Backyard Aquaponics
A place to discuss aquaponics

Hi guys, I'm a software developer and I was looking for a project to learn automation on and get into Raspberry Pi and Arduino. I thought a garden would be a good project and then I found out about aquaponics. I'm hooked before I've even started and after reading some books and anything I could get my hands on on the web, I think I've gotten a basic understanding of the concept. Now that I'm in between jobs and homes, it is the perfect time for me to lay the groundwork of the infrastructure of the electronic part of my future system: the server that connects and rules it all.

I got my ideas about what I want it to do, but when I started concepting it, I thought I could go opensource with this and contribute to the aguaponic community. Software development is more than just coding; getting your specifications right is half the work. With that in mind I don't decide what the specs are, but I should get the community involved. So here I am, asking you what you would want out of a server in your aquaponics system. To that end I've defined a goal for this project

The server should facilitate aquaponics in any way possible.

Please go bonkers with this goal. The main goal means that it is not only used in monitoring, but also advices the user what plants, fish etc to pick based on the knowledge it has. For example: if it knows its location it can advice which fish species can thrive in the local climate. If it already has measurements, it should be able to tell you specificly which species should be fine to use for your system. It should be able to notify you when some parameter goes beyond the safe levels. There should be documentation on setting it up, for people without experience with webservers. That kind of stuff. Everything for the completely noob and the grizzled aquaphonics enthousiast.

I hope to get some nice suggestions of features you'd like to see, which I can include into my development roadmap.

Welcome Todorus! To start, you should build a system sans automation to get a feel for it. Then you will get a better idea of what really needs automation. There are already a number of users working with Arduinos and Rapberry Pi's. I would certainly look at their work.

I understand your point that I must decide for myself what I want from an automated platform and build something using that knowledge. The result will be that it's tailormade for my system. I've seen other users of the forum that are working on that, but that isn't the point of this project. I'd like to build something that can be used by others. I want to contribute.

I'd like to design a standard server which is useable by any aquaponics practicioner from backyard to commercial. I intend to publish the API and instructions on how to install it on a Pi (or Ubuntu) and how to use the API on an Arduino. This way people only need to work on the code that is specific to their setup (the controllers) and have the server taken care of. It will also considerably lower the entry point for noncoders.

For all that I need input from people who just aren't me, to give me the full range of perspectives that people approach these projects. From a software development point of view, it isn't any different than any other project: get the specs, translate them to requirements/tests and implement them.

I'd also recommending building the system first, and then figuring out what needs automation.

You could automate fish feeding. I was going to do this before I built my system, but there is something oddly satisfying about feeding the fish and watching them eat.

Other than that, I want to further automate the topping up of the water, so it's all done automatically. I need a proper water tank first though.

Sensors are fun too. Check my sig. If I could find an easy way to sense Ammonia, Ph, Nitrate, and nitrites, I'd be a happy boy. But so far everything I can find for those jobs is horribly expensive, and doesn't seem to have the bugs ironed out yet.

Using a webserver, I'd love to set up a time delay camera/webcam on the grow beds. You could set it up to have to push a button on the webserver page, make it feed the fish, or top up water, but it's a two second job to automate it with timers/sensors once they get past a certain level.

For strict webserver though, I'd recommend going with the Rasp pi. I tried it with the arduino, and it couldn't handle it without crashing. Newer model arduinos with the intel stuff could probably handle it better though.

If you really wanted to get tricky, you could use sensors on the GB, and when the water is at a certain level, it'll cut off the pump. But to be honest, a bell syphon is a brilliant invention, leave the pump running 24/7, and it seems to work nicely. I'm a tinker, but I see no need to fiddle with that.

water chemistry and temperature monitoring, charting
flow control/monitoring, alert when pump fails
feed fish

There's already a half dozen threads on what people have done and want to do. there's also already commercial products that do this. Maybe you should check them out first and then if noone has done what you want the way you want it, then you have a reason to start your project and a goal in mind.

brian

Automatic water top up is easily done with float valves

Ill probably get battered for saying this but Im so anti-automatiion its not funny.

