Sulfur denitrator experiment

[Delphinus]

Update - I'm taking mine offline as soon as I have some time to get on my hands and knees to undo the plumbing and clean out the reactors and recover whatever media. In the meantime it's not hurting anything so I'm leaving it there but it's definitely no-good.

About two weeks ago I finally gave up hope altogether this would ever work out for me, so I decided I'd take it offline at the earliest opportunity (but left it on in the meantime since it wasn't hurting anything ... not doing any good, but not doing any harm, so I figured it could wait until I was looking for something to do).

Anyhow I stopped testing my effluent on Feb. 20. Sometime between then and yesterday the input line vapour locked and the siphon stopped so there was no flow. When I discovered this I did a nitrate test of the water in the reactor, and lo and behold I got (for the first time since NOVEMBER) a zero nitrate reading on the water inside the reactor.

So, I thought I'd give it one last kick at the can. Restarted the siphon and reset the drip rate to one drop per second.

24 hours later (tonight), the effluent is back up to 20ppm. (Tank is 35ppm.)

So ... I guess that's it. I'm done with this for now. My guess is that there's just not enough media, despite what people keep telling me that "its enough media" I think the numbers just tell a different story.

Psyire, how much media do you have in yours? Can you do a volume calculation? Even just a rough guess, like "1/2 container of LSM" or "3/4 container" or "2 containers" or whatever is fine. I'm just curious to see how it compares - you said your reactor was much bigger than this reactor so I'm curious as to how much bigger it really is.

[Psyire]

Sorry to hear....


But to answer your question.. 1.5 gallons (containers) of LSM. With a recirculation flow rate slighly lower than yours with the Mag 2. (this is where I believe your problem lies, but that's just my opinion) If you lowered your recirculation flow I think you would get closer to where you want to be. With a high recirculation flow I think you are getting oxygen present through your entire reactor. This is somewhat proven by your zero nitrate reading once your input was cut off.

I would be tempted to swap out that Mag 2 for a ehiem 1048 or something similar. Trust me, the results of lower nitrate are quite nice, as I have corals growing again that had stops for many months. It's made me a believer... so far.

[Delphinus]

You're probably right in that I should try a smaller pump, but I figure if I'm going to have to spend money to buy a pump I might as well also try a bigger reaction chamber because the small volume of media is really a wildcard I don't trust anymore. I could try swapping out this reactor for my calcium reactor and make this one a calcium reactor again or I could try to build something. I suppose I could try to find an Eheim 1048 or I wonder if I should just put a 1/2" ball valve on the output of the mag2 and set it to 1/2 or 3/4 to slow down the flow.

Oh well .. I went into this wondering if I could convert "this calcium reactor" into a functional denitrator, .. now I know the answer is "no - not this calcium reactor anyhow (but not to say one couldn't use another one)."

The next thing I want to work on is upgrading my G3 into a meshwheel and reworking the sump. I think it will be an interesting project. Maybe after that I'll look into trying the sulfur again, but for now I'm shelving the idea.

[Psyire]

Originally Posted by Delphinus

I wonder if I should just put a 1/2" ball valve on the output of the mag2 and set it to 1/2 or 3/4 to slow down the flow.

Why not do this before shutting the reactor down? It's a simple thing that may work perfectly... (and cheaply)

[Pansy-Paws]

Have you tried shutting the recirculation down for a few weeks to see if the media will seed more completely? I don't have recirculation myself, just a single pass through of the LSM (0.6% of the tank volume ... about 3 gallons), so I'm somewhat suspect of that component of the design simply because I've not used it myself. I do, however, have recirculation on the aragonite media to get the pH back up before re-entry into the tank ...

I've shut my unit down several times in the past six months, and during start up the key to reseeding is ultra ultra slow flow, and leave it there until you have a slight smell of sulphur (with your nose right at the outflow), then crack the flow up a bit (wait a few days for the slight sulphur smell to return), and repeat ...

The amount of sulphur relative to tank volume won't affect the zero nitrate formation (after all, the bacteria don't know how big your tank is), but with less volume you definately need slower flow, which can be a major obstacle. If I was to speculate without seeing your system, I'd say the flow is likely the factor that isn't correct yet and as a result, complete seeding of the media has not occurred. As well, if the length of the denitrator isn't at least five times the diameter of the cylinder, the drip rate would need to be even slower than usual to compensate and create the anoxic zone. I use a gate valve for flow control, but those can get spendy.

