Backyard Aquaponics
A place to discuss aquaponics

I set up my system using an IBC tank in May this year. I decided that I would attempt to make the system operate from solar, did a few calculations and purchased a 90W panel and a solar regulator from eBay. I have a background in computers and electronics so an Arduino system was roped in to act as a timer for the pump, switching the pump with an FET (solid state) switch. The first pump I used was a submersible in-line 12 volt pump, I have since changed to 12 volt bilge pumps with varying degrees of success, but basically you get what you pay for.





I ran the system for 3 weeks with some lettuce and caulies in, hoping to kick start the nitrogen cycle prior to introducing fish, I noted a slight increase in Nitrite and then a levelling off and concluded (wrongly as it turned out) that the cycle had completed and I was all good.


I purchased 30 Rainbow Trout fingerlings from the Ballarat hatchery and introduced them to the tank. All was well for a few days and then the nitrites went through the roof, I did a couple of 50% water changes over the next couple of days but the nitrites were still off the scale, then came the day I found a dead fish in the tank. As I fished it out I found the other 12 dead ones. I quickly did another water change and added a bag of salt to the tank. Over the next couple of weeks the nitrites cam down and everything has been tickety-boo since.




The single 90 watt panel I was using, mated to a 100 amp Deep Cycle battery was coping with the load early on, but a week of overcast skies quickly showed up it’s weakness and another panel was purchased. The CM3024Z regulator seemed overkill initially but the ability to instantly view battery voltage, charge current and pump current has proved very useful in the trial and error system I have been using. It gives the ability to check the pump current, which rises as the pump gets clogged and/or the bearings give up the fight, and change the pump before it fails completely. The Rule pump I am currently using seems to be the best of the bunch; I also have a Johnson which I only took out to go to a higher flow rate. The cheaper pumps from SuperCheap et-al don’t seem to be able to handle the constant work.

The biggest mistake I made was miscalculating the position of the sun in mid-winter, the neighbour’s new garage completely blocking the sun from the GB. Things are on the improve now that the sun is a bit higher. We had had crops of carrots, lettuce, cauliflower and parsley from the GB even with the bad sun. The trout are going great guns and are currently measuring 10 inches from tip to tip. I am planning on harvesting in November before the water temp start creeping up.





Cheers

Shane

Great looking setup mate, I'm a huge fan of arduino.. are you only using it for pump control or other monitoring/controls?

I'm waiting for the new Wifi Arduino to come out (this month?) and I will probably run a headless system monitoring the pumps temps etc. Toyed with the idea of Raspberry Pi etc but Arduino does enough for what I need at the moment.

The only thing aside from timing that the system does is to detect night-time (LDR) and change from 10 min on/20 min off to 10 min on/50 min off overnight to conserve battery power during the winter months.

I'm looking to crest the same level of automation when I upgrade to my stage 2 system. Wifi is a key element for myself as well, then accessibility to the controller won't be a problem.

I haven't seen any rasp Pi systems as yet so I'd be interested to see the results if someone did.

I've heard that some people have had difficulty with using light detectors for night time detection. Rather they switched to and arduino clock and had preset "night times"

For this problem I had considered loading in a data set containing all the sunrise and sunset times for the year (or shorter periods as necessary) and having the arduino consult the data set each day to determine when night is. This would save me from have to calibrate and maintain an LDR

Calibrating the LDR is not all that difficult for this application as I figured I only really need to know when it is getting dark/light. I set up the program with some debug code that sent the detected value of the LDR to the serial port, then whilst monitoring the reading simply put by thumb over the LDR. This gave a reading for bright and dark.

There really is no real calibrated light level value, compared to reading say temperature, and it doesn't really matter if you get a false reading, the worst that will happen if it gets shaded at the time for a switch-on is I would miss one half hour cycle.

I initially tried to detect night by monitoring the voltage from the solar panels, but this proved a non-event as the PV controller regulates by switching on the -ve side which I use as a common (ground) on the regulator output thus the solar oanel -ve cannot be connected to the regulator output -ve. Suffice to say that after mucking about a bit with voltage dividers and other isolation ideas I went bank to KISS and used an LDR.

