DIY Heater controller advice

[Nolan Shinn]

I was going to try to make my own heater/chiller controller. I've found a bunch of all purpose temperature controllers on amazon and I was wondering if anyone has a suggestion on which brand or controller to get. Anyone which past experience have any suggestions? I was most seriously looking at this bad boy:
Inkbird All-Purpose Digital Temperature Controller Fahrenheit &Centigrade Thermostat w Sensor 2 Relays https://www.amazon.com/dp/B00OXPE8U6/ref=cm_sw_r_cp_api_pn4OybVXSMCYY

Thoughts? I imagine pretty much all of these controllers have about the same temperature stability (+/- 3 degrees or so) though it's a hard number to find. Is that kind of spread acceptable for a reef tank?

 

[fragit]

I was going to try to make my own heater/chiller controller. I've found a bunch of all purpose temperature controllers on amazon and I was wondering if anyone has a suggestion on which brand or controller to get. Anyone which past experience have any suggestions? I was most seriously looking at this bad boy:
Inkbird All-Purpose Digital Temperature Controller Fahrenheit &Centigrade Thermostat w Sensor 2 Relays https://www.amazon.com/dp/B00OXPE8U6/ref=cm_sw_r_cp_api_pn4OybVXSMCYY

Thoughts? I imagine pretty much all of these controllers have about the same temperature stability (+/- 3 degrees or so) though it's a hard number to find. Is that kind of spread acceptable for a reef tank?

I built one, with this controller and it works great! Make sure you get the one with the plastic sensor as the metal even if it's stainless will rust. Also make sure it's got enough amps for the equipment you plan on running. Other than that it's a fun little project and mine has worked flawlessly for about 6 months now. Good Luck

 

[Bbaz123456]

I don't know anything about this but I do have an inkbird humidistat that is garbage. Temperature is probably easier to sense so it may work just fine.
Let us know, because $17 sounds pretty good if it does work

 

[Nolan Shinn]

I built one, with this controller and it works great! Make sure you get the one with the plastic sensor as the metal even if it's stainless will rust. Also make sure it's got enough amps for the equipment you plan on running. Other than that it's a fun little project and mine has worked flawlessly for about 6 months now. Good Luck

Awesome to hear! I think I'll try it out! Good advice on the plastic sensor

 

[Nolan Shinn]

I don't know anything about this but I do have an inkbird humidistat that is garbage. Temperature is probably easier to sense so it may work just fine.
Let us know, because $17 sounds pretty good if it does work

Will do!

 

[Skydvr]

I've got a similar one that I have had running off and on for a few years now. For a while I used it for my mixing bucket to keep from over heating the water. They don't lose their settings when they power down. It has been on my tank for the past two with no issues that I have seen.

There are tons of them, all re-branded I'm sure with no difference in parts or feedback logic. Search STC-1000. You can even find them 2 for $15-18. At least they are finally available in Fahrenheit, mine are in Celsius.

There is a guy on one of the homebrewing forums that developed a new program for them so they were better suited for some of the fermentation profiles. I think he may have set it up with fuzzy logic so it learned the proper cycling profiles, but that is kind of overkill for an aquarium heater. I have been considering seeing if I can set it up with an external clock to vary temperature slightly throughout the day to mimic temperature variations between night and day.

 

[cypho]

I use solid state relay that can turn on/off the heater many times per second. And a variable speed fan DC fan blows air from another room at the sump. Together they maintain the temperature rock solid.

My target temperature taken from a real-time sensor data from a weather buoy in Hawaii so the target temperature fluctuates over the course of the day/year. (the raw data is colder than I want in the winter and hotter than I want in the summer, so I dampen the swing a bit with a weighted average to a fixed set point).

The tank temp is the one that I target. The sump temp shows you what is going on. When the tank is warmer than the sump, I am cooling, and when the tank is cooler than the sump I am heating.

 

[cmcoker]

Here's my diy build with one of the ink bird controllers, it's on my QT tank and plugging along nicely.
https://www.reef2reef.com/index.php?threads/253261/

 

[Nolan Shinn]

Here's my diy build with one of the ink bird controllers, it's on my QT tank and plugging along nicely.
https://www.reef2reef.com/index.php?threads/253261/

Awesome!! Thanks so much

 

[reefwiser]

Just use a Ranco controller. Its made to do the job and is fool proof industrial temperature controller
I use temp controllers all the time in my work in industrial control. The style you are thinking of using will work its just not water proof so make sure to install totally away from water. I use the Ranco because it is water proof .

