Nutrient Pulse Reactor (DIY DyMiCo filter)

[OzIA]

Can you repost that graph. It doesn't show up in the post, or at least, I can't see it.

I looked at Klaus's graph of the water treatment system and it shows the same thing you are describing but I believe this was a fresh water reference, not salt so I am not sure how things will translate. In that graph the knee occurs at -30 mV.

I got the quinhydrone calibration solutions yesterday and ran a quick calibration (more of a practice than real calibration). I plan to do a good one today. Apex has advised me that the calibration factors, (an offset and slope) are stored in the channel and will remain until overwritten. I should be able to change the type to ORP, calibrate and then change back to pH and have those factors remain. We will see.

Also the filter is already starting to clog again. I am going to try once more time, removing the media and installing a bigger process pump. I got a Silent 1.5, which has a similar footprint to the one I am currently using but much better flow. I am also going to switch to a vinegar / CaOH mix (50/50) for my carbon solution. If these changes don't prevent the clogging, I am not sure where to go next. It seems that the more efficient pump has actually caused the clogging to occur faster.

Despite all the issues, I set up the Dingo Down version yesterday. I repurposed bio balls for my course media and used a 400 gph process pump. I made a few design changes. The lids are "showers" to defuse the water flow. Two trays will hold the sand so they can be easily removed. I added a heater and temp probe but it doesn't look like I will need it. I also moved the ORP probe and fitting to the overflow area for easier access.

Aaron

 

[dartier]

Doh. So no one can see that graph but me? Bad Google.

Lets try this again. I will look at this post from a browser not logged into my Google account and see if it shows.

Notice how the first 3 cycles, and especially the fourth, look a lot like your graph (except not inverted like yours).

Edit: yep all good, I can see the graph from Safari.

Dennis

 

[OzIA]

How accurate do you think that ORP probe was? I can see the first portion of that graph looking a lot like what I am seeing but I am well beyond where the -100 mV knee is taking place. That 11 on my graph equates to -230 mV.

How high were the nitrates when this graph was recorded?

I just completed the calibration on the ORP probe. Let's see how that affects things.

Aaron

 

[dartier]

I can't remember if I tested the ORP for accuracy. I "think" I compared it against a 400mv standard. I was not too worried about the accuracy at the time. I was cycling the sulfur denitrator offline using tank water. The nitrates were quite high, 50 ppm or so. Not like yours are currently.

Dennis

 

[OzIA]

Well, I have the other unit up and running now. This is the top down or Dingo down version that hangs on the back of the tank. I repurposed some bio-balls for the course media and built a couple of boxes to hold the sand above.

I got the Hach Total Phosphate kit in today and ran a quick ortho test. It was almost unreadable, somewhere between 0 and .02. The poly and organic tests are more complicated and require an acid / heat digestion process so I may fool around with that tonight.

Attached is the ORP wave from the new system.

 

[OzIA]

An update on the phosphate test...

I checked the organic phosphate levels with the Hach and got the same reading, near zero / undetectable. This test is agonizingly long. It requires a hot acid digestion, which means you add acid and boil the sample for 30 minutes, let it cool and then you have orthophosphate, which is now detectable and testable with the kit. So if i did it right, the amount is not measurable.

As a sort of experiment, I took the skimmate, from the skimmer I reinstalled in the third chamber and tested it for both ortho and organic phosphate. Since the unit had clogged again, I have been running a purge cycle (flush on all the time) for about 4 days now, with no carbon being added. As the bacteria mat broke down, it released a slight cloud that the skimmer was able to remove.

This was not like the dark skimmate you see most of the time. It was sort of looked like very watered down milk.

Anyway, the test found the organic phosphate in this skimmate to be 0.5. The ortho phosphate was about .04. My assumption here is that the bacteria is taking up the phosphate, in both forms, which can then be skimmed out.

I didn't check for meta phosphate.

 

[OzIA]

I was finally able to reproduce the nitrate knee effect. My over feeding finally got the nitrate up to 1.0 ppm and phosphate to 0.16. This was also immediately after I refilled the organic carbon container. The new mix contains 500ml CaOh, 500ml vinegar and 50ml vodka. My assumption is that available carbon was higher and combined with the increase in available nitrate, finally allowed the nitrate to be fully consumed and visually captured by the ORP probe.

