reef-pi :: An opensource reef tank controller based on Raspberry Pi.

[Ranjib]

Reef-pi is designed to run even without any network, so not having network is not an issue for reef-pi as long as you are not using network based drivers.
For the use case you are referring , it’s commonly known as external monitoring. I know adafruit has something called trigger that can notify you if reef-pi stops sending data. That will exactly what a typical external monitoring would do. If there’s a public Mqtt service we can do that with mqtt based metrics as well, since 4.0 release. In all these cases you need an externally hosted service /cloud like thing

Example : https://www.cloudmqtt.com/docs/alarms.html

 

[Ranjib]

Thanks everyone for the assistance. I have reef pi installed and was able to link one of my kasa hs300s in the software. Just figured I would give an update on my case. The whitebox probe cutouts are made and 2x 1/4" trs females for temp probes. I have to open up the pi connectors. FYI, if you use a whitebox hat, the pi will sit around 1/8" off the case, so your plug holes need to fit the entire plug housing, not just the plug ends. Waiting for my ds18b20s to arrive today.

Looks awesome, :-) . Thank you for sharing

 

[elysics]

Reef-pi is designed to run even without any network, so not having network is not an issue for reef-pi as long as you are not using network based drivers.
For the use case you are referring , it’s commonly known as external monitoring. I know adafruit has something called trigger that can notify you if reef-pi stops sending data. That will exactly what a typical external monitoring would do. If there’s a public Mqtt service we can do that with mqtt based metrics as well, since 4.0 release. In all these cases you need an externally hosted service /cloud like thing

Oh yeah, reef-pi did its job admirably in my absence, alkalinity didn't budge and the tank is fine.

A notification that data isn't sent anymore wouldn't help in my situation, I quickly figured that out on my own, I want to have fallback ways of sending data so that I know why no data is being sent, not absolutely necessarily providing the full Ui server. So I can figure out if I need to drop everything I am doing/get someone to drive over, or whether I can just leave it for a few days.

I might use a surf stick with one of those free low data simcards and see if I can get a fallback to work.

E: there are different scenarios with different likelihoods, requiring different levels of response:
Internet infrastructure problem
Router problem
WiFi interference
Power outage
Flipped breaker
Crashed/dead reef-pi
Burned down neighbourhood
And so on

In particular, it is important whether the tank still has power or not and additionally, whether reef-pi controlling the dosers is still running

 

[sfgabe]

I know whats happening, this is a common issue when the reef-pi starts before network is up during power up or reboot. You can fix this by updating the systemd unit file for reef-pi to make sure it’s started after network is up

Cause a script to execute after networking has started?

I am relatively new to systemd and am learning its architecture. Right now, I'm trying to figure out how to cause a custom shell script to run. This script needs to run after the networking layer ...

unix.stackexchange.com

Thanks @Ranjib - This is probably what my issues was as well. The MQTT setup has been working well for me in the past week or so. It seems like whenever there's a restart it sometimes takes up to 20 - 30 minutes to fully connect, and it is retrying that whole time and sending errors. I'm not sure whether its the sender or receiver that's causing that much of a delay, but once I restart and just let it go for a while it will eventually link up again.

 

[Sarlindescent]

Wiring done. Just need to solder the 1/4 trs to the temp probes and mount the box. That will be tomorrow's project. Need to rip out the apex to mount this. I also taped over the hdmis, for the time being since I am not ready to tackle that yet.

 

[Ranjib]

Oh yeah, reef-pi did its job admirably in my absence, alkalinity didn't budge and the tank is fine.

A notification that data isn't sent anymore wouldn't help in my situation, I quickly figured that out on my own, I want to have fallback ways of sending data so that I know why no data is being sent, not absolutely necessarily providing the full Ui server. So I can figure out if I need to drop everything I am doing/get someone to drive over, or whether I can just leave it for a few days.

I might use a surf stick with one of those free low data simcards and see if I can get a fallback to work.