 

[Skydvr]

My target temperature taken from a real-time sensor data from a weather buoy in Hawaii so the target temperature fluctuates over the course of the day/year...

Nice!
I have been considering doing something similar along with simulating moon cycles (coordinated with the actual cycle). I am curious to see if I can get corals to spawn in the tank like they do in the wild by simulating natural queues. Depending on how this next tank gets set up, I may try to mimic tides with a closed loop.

I had no idea where to get the data from for temperature fluctuations, thanks.

 

[reefwiser]

You need to be following Jamie Craggs and the Captive Corals Spawning facebook group. Lots more to spawning than just moon cycles.

 

[Skydvr]

Maybe if it wasn't on Facebook. I avoid social media for security and privacy issues.

Also, the more people set up these Facebook groups to discuss technical information, the more difficult it becomes to actually share and access ideas and information as the discussions get burried when new threads begin.

I know there is a lot more to it. The cycles are part of the natural queues from the little bit of research I have done.

 

[JBradford]

Interesting. I have one of these controllers on a shelf in my garage but hadn't ever thought about doing this with it. Might just have to happen.

 

[redfishbluefish]

I remembered seeing @cmcoker post on his DIY, and was going to link that....but he's already responded and linked. I'm following because I want to do this DIY as well. Please post your build with details. Thanks.

 

[cypho]

Nice!
I have been considering doing something similar along with simulating moon cycles (coordinated with the actual cycle). I am curious to see if I can get corals to spawn in the tank like they do in the wild by simulating natural queues. Depending on how this next tank gets set up, I may try to mimic tides with a closed loop.

I had no idea where to get the data from for temperature fluctuations, thanks.

My main lights, moonlight, and pumps are all synced to in real time to the same geographic location as I am pulling temperature data from. Direction and amplitude of wave-maker pumps and sump pump fluctuate with the tides (Big waves at low tide - waves breaking over the reef, and smaller waves at high tide - the waves pass over with less effect, when the tides are running, the pumps create stronger more uni-directional flows that alternate sides depending on the direction of the tide).

I get the temperature data here: https://tidesandcurrents.noaa.gov/api/

I started off downloading sunrise, sunset and tide predictions from the web too. But eventually I learned how to do offline calculations of the expected state of the sun, moon, and tide. In addition to being more robust (continues to work when the internet is down) my calculated values are more precise/granular than the published data. I use these two libraries for the sun, moon, and tide calculations.

https://github.com/sam-cox/pytides
https://pypi.python.org/pypi/pyephem

 

[JBradford]

My main lights, moonlight, and pumps are all synced to in real time to the same geographic location as I am pulling temperature data from. Direction and amplitude of wave-maker pumps and sump pump fluctuate with the tides (Big waves at low tide - waves breaking over the reef, and smaller waves at high tide - the waves pass over with less effect, when the tides are running, the pumps create stronger more uni-directional flows that alternate sides depending on the direction of the tide).

I get the temperature data here: https://tidesandcurrents.noaa.gov/api/

I started off downloading sunrise, sunset and tide predictions from the web too. But eventually I learned how to do offline calculations of the expected state of the sun, moon, and tide. In addition to being more robust (continues to work when the internet is down) my calculated values are more precise/granular than the published data. I use these two libraries for the sun, moon, and tide calculations.

https://github.com/sam-cox/pytides
https://pypi.python.org/pypi/pyephem

That is awesome! What type of controller are you running o program all of this?

 

[cypho]

That is awesome! What type of controller are you running o program all of this?

it is a DIY controller with a raspberry pi for brains

 

[VBMike]

My Inkbird controller melted. Luckily, it didn't burn my house down.

Here's my Amazon review: https://www.amazon.com/review/R2EOT6VXXXA4TU/ref=cm_cr_rdp_perm?ie=UTF8&ASIN=B00QQ02U42

 

[billwill]

DIY heater controller sounds like an oxymoron I have to say. Why not just trust the controller in the heater itself in that case? I've got a jci A419, which is same class as ranco. It's not worth the risk in my opinion.