 

[OzIA]

I have been thinking about the reaction and how things are affected by the dose and flush volumes and I feel like using the ORP values to trigger these may be the wrong approach.

It seems like a better way would be to have the system adjusting the dose and flush volume, based on the average slope of the ORP values.

For example, you can see the slight added volume or flush time, caused the peak values to shift more positive. You can see this effect in #67.

Additionally, if you exceed the volume by too much, you get a long pause before the bacteria population recovers enough to start consuming nitrate and driving the ORP negative. (ref #70) This appears to diminish over time as the bacteria population increases. One of the most difficult issues to overcome is the tendency to clog if the bacteria slime gets too large. On the Dingo Down version, the unit ran for about three weeks before I started to see a noticeable difference between the levels in the process chamber and the course media. The difference started out at about ½ an inch but is now nearly 4 inches. I can see long white strands of bacteria in the pump chamber, waving in the flow.

If the trend has been build to a point where it is starting to come into an acceptable range, then the program would make a slight adjustment to level things off. (Ref #71) Or if the opposite were true, additional carbon would be dosed. The trend would need to be consistent for several hours before any changes would be attempted. At some point, the length of the cycle could then be adjusted.

This would reduce the complexity of the algorithm and eliminate the need to react during each cycle to the normal deviations that occur. With this mindset, the primary thing you need to know is, what is the ideal ORP range for the reaction. Possibly the best way would be to run a few tests to determine where the nitrate knee was occurring and then use that as a set point.

Also, and then later, the length of the cycle could be reduced as the bacteria population increases. Since there appears to be a proportional relationship between the available nitrate and the amount of carbon needed to consume it. This control method solves that issue as well. As nitrate increases, the average negative peak ORP values will begin to decrease unless more carbon is added to match. This means that the system will also continually adjust, based on the available nitrate.

 

[OzIA]

Update

Update

I just wanted to update this thread and check to see if there has been any progress with the others that were actively following...

I have three variants up and running now. Two are HOB versions plus the original conversion that I started with.

I continue to have difficulty with the bacteria bio mat clogging. Even the second version, which used bio-balls instead of the Reborn media; clogged after about three months of operation. I have added a new routine on the process pump, which turns it on for 30 seconds to a minute, then turns the pump off for a few seconds. This creates a sloshing effect from the pump side to the course media side. This seems to be reducing the clogging effect of the bio mat.

I also changed the way the systems work. All three are now using this simple process:

Process pump: On for 30 to 55 seconds, then off for 5 to 10 seconds. Also, if the Flush pump comes on, the process pump goes off.

Flush pump: Every 10 minutes I look at the ORP value. I set a target of 0 mV. If the value is below 0 mV (negative) I will cycle the flush pump for 10 to 15 seconds. If the value is positive, there is no flush that cycle.

Carbon: Three minutes after the flush cycle occurs or would have occurred, I look at the ORP and either add carbon or pass to the next cycle, again, based on the 0 mV target.

This program is the simplest of the routines so far and seems to be yielding acceptable results.

 

[dartier]

Just want to check in and say that my project is still underway. I know that I have not posted to this thread in quite awhile (my bad), but my tank is now in place, although empty.

Just working with my contractor on getting supports in place for a track system (for an above tank sliding work platform) and height adjustable light rack.

Ventilation went in last week. Hopefully a 110 CFM fan is going to be enough to mitigate the humidity given off by the large tank. I have a dehumidifier standing by incase it is needed as well.

Dennis

 

[OzIA]

Well, I came back from holiday to find my basement wet. One of the canisters on my Kent Marine (Coral Life) RODI system split near the threads. This is the second canister that this has happened to. I was home for the last one. This time I wasn't so lucky. I had a catastrophic failure in my Frag tank, which was auto-topped by the RO system and lost everything. I'm going to replace the remaining canisters before they fail too.

