Problem with external CO2 reactor

[Eudmin]

I built an external CO2 reactor like Tom's one listed here but didn't include the de-gassing attachments. Basically a 1ft long 2" clear PVC chamber with the output from my Eheim 2215 going into the top, coming out the bottom and up to the tank and some 3/16 " rigid tubing glued into a hole drilled into the top which I have hooked up to my CO2 tubing->check valve->more tubing->needle valve->brass check valve->solenoid->Micromatic regulator.

I am having a problem with the gas starting once the solenoid opens. What happens is that for the entire time the solenoid is off I can see water creeping up my injection tube and then up the CO2 tubing all the way to the check valve. It takes a while for it to get to the check valve, but then it stops. Then once my solenoid kicks on it seems like the water has made some kind of an activation barrier for the check valve to open (just a guess) or something because the gas doesn't push the water back down the injection tube. Each morning to get the gas flowing I have to open the needle valve wide to get a good enough flow to push back the water which has creeped up the tube and then close it back down to try to get the bubbles back at the right level.

What should I do to solve it? Is the water creeping up there because I've got leaks in my CO2 tubing or something? I know that there is some pressure pushing up that rigid injection tubing because when I take the CO2 line off it likes to spray water out. Maybe I need a better inline check valve? Or maybe I should just switch to one of the inline diffusers like GLA sells since that one has the gas going across a membrane? It seems like these external reactors work fine for most people, though, so can someone suggest what to do to get the gas to flow reliably after the solenoid opens?

Thanks.

 

[nipat]

I have the water creep issue too. But it is pushed back fine when CO2 is turn-on.
BTW, I use two (cheapo plastic) check valves. One at the reactor and
another before solenoid.

http://www.barrreport.com/showthread.php/8586-Water-creeping-into-airline

 

[Eudmin]

OK. I'm using an 1/8 NPT brass check valve at the solenoid which opens at 1PSI from McMaster and a cheapo plastic one close to the reactor chamber. Looks like from your thread that you linked we're using the same kind of cheapo silicone tubing. I also couldn't justify expensive tubing for such a short run, but maybe I'll do that and see if I can get some better results.

It happened again this morning, by the way. Didn't push back when the CO2 was turned on until I opened my needle valve wider to get it moving. It's kind of a shame because for a while I was just running it without the solenoid on the needle valve. Would just bubble in the CO2 24 hrs and would "rinse" it out at night with an air pump and air stone in the tank. That was working great, and I had a solid couple of weeks of very consistent CO2 levels and low-algae growth, but then my 5lb tank ran out in only 2 weeks. So now I'm trying to be more conservative with the CO2 usage and having problems.

I've thought about eliminating the needle valve and solenoid altogether and just buying a CarbonDoser from aquariumplants.com which I understand is basically a solenoid with a timing circuit that just opens and closes at the rate you set to inject a bubble every second or whatever. Before I spent that money, though, I'd give serious consideration to switching out my plants, putting a filter in front of my light, and going low-tech so I don't have to fiddle with CO2.

 

[nipat]

How long did you wait before concluding that the water couldn't be
pushed back without adjusting the needle valve? I never observe mine
because I'm still sleeping at that time. But I suspect it will take some time.

BTW, my working pressure is around 25 PSI.

 

[Eudmin]

Yikes. My working pressure is only a few PSI. Sounds like I really do need a better needle valve because if I go up to 25PSI I can't regulate the bubble rate at all. That must be why I can't push it back down the tubing and you can.

The solenoid comes on at 5am, and I typically check it at 6am and see that nothing is happening.

 

[nipat]

My valve is not that good too (stock one which comes with Azoo reg).
It is very coarse. Much patience is needed to adjust it.
So I understand perfectly the want of better needle valve.

 

[hbosman]

Try 10 PSI. That should be enough for a reactor and shorten the tubing between the check valve and reactor if, you haven't already.

 

[Eudmin]

Will do both of those things. I have it set to 20 PSI right now and after much teensy turning got it to bubble out about 1 per second. I'll shorten the tubing between the check valve and reactor next. It's about 6 inches right now.

 

[Eudmin]

That worked! Got my regulator set to 20PSI and it pushes the creep back down in the morning. Two mornings in a row now.

I had been trying to set the regulator at a level that would let me adjust the needle valve bubble rate in the middle of the range for the valve thinking this would give me much finer control over the rate. Turns out I don't really need that fine of control. The fish don't seem to care if I just bubble it in until my drop checker just starts turning into Mountain Dew color and then tweak it back a bit so that it's light emerald green color in a few hours. This is with 5dKH solution in the checker, so that's gotta be up there in CO2 ppm, but they don't seem to notice.

 

[hbosman]

You might double check for leaks with soapy water. 20 PSI shouldn't have had problems pushing water out of 6 inches of tubing. i was imagining a big loop of hose.

 

[Eudmin]

hbosman, you're right. It doesn't have any trouble. It works just fine once I upped the pressure to 20 PSI. When I had the problem I only had 2-3 PSI because I was trying to get the bubble rate right with the needle valve open to half its full value thinking that this would give me finer control. Turns out this isn't the right way to go. Now with the regulator at 20 PSI my needle valve is just barely cracked open but with some fiddling I can set the bubble rate I want, and it also pushes the water out just fine. It just hadn't occurred to me that this would fix the problem, but I guess it should have.

Here's a picture of my cramped little setup beneath my tank, reactor in the back left and you can see the plastic check valve on top, tank in the front, Eheim on the right, inline heater not visible behind Eheim, regulator, solenoid, needle valve are all level with the floor going back into the cabinet since they end up being kind of long with all of the fittings:

[attachment=1025:name]