Yep, kiss is the answer and I understand where you're going with the LDR, I'd be interested to see how effective it is with the setup.

If you're not concerned with missing a cycle then your setup will work fine,

How much did the controller and its accessories cost?

The light sensor on my Arduino is great at night/day transition, but rather useless for discriminating sunny vs. cloudy days. When time permits, I have a Pi just waiting to set up for remote monitoring (currently light and temperatures are logged on an SD card).

I have a small solar panel, charge controller, and UPS batteries to play with, so I'd like to monitor the panel and battery voltages and perhaps the current drawn by the loads. It looks to me like the easiest way to do this is to hook all the sensors to the Arduino and have it push data to the Pi via serial. I'll have to resurrect my dormant programming skills and pick up some new tricks to do anything useful with the serial data, but that's part of the fun. Setting up a headless Pi with Wifi looks fairly trivial, though.

Nice work Shane! Looks like we've both gone through the same issues (bilge pumps, a week of cloudy days, etc)

The only thing I'd suggest is to get a brushless 12v pump instead, mine was about $50 from eBay and it should last ten years or so. Bilge pumps are lucky to last ten weeks!

Hi guys

Yes the cloudy day thing was a lot more of a problem than I thought it would be. Currently the panels are lying on a 10 degree shed roof, I plan to increase the angle prior to next winter to see if that improves the time between clear days capability. You will notice on the schematic that I have ann external supply capability to cover these emergencies. I have a switch-mode 12 volt supply I got from Jaycar. I have made use of the same two pin connectors throughout the system so that I can switch pumps quickly and/or connect a pump directly to the switchmode for emergency pumping.

The regulator was aroud $75.00 off Fleabay, the panels were $120.00 each, by far the most expensive part of the whole project, but I figured if I decide to ditch the solar Aquaponics thing I'll just use them for some other project; there is always another project. The case that the control electronics is housed in is an old 3.5 inch floppy case, we have stacks of them lying around unwanted now that noone uses floppies anymore.

I had read a fair bit on here about bilge pumps but thought I would give them a go as they are reasonably cheap and easy to replace if needs be. Supercheap and most fishing places sell them. Interestingly I pulled one of the failed Supercheap pumps apart and found that the bearings (which were shot) are easily replaceable. I haven't put the repaired pump back in the system but it will be interesting to see how it goes with a good set of bearings fitted.

Hi All

Bit of an update. Have added another grow bed to the system. Till now I've only had the one 'half-an IBC" GB siting on top of the FT. Although it seems to work okay the water is pretty murky, hoping an extra GB will improve this.

I am running two pumps, one for each GB, with the timer alternating between the two on a 7 on, 8 off cycle, though I will probably be dropping this back to 10min on, 20 min off. I figure by running two separate pumps, should one fail I'll still be cycling FT water through one GB and keeping the aeration up. The plants can stand a day or so without water but I don't think the trout would last long without aeration.





I'm hanging out for Melbourne Cup weekend, when I plan to start harvesting the Rainbow Trout......

Cheers

Shane

Hi Shane. That is a nice solar setup you have going there. Do you have an air pump in your FT?

Hi Cobus

No I rely on turbulence from the draining of the GBs to provide aeration, the return pipes are always clear of the FT surface. Seems to work okay, and with the couple of failures I have had with pumps (or my stupidity) so far, the fish let you know when they haven't got enough air.


Cheers

Shane

Nice system you've setup there Shane



Couldn't you tap in to the wires coming directly out of the panels to measure their voltage before the charge controller?

Looks great Shane, I am looking forward to following your thread as I am hoping to do a Solar system trial one day.

Good luck.

Humm 30 fish on a timed cycle.. with no additional aeration....

Hope the water temps don't get too high before Melbourne Cup day....

I'd be starting to eat your fish... now....