So months of discussion and no photos of the actual aquarium? That is really disappointing! 
I’ve been running a reef angel controller on my systems for several years now and have really liked the ability to customize my coding with it since it is based on arduino.
Being able to integrate all that into home assistant seems like a great idea. I would definitely like to see more details on what you have done. Would you be willing to share some of your code if you get the chance?
You are 100% right. I’m sorry.
Full disclosure : I’m actually holding down two professions at once, and am trying to break into the coral farming biz as a 3rd. Time suckage at the moment is huge, and then my wife returned from her summer holidays with her own agenda. lol
Nonetheless, I’ll post some pics here now…
Here’s better than a photo… a video.
The top tank is 2.5 meters by 60cm (H) x 50(D), and holds 750+ liters of water. The tank underneath is a 200liter - which at the time of this video was standalone with it’s own mini sump. One of the distractions that kept me busy jsut two weeks ago, was removing the sump & extra electronics and integrating the plumbing of this 200 liter tank underneath into the overall circulation system that drives the tank on top and another Anemone dedicated tank (not in this video) that also is in the same living room space, but on the other side of the TV. It took more than a week of planning, execution and fine-tuning water flow, because I had to re-engineer the height of the sump in the “support room” which is located another 6 meters away from these two tanks.
This change in my aquarium system necessitated some overhaul in my HomeAssistant setup. However, I’ve been giving a big re-think on how to overhaul my existing setup to make it more object oriented, to be smarter about dynamic changes to my setup.
For instance, previously Home Assistant was previously controlling the two seperate tank systems as complete seperate entities, which was what I wanted. But when I moved my 200liter tank into my existing 1300liter system, making the overall system 1500 liters - it would have been super cool if my 'AquaSystem" controller could have been “smart” about my migration of tanks into a single system, removing the need to change the Home Assistant config file. So if I said “these two tanks are connected now”, Home Assitant would say “Oh, I see there will be two Protein Skimmers and Two heaters on the same system then - which would you like to make the defaults for this new system, or use both?” Anyways, still giving that a bit of a thought how to best do that & trying to get some other stuff out of the way so I can get back to hardware hacking and improving the fail over and redundancy features first.
Overall, Home Assistant has proved quite a rather reliable controller for an aquarium this past year, as long as everything is built to ‘fail safely.’
On that note, I did have a hardware failure I should also mention. About 2 months ago, one of my network controlled APC power bars gave up it’s ghost. And it wasn’t the 220V relays that gave out - it was the management card controller. I’m still trying to figure out what part of the hardware is fried and if I can replace it, but I had to shuffle things around in regards to 220V control and replace with another make. Another time vortex that popped up and I didn’t plan for. Here again, I want to make changing out such hardware components a bit easier for the end user experience.
I’ll make some more videos showing off my Home Assistant setup & hardware constructions
Very cool. Lots and lots of euphylia in there. Love the torch corals. It looks like you’ve had success with raising the clownfish too. Are they big enough that they don’t require a rotifer diet yet?
Home assistant does almost seem like an ideal platform for an aquarium controller when you look at it. How I think I’m going to approach this is by keeping my reef angel controlling most of the main life support systems of my tanks still, but then start using home assistant to control other parts. For instance, I’ve been looking at the sonoff 4 chan controller and thinking that would be great for controlling pumps in my water mixing barrels. Home assistant may also be a good option for adding a backup heater (I only have one heater controlled by the reef angel right now).
If you were able to find a solution for home assistant with pH probes that would be great (I’ll probably do a little digging myself but I’m still pretty new to the whole thing). I currently have my pH probe monitoring the pH in my calcium reactor since the solenoid needs to close if the pH drops too low. So I don’t monitor the pH in my tank as closely as I should because the pH expansion for the reef angel would cost me almost $200.
I’m very excited I stumbled upon this. I think there is a lot of potential here!
I agree there is a lot of potential and I plan to start similar to both of you- slowly and with non critical items, with a ReefKeeper (old school!) manning the heaters.
I’ve looked into the pH meters and was waiting for someone on here to try it first as some options are more expensive than others (Atlas Scientific).
I did attempt to create a variable speed pump controller via HA and MQTT and have it semi-working… maybe I’ll do a write up when complete.
This will be useful for automating some flow patterns maybe but also for slowing down a pump if a float sensor detects a problem, etc.
