like that, almost like a fluidized sand filter. I totally agree that an upwards flow idea has merit.

I've thought about changing the flow direction on this reactor design to see what would happen. I suppose, for one thing, one might not need the recirc line anymore (except that bubbles may still accumulate above the pump intake so I guess the recirc line would still have a place). In the end I don't think I could make the "gravity feed" arrangement work in that configuration, so ultimately decided against it for this time around.

Hmmm, maybe the answer is a second recirc line is called for? One at the top of the top chamber and one somewhere in the bottom chamber...

[Canadian Man]

My bubble was quite small.
Like 4 toonies stacked together.

Maybe I should come over and check it out tonight and see what's up.
I think closer inspection is needed.
Maybe media has something to do about it?