Open-Source Professional-Grade Aquacontroller : "SeaDragon"

[jcglisson]

I was a little dismayed with Neptune Systems Apex platform for a couple of reasons (proprietary, expensive, lack of options), so I decided to build one for myself!

100% open-source (P1AM is literally just an Arduino MKRZERO in IP20 housing.)
100% off-the-shelf components

 

Here's the latest video of the system:
https://youtu.be/GcSTdc1w4YA

 

My official build thread on AtlantaReefClub:
https://atlantareefclub.org/boards/index.php?threads/open-source-professional-grade-aquacontroller.1176280/

 

GitHub Repo.:
https://github.com/JohnnyG89/SeaDragon

 

Reddit:
https://www.reddit.com/user/TheReefySeaDragon

 

Youtube Channel:
https://www.youtube.com/channel/UC_xClikPNXl30xSlmCY1CwQ?view_as=subscriber

 

Some of the features:
Thingsboard MQTT API through Shiftr.io
Function block abstraction for autofeeder/dosepump control and scheduling
8x 120V outlet control
4x Temp probe monitoring
4x pH/ORP inputs
2x variable speed DC motor or PWM-controlled lights
8x switch monitoring
8x valve/actuator control
16x physical switches (auto/manual mode, Remote defeat, manual control)
Cost: ~$600

 

Building a stable, failsafe system is my #1 priority, which is why I'm not doing anything too fancy with the web interface right now. The web interface is going to be an auxiliary system that interfaces with the unit without affecting the core life support systems. I'm testing it on my tank, which I wouldn't do if there were any chance of something bad happening (i.e. if the controller goes down, all of the livestock will be fine). That's why I included the pushbuttons (start, stop, reset) and light tower (Red, Green, Amber, Blue), and the system will be 100% controllable from these buttons/indicators 100% of the time.

 

It's kind of a philosophical conundrum for me: On one hand, reefing is my passion and the chance to make money from it would be a dream come true. On the other hand, though, is the more that I "chase" the money will detract from the authenticity, which could diminish my passion for the hobby altogether.

Best-case scenario is that someone sees my project and wants to invest in it, so I could make a company and do this full-time and retain full control of the end product. A few people have expressed interest in purchasing units, but they've all got some strange feature requests that are just out of my core skillset (NOT a web developer, don't have the time and/or energy to fabricate multiple units on a recurring basis). Also, the "WhY DoNt YoU uSe ThE RpI" cult is a little obnoxious.

 

I have been running the Apex Jr. for almost 2 years now and I have absolutely loved it.
-No local network-related hassles (e.g. port forwarding)
-100% "out-of-the-box" functionality
-Easy programming
-Convenient 120V plugs

 

Some of the things I really didnt like were:
-Limited offering for IO modules: In terms of electrical control, it's a 120V mechanical relay or nothing. the P1AM has options for 5V, 12V, 24V Sinking/Sourcing Inputs/Outputs; 4-20mA and 0-10V Analog IO, Solid State Relays, DC motor drivers, etc.
-Proprietary ecosystem (not going to be extensible for hobby users)
-Very disappointed with the quality, durability, and versatility of the AFS... unless it's their proprietary food you're using!
-Cost per controlled signal is exorbitant (See: $250 for 8x 120V relays for Apex EB32 vs. P1-08TRS for the P1AM at $46.50)
-Connectivity with other web platforms is impossible
At the end of the day, $600 with this system is comparable to a $2000+ Apex system when you take functionality into account.

 

A lot of my solution was built around the ReefPi:
-Open-source software
-Open-source hardware
-Community-Driven by passionate people

Some of the things I changed:
-Every component is off-the-shelf. Even the core controller - the P1AM - could be "drop-in" replaced by the Arduino (since it's literally just an Arduino w/ ip20 housing). The only "contribution" of mine is the architecture.
-Since it's completely modularized, if any component has a problem, just that single component can be replaced. If you fry a single GPIO pin on your RPi, you've got to replace it entirely!
-Significantly fewer parts total, and each of them connects easily and durably. significant time saver as well. No soldering!
-No additional hardware to connect a new unit. For a standard 120V peripheral (i.e. return pump), just sheath/strip/ferrule the cable and screw into the terminal block.
-Enterprise/Industrial quality hardware.
-The ReefPi project seems to encompass the full gamut of features - control, monitoring, reporting, alarming, etc. I'm just covering the control portion to focus on long-term reliability and durability in the unit that actually interfaces the aquarium. So really I'm just breaking off the "control" part of ReefPi and trying to perfect it!

 

[mcarroll]

Very cool!  The main problem is that this is so far over most folks heads.  (Anyone who's not a programmer, really.)

 

Do you think you could build one and give it to a newbie reefer to use and they could use it without issues?  If not, I'd keep working util you get it up to that level of "slick".  

 

If you can get your interface to be "even slicker" than what's on the Apex, then you'll be onto something real!  Apex is WAY complicated, so that is a real competitive point you could have, if you can do it.