So the advantage is I wouldn’t necessarily have to turn off the return pump entirely if something triggered (overflowed) while I’m gone.
We’d love to see that project J. I bought a reef keeper first then quickly sold it on ebay to go with a reef angel. I do not regret the move. The biggest thing that sold me on the reef angel was their water level sensor no other aquarium controller uses a pressure sensor to indicate water level so I can know where the water is in my sump by % not just high or low.
So I was doing some brainstorming on my project and came up with an idea I wanted to share and get opinions. I was thinking the sonoff 4ch flashed to work with HA would work great for turning on lights, pumps, etc much like the relay boxes on an aquarium controller work. The issue I was running into was I didn’t want to cut every electrical cord on the devices I wanted to control. It makes it so that’s the only way I can use these devices and its probably not ideal from a safety standpoint either really. So my solution was to build my own 4 channel relay boxes using standard electrical outlets and new work boxes. Sort of like this project: http://www.instructables.com/id/Arduilay/
Finally in fairness here are a few shots of my tank and fish room. I am in the process of a tank upgrade since we moved this summer. The display tank will be in the finished basement this is my old 90g that is temporarily set up to keep all the livestock happy.
Okay, this week I should be good for making a decision on Ph controllers and probes and getting the order out. Already been looking and will post some links to options I found and am considering / decide on.
I’m curious @jnvd3b what sort of variable speed pumps you’re working with? Do you have IP access to the pump or a hardware link?
It’s a variable speed DC pump
It does not have IP control on its own but you can control it via a 0-10v input, meant for third party controllers like the Apex. This would apply to most DC pumps including the Gyre pump which I do not have yet but plan to get one soon…maybe.
What I have attempted to do is basically copy the idea of the Apex VDM which is an add-on of the Apex meant for controlling these pumps.
What I have created (spent most of the weekend on this, at the expense of my wife, so won’t be able to finish this until next weekend at the earliest at which point I hope to do a short write-up) is a simple 2 channel DAC that pushes 0-10v signal to the pump’s ‘dumb’ controller via MQTT/Home assistant.
I can say that it works quite well so far on the benchtop but I haven’t actually used it on the pump yet. Total cost was under $20.
@jnvd3b I think you and I are thinking a lot alike on many aspects. 
I was looking to do the same with my Tunze Nano Stream pumps and while I haven’t built anything yet, I do have the Din5 plug pinouts mapped out now & the general build plotted in my mind now.
I understand the wife debt completely, and have had to pay that debt back many times on this project of mine already.
I do look forward to reading and sharing myself more here soon. 
So this is the first week I’m really back at my work bench and I’ve discovered much to my surprise, it seems I’ve been using my workbench as a ‘dumping ground’ for various items for the last 2 months. This will of course immediate attention so I can break out all my more delicate electronics gear & get to work. Hopefully I’ll get this wrapped up within 24 hours so the work can begin again.
But during the last two months I’ve been quite busy with my aquarium and a few substantial changes have taken place. The first, being I had to suspend rearing out all the baby clownfish my two pairs were producing - too much supply!
Secondly, I found it necessary to re-engineer the water conduits under our floor to support a 3rd display aquarium to the overall “main system”. Until that point, the 200 liter grow out tank for the baby clownfish was an independent standalone system. And since this 3rd display was already close to the ground (under the big reef tank) - I had to go through the trouble of lower the existing pipe fittings and even the sump, to make the water levels and flow rates work out & safe.
And while I previously had at one point up to three stand alone 10 liter Fry tanks, I also added one of those fry tanks also to the main system (I still need to update the “Tank View” to reflect the fry tank changes).
However, the expansion of the system to incorporate the 200 liter allowed me the first time to investigate my pipes for the risk of something (not found to be present, btw) that I had long wondered about: How will using my main circulation pipes as a food delivery path, negatively be affected by encouraged organic growth, buildup and blockage? I got to find the answer to that when I cut into the pipes I’d been using the last 10 years for that.
It should be mentioned, I did not just start doing this and achieve success, without a reasoned analysis long ago into the potential problems and how to successfully mitigate them by making some additional trade offs (read: investments of time and solutions). I’ll write more about this and explain in greater detail later, but in short when applied with the correct methodology to mitigate against the risks of impact, I found success without realization of risk (blocked pipes or accelerated growth within the pipe lines). And after my findings just a couple of months ago, I wrote those mitigations into part of the “Centralized Feeding Mode” that is part of my HA Aquarium Controller, so I don’t have to manually perform them anymore. For people with Interconnected tank setups, this could be an attractive and sustainable solution. 