E: there are different scenarios with different likelihoods, requiring different levels of response:
Internet infrastructure problem
Router problem
WiFi interference
Power outage
Flipped breaker
Crashed/dead reef-pi
Burned down neighbourhood
And so on

In particular, it is important whether the tank still has power or not and additionally, whether reef-pi controlling the dosers is still running

for what is worth , adafruit has a nice gsm addon that you can use to attach any oridinary simcard (like mint mobile) and give the pi direct internet address

PiPhone - A Raspberry Pi based Cellphone

This project takes you through building a functional cellphone using off-the-shelf parts,

learn.adafruit.com

and then there is this: https://www.adafruit.com/product/4521 , satellite phone that will work from most part of the world :-)

 

[Ranjib]

Wiring done. Just need to solder the 1/4 trs to the temp probes and mount the box. That will be tomorrow's project. Need to rip out the apex to mount this. I also taped over the hdmis, for the time being since I am not ready to tackle that yet.

you dont want to test the probes before mounting? What are the three probes you are considering?
Looking good

 

[InactiveAcct]

Got my reef-pi build started this week, really impressed. Started with just getting the Pi going and connected to an HS300. Was working really well for a day, then today it wasn't working anymore and throwing an error that it couldn't get the output pin when I tried to turn equipment on/ off. I set the hs300 to a static IP and verified that it was still in the same IP position and reef-pi was calling that position.

I deleted the hs300 driver and added it again with a new name; however, I'm noticing that it seems to be calling the pins from the deleted driver when I try to load new equipment, and it's still throwing the same error. If I delete all drivers it's still pulling up as having the HS300 0-5 pins still available to add new equipment. Any tips much appreciated.

Thanks!

 

[Sarlindescent]

you dont want to test the probes before mounting? What are the three probes you are considering?
Looking good

I would like to test them before, but at this point, my apex classic is useless. The display does not work and the nic is dead, so I cannot adjust anything. My ato has been on an outlet that is permanently off and my scheduled outlets are only working 1 hour on and 2 hours off(this is not what it should do).

For the whitebox board, 8 have 2x ph probes on the isolated outlets. The third, non-isolated is a rtd temp probe. I bought it before I new it wasn't supported, so figured I would see if I could integrate it with reef pi. The 2x 1/4 trs are standard ds18b20 temp probes on a pre-configured resistor board.

Edit: Additionally, I mount all my equipment, whenever possible, using threaded inserts and bolts. This prevents me from stripping out the cabinet if I need to remove something and add it back to the same location.

Edit x2: Additionally, using the kasa bars, i can set some conservative timers to put me in a better spot in the event that I have to tinker with the probes.

 

[Ranjib]

Got my reef-pi build started this week, really impressed. Started with just getting the Pi going and connected to an HS300. Was working really well for a day, then today it wasn't working anymore and throwing an error that it couldn't get the output pin when I tried to turn equipment on/ off. I set the hs300 to a static IP and verified that it was still in the same IP position and reef-pi was calling that position.

I deleted the hs300 driver and added it again with a new name; however, I'm noticing that it seems to be calling the pins from the deleted driver when I try to load new equipment, and it's still throwing the same error. If I delete all drivers it's still pulling up as having the HS300 0-5 pins still available to add new equipment. Any tips much appreciated.

Thanks!

did it stopped working after you rebooted pi? if so , this is due to reef-pi starting up before network. I just mentioned how to fix it in few posts earlier.

 

[Ranjib]

I would like to test them before, but at this point, my apex classic is useless. The display does not work and the nic is dead, so I cannot adjust anything. My ato has been on an outlet that is permanently off and my scheduled outlets are only working 1 hour on and 2 hours off(this is not what it should do).

For the whitebox board, 8 have 2x ph probes on the isolated outlets. The third, non-isolated is a rtd temp probe. I bought it before I new it wasn't supported, so figured I would see if I could integrate it with reef pi. The 2x 1/4 trs are standard ds18b20 temp probes on a pre-configured resistor board.

Edit: Additionally, I mount all my equipment, whenever possible, using threaded inserts and bolts. This prevents me from stripping out the cabinet if I need to remove something and add it back to the same location.