AP for me is a hobby that gets me and the kids out of the house and into the garden, being active and learning about fish and veg. In this hyper speed automated life we live in- AP is one of those things that get me away from all that. I get the techy stuff is fun and Im involved in a lot of that in my day job but when I get home I get great pleasure in throwing a handful of feed in my FT and just watch the fish swim around.... and after a system establishes thats about as hard as it gets.

I enjoy reading these threads and discoveries but I doubt Ill ever have any electrical automation on my system (other than back up of course). I hope you are successful though and Ill be watching with interest.

Well Charlie, let me join you on your side of the fence then. IMO KISS is the main theme for Aquaponics. Sure automation of AP is a great add-on to the hobby but it is really not necessary. Don't get me wrong. I graduated as a Mechatronics Engineer so automation is what I learnt. But most importantly what I learnt throughout my career is that you don't need to automate everything and you cannot automate everything. Especially not Aquaponics. By natural design, AP is a simple system with numerous variations but the main principle remains the same. After all, would AP or for the matter "gardening" be as rewarding and therapeutic to the human "soul" if we have equipment telling us what we need to do next? Where's the fun in that?


Sent from my iPad using Tapatalk

Hi todorus,

I also wish you well, if you can produce something that makes people's days easier that would be great.

The only problem I see is that the biggest automation need in AP is automated testing of the water parameters and like someone has mentioned already there aren't any sensors readily available for that. The rest of the system is up to the user but I suspect even a luddite like Charlie might be tempted if they could have a warning message at work that the tank is too hot and the ammonia is climbing. This however seems impractical at this time. Then again no one has ever made a big impact by producing something that is freakishly easy that anyone can do.

Good luck.

Regards, Martin.

Unfortunately for me, automation is going to have to become a bigger part of my system. Moving back to Perth means i will only get 2 out 4 weeks to physically check the system. I sure as hell am not going to pack my system up, so alternatives have to be found. Any clues i can give myself to help the mrs rectify problems is a win for me. She is not interested in learning about fish keeping.

A large part of the automation comes from design. We all do it, knowingly or not. Think shell grit, filters, top up floats etc. A well designed system will not need much input to keep running reliably. Especially a seasoned system, which tends to be very stable. As any of the more seasoned AP people around here will know, once the system is stable it is not uncommon to check the water parameters monthly if at all.

todorus, I have already started working on the project you have described. PM me for more details if you are interested. I mentioned it in another thread recently, but it went by unnoticed. It is a hosted on cloud9, so i can show you the source is you wish. It is a FOSS project i have named Aquaponolog.

@general
I understand that electronics aren't for everyone, as some like to be in touch with your garden in their way. I got the same vibe when I explained aquaphonics to my girlfriend, who loves to work in her (conventional) garden every sunny day she can. With that experience I find it a bit surprising to defend my want to tinker to a community which loves to build DIY systems (that are automated, using syphons, valves, pumps and a little gravity). From my perspective, I don't see anything strange about throwing electronics in the mix.

I do try to follow the KISS principle. Electronics should add functionality to the system. If you replace systems that can be used by more dependable and/or simpler solutions, you're just adding complexity. My first release would be focused on monitoring. Notifications are a second step and actuators a distant third, as there are simpler solutions for most.

With that bieng said: I hope I can avoid the discussion of "to automate or not to automate", as this is a more personal decision and should be weighed by looking at what someone wants from his/her setup.

@Columnmn
Good call on the webcam! I hadn't anticipated other data than integers or floats. I'll add binary data to the list.

@bcotton
Thanks for the tips. I'd like to build something which is n00b friendly, both in setup and flexibility. With that in mind, I'm not entirely impressed by the projects I could find on this board as they're specific to their setup and/or hard to install. I'm certainly interested in commercial solutions you've come across, so I can take a look at their approach, as I consider them the competition.

@MartinC
Thanks for the love!

@rsevs3
You posted during me writing this up. I'm interested

My project comes in 3 parts:

Hardware:
What ever you want, arduino, propeller, pic etcbasically just doing the data grabbing from the sensors and formatting the data. Initially using a serial link to a PC/RaspPi/Beaglebone for comms.