That said, I am thinking of adding some flow meters eventually to the solution. I could see where it might be interesting to monitor flow rates over the short and long term.
Sadly, one of my 3 recycled APC Masterswitches gave up it’s ghost, and I had to substitute a couple of powerswitches instead, which I find aren’t as quite as responsive as the APC gear.
I’m looking at some more modern and perhaps customized replacements such as some of these for power control, since they’re certified for both AC/DC usage.
And lastly (for now) here’s the latest screenshots of my setup. As you can see, HA takes care of a variety of items, like co-ordinated mixing of Kalklwasser & dosing (so as not to do at the same time), automatically pumping collected skimmate after a certain level has been collected down the drain, temp regulation and monitoring of all the aquariums, water level monitoring (I’ll be adding more sensors for this soon) at various points, and even automated reactions if an event happens. If any of the Pi controllers goes off line, even elements of the network, it’ll get hard rebooted, etc. The water change automation has been awesome.
BTW, the Anemone tank is not overflowing as indicated here in the screenshots. My cat did that by chewing through the sensor wire & I still need to fix and better secure that wire.
On the topic of PH monitoring… this looks like it could be just what I might be looking for…Tentacle for Raspbery Pi. Oh yes.
That does look nice. What is the estimated cost after 3 Atlas Scientific chips and probes? I would guess $300USD? Not horrible j/w what the cost would be to implement.
it doesn’t look like you have to use Atlas’s probes. But it looks like they have some pretty high quality ones.
What are you intending to put on the tentacle board? Obviously pH, but also D.O. And E.C. for salinity?
I was looking to add Ph, and after that, ORP. Torn between EC and DO. Probably EC since I run Kalkwasser combined with Baling, as opposed to running a Calcium Reactor. This way instead of an Auto-Top off solution, a “Smart Top Off” solution could be created instead.
For instance, if say the Protein Skimmer goes nuts (something spawns gazillion eggs into the water) at 2am at night - or during the weekend get away with the wife. And the Protein skimmer’s cup is one of those setups that is automatically set to drain all waste down the drain (no auto cut off). And since the PS has gone nuts, it’s just pumping extremely wet skim-mate down the drain. Water level drops in Sump and Top Off Kicks in, and salinity start to go out of whack.
Doesn’t have to be a PS going nuts either. A small leak in the plumbing occurs and water level drops in the sump. Bad enough salt water is going to end up where it should not, but even worse if the Top off fills the tank back up with fresh water, ensuring further tragedy through hyposalinity in the tank.
A Smart Top Off solution could monitor the salinity as a sanity check to prevent the above from occurring. It might even take a decision to add fresh saltwater instead of Kalkwasser or fresh top up water, to bring the sump up to normal operating water level without knocking salinity off.
That all said, eventually and ultimately I’d like to realize support all the probe interfaces on HA.
Thoughts? other opinions? suggestions?
After seeing the cost of one DO probe ($200-$600+) I would also probably defer to ph, ORP, and EC
The problem I foresee is that the RaspberryPi Tentacle shield has 3 ports, but only 2 are isolated. The third port is for Temp or Flow measurements only. That’s a bit disappointing, and has me thinking about a possible alternative based on an Arduino Tentacle connected to a standard Arduino or ESP32 even. This setup could provide up to 4 isolated probe channels and a little fail-over independence from HA (i.e.- water top up operations and monitoring could continue uninterrupted while HA is being restarted or crashes), as well as not require the HA RaspberryPi to be near a water environment.
Of course then there’s some code to be developed on the ESP32 / Arduino but hopefully most of the basics are already there from other projects. Hmmmm…
Please go that route! i’m not a programmer by any means I struggle with arduino sketches and python even more so.
I think one helpful thing is that there are pre-built sketches you could use to set the tentacle up
@jnvd3b Just to confirm my understanding of your wish - you’re saying you’d prefer the Arduino solution? You think this would be easier for yourself to implement?
No I was joking go whatever route you are most comfortable with don’t let me sway your decision.