On a lighter note, the other tanks are doing fine. I came back to find nitrate at near zero in both, despite being on auto feeders for almost three weeks. Cycling the process pump seems to be keeping the slime buildup to a manageable level. Phosphate continues to climb when the GFO starts to become depleted. It seems GFO is still a necessity. I was really hoping to get away from that expense.

Aaron

 

[dartier]

Well, I came back from holiday to find my basement wet. One of the canisters on my Kent Marine (Coral Life) RODI system split near the threads. This is the second canister that this has happened to. I was home for the last one. This time I wasn't so lucky. I had a catastrophic failure in my Frag tank, which was auto-topped by the RO system and lost everything. I'm going to replace the remaining canisters before they fail too.

On a lighter note, the other tanks are doing fine. I came back to find nitrate at near zero in both, despite being on auto feeders for almost three weeks. Cycling the process pump seems to be keeping the slime buildup to a manageable level. Phosphate continues to climb when the GFO starts to become depleted. It seems GFO is still a necessity. I was really hoping to get away from that expense.

Aaron

Oh, no that is terrible Aaron. Coincidentally, I just placed an order through BRS This morning for a bunch of RO/DI stuff (new filters, membranes, 150GPD upgrade, auto flush, etc.) and one of the things I ordered was an automatic cut-off for leaks. I am switching to an Avast barrel tender as I expect my new tank to require far more frequent top-offs and I did not want to have to remember to keep an eye on my supply barrel. Included in the order is the LBS-10 Leak Block sensor, that threads into a normal 3/8 connection. Something like this may help if your recent issue were to re-occur.

I am glad to hear that your filters are still working well. I have been considering my options on what to do about phosphate once I get my new tank running, and the NPR filter done (still working on the tank). I plan on trying an experiment to see if I can use a 80 Gallon with some fast growing hard coral (like cyphastrea) attached to the main system through a second pump in the return chamber of the NPR filter. The thinking is that a few times a day the flush water would be returned to the coral tank rather then to the display tank, and then a small amount of NO3 would be dosed into the coral tank. The hope is that the coral tank would deplete a portion of the PO4, and whatever NO3 is unused (in the coral tank) would get processed by the NPR filter the next time the coral tank is flushed. I plan on trying some eggcrate to give the cyphastrea plenty of surface to grow on. Should be a fun experiment.

Anyway lots to get done before I get to that point.

Dennis

 

[OzIA]

Oh, no that is terrible Aaron. Coincidentally, I just placed an order through BRS This morning for a bunch of RO/DI stuff (new filters, membranes, 150GPD upgrade, auto flush, etc.) and one of the things I ordered was an automatic cut-off for leaks. I am switching to an Avast barrel tender as I expect my new tank to require far more frequent top-offs and I did not want to have to remember to keep an eye on my supply barrel. Included in the order is the LBS-10 Leak Block sensor, that threads into a normal 3/8 connection. Something like this may help if your recent issue were to re-occur.

I am glad to hear that your filters are still working well. I have been considering my options on what to do about phosphate once I get my new tank running, and the NPR filter done (still working on the tank). I plan on trying an experiment to see if I can use a 80 Gallon with some fast growing hard coral (like cyphastrea) attached to the main system through a second pump in the return chamber of the NPR filter. The thinking is that a few times a day the flush water would be returned to the coral tank rather then to the display tank, and then a small amount of NO3 would be dosed into the coral tank. The hope is that the coral tank would deplete a portion of the PO4, and whatever NO3 is unused (in the coral tank) would get processed by the NPR filter the next time the coral tank is flushed. I plan on trying some eggcrate to give the cyphastrea plenty of surface to grow on. Should be a fun experiment.

Anyway lots to get done before I get to that point.

Dennis

I wonder how different that option would be that just running an Algae Turf Scrubber?

I have been contemplating another option.

Since the bacteria seems to be building this slime mat that can become clogged, I was thinking that if it could be periodically removed, that might be a solution to the phosphate buildup issue.

I tested some of the slime and it is loaded with phosphate that seems to be bound up in the bacteria bio mat. If you could break it up and filter it out on say a weekly basis, it would remove phosphate from the system.

