Have you tried to bring all the temp probes back together at once in the same container of water controlled to say 77.3 and see where they measure in that container? If not, try that and let’s see if they’ve changed since deployment.
BTW, I’ve done this on mine a couple of times now & they’ve held their calibration rock solid. If yours don’t, you need to return them probably.
I’ll try that if I get a chance soon. I don’t know as its a calibration issue as again I look at the graphs and they aren’t reading the temps correctly aside from the curing tub. I’m starting to wonder if most of the probes didn’t like the saltwater. I didn’t put any additional covering on them and they look exactly like the probes that came with my reef angel (which seem to be working exactly as intended)
So I have been busy starting a new job and haven’t had a chance to do much with this lately.
I did put the probes next to each other in the same tank and take a screenshot
Aside from the sump temp, which I recall now was not in a tank and just in the air those are pretty close and pretty accurate to my display tank temp.
When I put them in their respective tanks I would like them however I get these graphs:
So my readings are very similar to what I posted before. The display probe appears to be somewhat working it is still different than what my reef angel probe is reading though and I’m fairly sure that one is accurate. The curing tub seems to be working as well since the heater turns on and off as commanded by my automations. The other two really don’t seem to be accurately measuring though.
Did you happen to use / change a calibration function in your python script while you had them in the same water column next to each other? If not, I’d recommend that. I think in one of my previous post about my script, you can find such an example.
You do also realize that the scale on the left side is changing dynamically based on what data there is to display? So in this way, the graphs aren’t one-to-one equivalent scales to each other. The Scales displayed are optimized for the data that needs to be displayed. I mention this because I’ve noticed this has confused some users in the past. Apologies if I’m pointing out the obvious tho. 
Yes actually the scales on the left solidify the point I was attempting to make further. On the two probes that I believe are giving erroneous data the temp range is less than .1 degrees. That is impossible for an electric heater to hold (well at least the common aquarium ones). So I’m just wondering why when I separate the probes some tend to not behave as they should? Because when they were all next to each other they appear quite accurate.
I’d say it depends. It’s actually crazy cold here now, so the Fishroom is a few degrees cooler than normal, and hence, today my QTank has been swing temp wise a bit more than it usually does, but I have seen as little as 0.05 or 0.1 degree of temp variation in my QTank. It depends on your hysteresis functional settings and your temp set points and the external atmospheric temp. Also, it’s possible if it’s a small tank your powerhead could be a stable heat source causing exactly what you indicate above.
How often is that probe taking readings in your post?
I also just noticed something while checking the temps again. Take a look at these graphs side by side:
They have almost identical “valleys” from 2000-0800, then again from 1200-1600. The tanks these are in are not linked in any way. So for some reason when the probes are separated from each other they do not all read as they should.
Next time I’m home I will try putting them all together in the same tank again to see how close they get.
Hmmmm… What’s the ambient atmosphere temp around those tanks?
That is strange tho. I’m just curious if something’s wrong with your circuit.
Can you provide a wiring schematic of how you hooked those temp probes up?
Ambient is about 71*.
I will have to look at the circuit when I get home. Around post 120 I do have a messy picture of it uploaded but probably not enough to get any idea what could be wrong.
Okay, I’m about to get pretty saturated with work for the next 8-9 days.
If I don’t make it back on here and respond soon, I should be back and free to follow up after the first week of March.
Hang in there.
No worries. I’m in training for a new job and that is taking the majority of my time.
Overall things are working, just would be nice if everything was working! lol
Hi,
Just wanted to say great thread.
I have been following along every now and then and I have found a ton of useful information.
Did anyone ever get Ph monitoring to work in HA without Seneye? Not a big fan of consumables
I am almost ready to get my project started for my reef but I wanted to include pH monitoring. I am still not sure if I am going to use just the Rpi or Arduino, Rpi and MQTT. I think the second scenario would be better for redundancy but I do not want to over complicate the whole thing.
Best of luck to all of you and keep those updates coming…
Hey, great thread.
I’d been building out some a custom aquarium controller for the last year (https://github.com/mcclown/AutoReef) but as I was expanding my 90l tank setup, to include an new 4ft tank, I realised I just wasn’t getting enough done on the automation system. A little bit of digging around led me to take a look at HA and this thread.
I just started building out what I needed, using RPi 3 B and Hass .io and I’m impressed so far. In one evening I managed to get 50% of the way to feature parity, with what I already in my custom controller. I’d guess I’ll have close to 100% of functionality sorted in a week or two, at the current pace.