Python Script:
A simple script that polls the micro and dumps the results into a MySQL database. Run every x time using a cron job.

PHP Server:
Provides a front end to easily view the data. Also generates the python script depending on the connected sensors.

It isnt far enough along to show everyone, but anyone interested in development can take a look.

Hi todorus,

As a fellow aquaponics entheusiast in the Netherlands, it is great to have you join the group and share your experiences. I got hooked into the idea of aquaponics at the end of last summer and was determined to get started immediately. In addition to desgining the AP system and sourcing all the parts, I also needed to bulld a greenhouse to get through the Dutch winter. As a software engineer and electronics hack, I also wanted to monitor some of the environmental factors in the greenhouse. Here is what I have discovered in the first 6 months:

- I want to move back to California or perhaps move down under to Australia! I am jelous of all the AP posts that i see.

- With the exception of my humidity controller in the Greenhouse, automation has been a luxury. The sensors and image updates at https://xively.com/feeds/1205695996 give me a bit of piece-of-mind when i travel. Especially the sound sensor that tells me that the bell siphon is working.

- Suprisingly, the chemistry of the FT has been VERY stable. I take manual measurements every other week and the numbers are always spot on. All 40 of my goldfish have survived and grown with no deaths. I used fishless cycling with ammonia from AH and everything was stable in 2 weeks. pH was my only issue. The city water is relatively soft with a pH of 8. After about 3 months, the it stabalized to 6.8.

- My greenhouse is too humid. Despite adding a relay controlled fan to my raspberry pi, it has been very diffcult to keep the relative humidity at acceptable levels.

- Sun light has been an issue. My back garden is northern facing and the greenhouse gets no sun from 1 November to 1 February. i put in some basic floresent lights from Gamma but I think I would need more. The good news is that the Sun is back. The green house hit 30c several days last week.

- In the first few months of operation, Mint grew the best and i was able to harvest weekly for tea until December. Once the sunlight want away, it nearly died.

- I tried lettuce which while growing slowly has reacted well to the artificial light. humidity is an issue however.

- The one green leafy plant that has done well despite the humidity are strawberries. while there is no fruit, they continue to grow. for next winters maintenance period, I will go heavy on the strawberries to keep all the Nitrates under control.

Now that the sun is back and we are approaching the growing season, I am rather excited to reinvest more time in the hobby. I am planning to build some grow towers in the next month to put in place in April.

I sent you my phone number. feel free to call. i would be happy to give you a skype tour of the setup.

Mark

I read this topic and found it interesting because I have helped a friend set up an AP system close to my home, but he travels a lot for his work so he has to rely on his wife and young daughter to look after the system while he is away.

I live close by and so I am available if needed, too. I got a call a couple of weeks ago from his family to ask if I could help catch two fish that had travelled over into the sump from the main tank. We got them back in, but after that he and I were discussing the possibilities of a pump failure, a leak, a low water condition, automated fish feeding, and agreed that sensors and monitoring would be helpful.

I have since tested a remotely accessible web camera and an arduino server which could potentially give monitoring and remote control functions to his system. I have some parts of it working, but there is a lot of possible issues with designing such services to interface with an AP system. Many of them are related to the design of the AP system in question, too.

I agree with others that PH and other chemistry would be good to monitor, but I am less concerned about that than I am with the obvious major potential problems like mechanical failures.

I understand that electronics are not necessary, as demonstrated by the many systems up and running for years without them, but it make the possibility of leaving the system for several days or weeks much more attainable and convenient.

Cheers!

Hey rsevs3, your project has me interested (coming from a PHP/MySQL coding background)..

Could one poll the micro with a php script running from a CRON job or is python definately required?

I've got one foot on either side of the fence on the automation side of things - I love keeping it as simple and natural as possible but also like the idea of 1 man being able to run a lot more stuff than normally possible (ie less labour) with the aid of appropriately chosen tech etc

Prob not required for your average backyard setup though.. most of my vegies are grown in soil and wicking beds but I love and always will love AP, it certainly has its niche area and benefits