kcschwabe
Community Member
Sure looked like it
Making things that other people buy in this hobby..
- QUESTION OF THE DAY
- Thread starter revhtree
- Start date
- Tagged users None
CraigJoh
Community Member
I DIY out of necessity, because I have way too danged many hobbies, just not enough cash to go around.
I've build everything but the display tank (and that I got used) :
Stand,
sump,
skimmer,
lights,
hood,
frag tank,
frag racks,
ATO,
RODI storage/mixing (50 gal fresh, 50 salt, 50 waste for watering plants)
I've build everything but the display tank (and that I got used) :
Stand,
sump,
skimmer,
lights,
hood,
frag tank,
frag racks,
ATO,
RODI storage/mixing (50 gal fresh, 50 salt, 50 waste for watering plants)
cryptodendrum
Community Member
Most everything, except for the glass tanks themselves.
The single most impressive "thing" I've made for my hobby, is my own High Availability Aquarium Controller. It leverages a Distributed Controller architecture (5 nodes currently, but scaleable between 2-20+ nodes on demand and needs), which replaces traditional "remote probe boxes" (as seen with more traditional controller architectures) with full featured redundant controllers instead. As far as I have been able to determine, this is the first instance I am aware of of a Decentralized and Distributed aquarium controller architecture, that has existed. I'm aware of one brand that recently launched that copies this architecture model a bit, whether by coincidence or intentional, but by then my controller was on it's fourth year of operation already. But they also made some fundamental design choices I would definitely and deliberately would not do myself.
In my architecture, whenever and where ever I install a remote probe box, that also becomes a redundant stand-by aquarium controller "node", which has the fail-over capability to take over control of the aquarium in the case the primary controller were to fail for any reason (i.e.- power failure, software bug/crash, hardware failure, etc). It basically creates a Cloud-like architecture implementation for my aquarium controller in my own home, and without dependancy on the Internet.
When the primary controller is online and working as normal, all the "secondary redundant controllers" operate as remote probe boxes, forwarding sensor data on to the primary controller. But they are also always checking and monitoring the primary controller. If the primary controller fails and ceases to function normally & within expected operational parameters, one of the remaining four secondary controllers will be designated as the new Primary controller and take over from the failed previous primary controller, and the remaining secondary controllers will send their sensor data to the new primary that has just taken over.
Further, the new primary controller will attempt to self-heal the system, and attempt to restart the failed controller, first in software, and then finally by hard cycling the power supply. If the failed primary returns to service and remains stable in operation, the previous hierarchy will be restored: the new temporary primary will revert to secondary standby and monitor behaviour, and the previously failed but now restored primary will resume it's former role as the primary, once again.
In this regards, it blows all other aquarium controller architecture models out of the water in terms of reliability. In fact, as I do in my day job, I measure uptime of service availability of my controller, the same way I measure it for systems on Telephone networks and since 2017, I've measured a 99.993% total service availability of my controller. That puts it as good, or even better than some telephone networks, with the majority of that 0.007% of unexpected downtime occurring in just the first 2 years, when my software failure detection code was a bit less mature, and had fewer "failure detection" scenerios built into it.
Going to next level of Aquarium Automation...who's with me?
So I saw a few posts about controlling aquarium lights with HA and I was curious if anyone else has taken HA to the next level with their Aquariums. Either way, I thought I’d share a little project of mine - perhaps it could be inspiring to others. My Integration is a little insane, but that’s...
community.home-assistant.io
One of the primary benefits and problems to solve of this architecture model is the proverbial "My controller did it" failure situations we still read about on forums. "My aquarium controller freaked out and dumped all my CA into my tank!" or "My ATO went bananas and now there's water all over the floor." People buy controllers because they are supposed to prevent these kinds of things from happening. But they continue to happen because most aquarium controllers are not / were not embracing "Designed to Fail Safely and Securely" principles. I wanted my aquarium controller to have this principle embedded at it's core, from the ground up. And these kinds of failure scenarios do not happen to me, anymore. My aquarium controller does more than monitor and control the aquarium, it monitors and controls and will attempt to even heal itself. And it works so well, I use it to automate, control and monitor my entire house as well. It even talks without the aide of Siri, Google, etc, because it's architected to also work without the Internet. And my Seneye even talks to it directly, instead of Seneye.com
I like creating, making and innovating new ideas and solving new challenges. I also have a passion for making my hobby as sustainable as possible, for myself, and the planet.
Last edited:
Made with PETG FOOD GRADE FILAMENT
3 magnets on inside of tank and outside of tank from amazon.
Check this out at Amazon.com
LOVIMAG Super Strong Neodymium...
Programs are tinkercad and cura
Printer is an ender 3 pro
I don't know if they are the exact same, but I used similar magnets and they rusted within weeks. Be on top of them
Similar threads
-
- Poll