Edit x2: Additionally, using the kasa bars, i can set some conservative timers to put me in a better spot in the event that I have to tinker with the probes.

reef-pi does not have a driver for the rtd temperature probe, but it should be possible to integrate it by employing the filedriver (which reads sensor value from a file), and use atalas scientific python library to read the sensor and write it in a file

 

[Sarlindescent]

reef-pi does not have a driver for the rtd temperature probe, but it should be possible to integrate it by employing the filedriver (which reads sensor value from a file), and use atalas scientific python library to read the sensor and write it in a file

Cool. I found the atlas ezo global driver library last week before I went out of town. Will have to give this a try. They have a whole suite of modules from ph to temp, salinity, orp, O2, CO2, and others. I personally don't see the need for them given price and my particular setup, but with whitebox boards being stackable, added functionality is almost always a plus.

 

[InactiveAcct]

did it stopped working after you rebooted pi? if so , this is due to reef-pi starting up before network. I just mentioned how to fix it in few posts earlier.

yes but it still wasn’t working after rebooting pi

 

[Ranjib]

yes but it still wasn’t working after rebooting pi

did you change the systemd unit file??

 

[InactiveAcct]

Wiring done. Just need to solder the 1/4 trs to the temp probes and mount the box. That will be tomorrow's project. Need to rip out the apex to mount this. I also taped over the hdmis, for the time being since I am not ready to tackle that yet.

wow very impressed at how tidy this looks

 

[InactiveAcct]

did you change the systemd unit file??

I just did. But of course the issue was something much simpler, I didn't update the connectors after updating the driver. Hopefully the systemd update will prevent losing the connection in the future.

If others need to update systems, the more detailed process is post #30 by Ranjib here:
https://www.reef2reef.com/threads/tariq’s-reef-pi-journey.666421/page-2

 

[Ranjib]

Happy new year everyone. Thank you for making this awesome community

 

[Urtoo]

@Ranjib sir, thank YOU for making this community possible!!

 

[codster]

Hello all! I just set-up a reef-pi (v4.0) I got for xmas, and love it so far. I have a temp probe and pH probe, but hope to add some dosing pumps and a PWM dimmer for my BlackBox LEDs in the future. I have one small question which I can't seem to figure out.

The graph ranges for the temp and pH seem to always show zero. The ranges are so large, its hard to get any useful info from the graphs. I waited a day (1 min intervals on both probes), but they still have insane y ranges. I do see other people's graphs and see their ranges are fine... I think in older versions of reef-pi you could set the range value at the bottom of the probe page (under the graph), but I can't see that on my reef-pi...

Is there a way to manually set the ranges on the graphs? Or how long do I have to wait until the ranges will correct themselves (1min*24h = 1440 data points so far, and it still won't correct, and no problem data are in the graphs, they are just flat lines).

Ideally, I would like temp to be 73F -> 83F, and pH to be 7.7 -> 8.5.

 

[Mikeneedsahobby]

Hello all! I just set-up a reef-pi (v4.0) I got for xmas, and love it so far. I have a temp probe and pH probe, but hope to add some dosing pumps and a PWM dimmer for my BlackBox LEDs in the future. I have one small question which I can't seem to figure out.

The graph ranges for the temp and pH seem to always show zero. The ranges are so large, its hard to get any useful info from the graphs. I waited a day (1 min intervals on both probes), but they still have insane y ranges. I do see other people's graphs and see their ranges are fine... I think in older versions of reef-pi you could set the range value at the bottom of the probe page (under the graph), but I can't see that on my reef-pi...

Is there a way to manually set the ranges on the graphs? Or how long do I have to wait until the ranges will correct themselves (1min*24h = 1440 data points so far, and it still won't correct, and no problem data are in the graphs, they are just flat lines).

Ideally, I would like temp to be 73F -> 83F, and pH to be 7.7 -> 8.5.

This is on the list of things to be changed. Right now it can’t be adjusted.