That zero reading you had would indicate that a super slow flow rate would give you success. For readings, I've found that the Jungle test strips aren't affected by the sulphur. Of course, you need to extrapolate between colours rather than simply reading from a digital device

I still swear by these devices for FOWLR systems with high bio load and heavy feeding.

[Delphinus]

New update ... This has taken a bit of a backseat since March but I got it going again recently.

I've reworked the sump baffles, I've modded the skimmer to meshwheel, and now I use a 100micron filter sock on the sump intake. I'm now happy with the tank and removed what I hope were stumbling blocks in its ability to process bioload.

I took offline my other calcium reactor from the other tank, and replaced it with the smaller unit (left it as upflow). Using the larger reactor now, I modded it to upflow (switching the tubes at the bottom, using a T at the top of the chamber for the intake, cut some eggcrate and got a filter at the top).

It's -just- large enough to hold an entire container of LSM media now.

The recirc pump is a mag5, the feed pump is a lift pump (looks like an airpump but for water) that pushes water, and flow control via an irrigation dripline needlevalve.

Using a 5g bucket of outgoing water change water, which tested at 21ppm NO3, I cycled the reactor (@1 drop per second) on the bucket and the results look like this:

(Date/Effluent Nitrate ppm)
2007/04/29: Not tested (source water = 21ppm)
2007/05/02: 15
2007/05/03: 10
2007/05/04: 7
2007/05/05: 0 (source water now = 4ppm)
2007/05/06: 0 (source water now = 0ppm)

So ... quite an astonishing change from the previous trials. Of course this is on a small volume of water with a static nitrate condition (ie. not being produced), but still, a reduction of 21ppm to 0ppm in 7 days to me seems pretty keen.

I'm more convinced than ever before that the problem in the previous attempts was inadequate media volume.

Incidentally the lift pump seems to be the answer to producing a nice slow flowrate. It has been rock steady at 1 drop per second all week and what tickles me most is that it's a $10 item.

I moved the reactor over to my 110g tank tonight, which tested at 33ppm NO3 (there is a good chance of error on this reading as I did it with the unit uncalibrated on the high range reading, but it is definitely in the ballpark).

It will be interesting to see how it does now that it's moved over into "production mode."

[Psyire]

Awesome News/Results

Glad to see you are giving it another go, hopefully with much better results this time around. I haven't posted any updates to my situation, but things are still ticking along just fine with my reactor and my tank has never looked better. (and tested better)

[Delphinus]

Murphy's Law had to kick in sooner or later.

This week's run started off promising, but has ended not so good.

Date/Reactor Effluent/Tank nitrate readings:
May 7: 10 / 33
May 8: 1 / 31
May 9: 0 / 31
May 10: 14 / 34
May 11: 21 / 32

Not sure what happened on the May 9/10 there, but ever since then Murphy's Law has been showing up. Haven't been able to keep a consistent drip rate since then either, so last night I switch the feed pump / flow control valve around so that the pump is Aqualifter pushing into the reactor (as before) but with the dripline irrigation valve on the output.

Worked awesome to get a nice steady rate, but, it looks like the reactor just can't handle the pressure. Although I have my flange bolts as tight as I can make them, water just seeps through the flanges like crazy. I wish I had noticed this last night (or this morning). I came home from my bike ride today (I left before the tank lights turned on), checked on the tanks and found a good 5-10gallons on the floor. Argh. And, I found that the intake tube had wiggled its way out of the sump and the reactor was just full of air bubbles. I figured at this point the bacterial cultures are a total writeoff and I have to start the cycle from the start all over again.

Bummer. For an inanimate object, it sure seems as if it's quite determined not to work. I tell you, between the nuisance crabs (who just killed a fish on me .. I have to declare an all out war on the little bastards now), the out of control nitrates and the out of control phosphates, and faulty test meters, my patience for this hobby is wearing dangerously thin.

Oh well, next week's another week, we'll see where it goes...