Aaron

 

[dartier]

I wonder how different that option would be that just running an Algae Turf Scrubber?

I have been contemplating another option.

Since the bacteria seems to be building this slime mat that can become clogged, I was thinking that if it could be periodically removed, that might be a solution to the phosphate buildup issue.

I tested some of the slime and it is loaded with phosphate that seems to be bound up in the bacteria bio mat. If you could break it up and filter it out on say a weekly basis, it would remove phosphate from the system.

Aaron

I have used an ATS in the past to export nutrients. The parts that I did not care for was the mess, and the possibility of tossing out lots micro fauna on a regular basis. Now I have a fish room, so the mess would be less of a problem, but I can still see myself picking through a bunch of GHA for micro fauna. As well with an ATS, you have to stay on top of it to keep the GHA growing in the ATS and not the display.

If you are getting that much density of bacterial mats, your filter must be tending more towards carbon dosing rather then denitrification. Carbon dosing exports nutrients by skimming the produced bacteria where as denitrification uses less dense bacteria populations operated in more of the anoxic range to cause the bacteria to liberate the oxygen molecules from the nitrate leaving only nitrogen gas.

If you have the means to export (a portion) the bacterial mats easily, then it can be a means of exporting phosphate as well. However you may need to supplement (dose) nitrate in order to balance the nutrients and allow for more growth to be exported.

Dennis

 

[OzIA]

Skimmers have been turned off so the export of the Nitrate is not a result of skimming. I am operating (cycling) the media bed from 100mV to -100mV. I don't know if that is an indication of anoxic range but it seems to be working.

i am all in favor of less of a bacteria mat but I haven't been able to completely eliminate it.

 

[dartier]

I guess the question would be to see if your nitrate stays low and your bacterial colony remains stable and does not continue to grow. Since the bacterial colony remains about the same (as far as volume occupied), and your filter continues to reduce nitrate, then the explanation is that denitrification is occurring as your skimmer is not running so no export of bacteria can be taking place. -100 mv is low enough for denitrification to occur.

How are your corals reacting to your water parameters?

Dennis

 

[Thadude0226]

Npr

Npr

Hello I was scheduled to get the Dymico system shipped to the States but they are having production issues. So I have elected to try to create a build like you guys and was wondering how are the systems performing and do you have pics of the build and some advice

 

[OzIA]

I have never been able to get these systems to run long-term. They clog up when I try to drive them and I have to back off on the organic carbon.

I haven't heard much from Dartier in a while. He was working on a new tank (Blue Window) but I haven't seen an update in a while.

I started another thread, (search for Automatic Biomass Scrubber) if you want to look at that one. The ABS ran for about 6 months and then I developed a leak in the remote DSB so it's been limping along without the DSB while I work on the newest ABS version. I haven't posted anything about that one (Mark IV) yet.

Good luck.

 

[Thadude0226]

Ok so it's better just run a calcium reactor in a DSB sump?

 

[OzIA]

You can probably do that but if you add a form of organic carbon to the DSB, I believe it will clog.

As I understood it, the DyMiCo system did put CO2 into the second course layer and melted some of that as a form of calcium reactor but I only built one of my systems that way.

DyMiCo also used two layers in their system. The top sand layer is where the ammonia was converted to nitrite then to nitrate or nitrification process was accomplished. Then that water filtered down to the coarse layer where the bacteria, fed organic carbon under low oxygen conditions, broke down the nitrate (denitrification) into oxygen and nitrogen gas.

All of the systems I build and tried worked well in the beginning. All removed nitrate almost immediately and I assume increased the bacteria populations in the water. I can't be sure about that last comment as I have no way to count. The problem is that a biofilm built up in every system to the point where flow was severely limited. At that point, the systems capabilities were diminished and nitrate would start to increase. Once I manually cleared the biofilm, the systems would start working again but clog more quickly the next time. Also phosphate would start to increase. I believe this was because there was no way to remove the phosphate in the same way as the nitrogen and just have it bubble out.

What i was trying to do with my current system (soon to be upgraded) was to automatically remove the biofilm which I believe is where the phosphate is bound up.