@cowboy I noticed earlier in the thread that you talked about trying to automate some AI Hydras and having no luck getting an API from them. Did you ever get anywhere with that? I’ve managed to reverse engineer some of the AI APIs (really need to push that code to github). I’m thinking that I’ll add that as a new platform/component for HA shortly, as a project…unless someone else has done it already. Here’s the API abstraction that I built out for my Hydras, all the shown functions are working in my local build…just need to port the functionality over to HA.
https://app.swaggerhub.com/apis/AutoReef/AutoReef.ApiGateway/1.0.0
The tentacle add on for a Rpi seems to be the best solution I’ve seen posted here. Although I don’t know as anyone has posted any results of it working. It’s a bit of an investment up front but has decent expand-ability for salinity and ORP. It’s the way I’m planning on going if I continue to expand on my system.
But as you can see from this thread I barely know what I’m doing in regards to any of this anyhow! 
@mcclown Nice if you could make the HA plugin. I have AI Hydra’s too and although the included app is good integration is always nice to have. I will check out your AutoReef controller.
@Ciwyn I have seen the tentacle as well but I don’t want to be the trailblazer as I do not have much time to diagnose error’s etc… I want to be sure that it at least can work and then I will figure it out from there. It is a bit on the expensive side. If we are going to spend lot’s of money on this stuff then we might as well just get an Apex.
I am building an auto RO/DI tank level monitor / re-fill system with an Arduino, LCD and HC-SR04 ultrasonic sensor, maybe I will add some extra safety float switches.
So far I have only written the code(I hope, not tested lol) and have the parts but I have not assembled it yet. Have to get round to soldering onto some perf board and installing the water shut-off valve on the RO/DI inlet. Tunze make a unit to do exactly this but is costs >100€ for what is essentially 2 float switches, a solenoid valve and a micro controller with some preset timeouts.
@cliffdude I have that Tunze ATO…but it came with my tank. I live in Oz and they cost the equivalent of €200+ here. I was never going to voluntarily pay for that myself.
Sorry I was sucked away onto other things for awhile, but work duties demanded it and nonetheless was glad to see there’s been some activity on here anyways while I was away.
Something else that happened awhile I was away - I strolled by the 1 year mark of Home Assistant controlling and automating all my aquariums, as well as my house. That is over 100 electrical sockets and lights alone. Further to that are 24+ Temp sensors, 8 float level monitors and another dozen or so other inputs or sensors. HA has not just automated and monitored it all, it’s also raised alarms and saved our asses, a couple of times now during electrical blackouts in the middle of the night - forcing me out of bed and discovering everything silent and dark. And helped me quickly determine one of those failures was do to a heater that was actually starting to leak. And this last winter, a I experimented with outdoor farming of phytoplankton on my roof in near freezing, HA made it super simple to monitor my saltwater based Nannochloropsis phytoplankton with crazy precision on the temps and discover how far I could push my phyto up to the freezing point of salt-water & alerting me when it was finally time to bring it indoors. How did that go in the end? I moved my entire phyto farm to the roof this spring & shutdown my artificial lighting indoor farming to save electricity. And this was the first year we could really precisely control our electrical usage and costs & actually did succeed in saving a lot of money this last year because we changed our strategies - particularly in regards to heating 10 aquariums, and used HA to get us there.
HA as an aquarium controller? Not just doable, but even beyond practical. Even a luxury. I can’t tell you how wonderful it was when overhauling one of my tanks - and removing & displacing water in the Display tank - to leave HA to managing water levels in the sump as I either removed 100 liters in volume of rocks from my display tank & not have to worry about my circulation pump sucking air OR overflowing my Sump (located in another room, about 7 meters away). When HA would detect the sump unexpectedly approaching levels beyond a preset limit because I displaced a bunch of water with rockwork - HA would fire the purge sump pump to prevent an overflow. Gone are the days of accidentally flooding my fishroom floor.
@mcclown Oh very cool news indeed, thank you very much. No, I hadn’t worked on any of that yet. However, with that in hand, I should be able to get somewhere with coding a python or even bash/curl script to perform that functionality. The biggest challenge will be to find the time to sit down and do it. I sort of unexpectedly kicked off another Open Source project, of a completely different sort, and publicly committed before a live audience to get the code I’ve already got working and demonstrated, polished up and published by later this year. Oeps. LOL
Let me see where I can find an opportunity and I’ll see if I can knock out some basic functionality in python. I do want that functionality in HA, very much. Again, Thanks!
@cowboy I’m pretty close to publishing a python library for AquaIllumination lights, just a few small kinks to work out. Once that is done I’ll work on adding a HA platform for it and see if I can get it merged back into HA. It will be nice to get AI support, all in one place.
I’m also working on building out a library for the CoralBox doser, with the aim to also get this support into HA.
Work is a bit hectic at he moment, so I’m just fitting in work on this when I can. I’ve been impressed with HA so far. I love having control of the fishtank using Google Home. Very handy when doing water changes, no more fumbling with switches etc.
Id like to see a photo of a rooftop phytofarm thats a new one to me
