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

[hhaase]

Hmmmm..... I think I'll take my dive into this via just building one up. This way I can get a good feel for areas that might be well served by having a dedicated PCB. I've never really used a pi before but no reason I can't start now. From what I can tell there are a few main system groups, which are all independent standalone systems right now. Is that a safe assumption?

Here's the systems I see at current in the adafruit guides that would line up as their own schematic/layout sets. Feel free to correct me at any point. I'd like to start doing up schematics for these individual units.

A main hat, which includes the external power controller, temp controller, and level controller. This should be very easy to incorporate into a single dedicated PCB.

A PH monitor controller. This looks like just a basic breakout board.

A dosing controller. Very simple looking single board system to be honest.

A lighting controller system. Most of this could also be put onto a single board, with the PCA9685 as a daughter-board stacked on top. Once the design is nailed down, given that adafrut stuff is typically also open-source based, it could potentially be integrated into a single board system, but I'd probably not do that yet. I also don't have lights that are compatible, but doesn't mean boards can't be made. Just that I can't test here.

 

[Blaxkin]

So now iam also finished my build

With 3.0 and pca9685 as outputs i integrated 5 12v outputs thanks for that

i designed 3 pcbs, two of them are stackable

a jebao dp4s can be directly connected.

thank you ranjib for reef pi :-D

 

[Ranjib]

So now iam also finished my build

With 3.0 and pca9685 as outputs i integrated 5 12v outputs thanks for that

i designed 3 pcbs, two of them are stackable

a jebao dp4s can be directly connected.

thank you ranjib for reef pi :-D

Woohoo . thats awesome . Thank you for sharing

 

[Ranjib]

Hmmmm..... I think I'll take my dive into this via just building one up. This way I can get a good feel for areas that might be well served by having a dedicated PCB. I've never really used a pi before but no reason I can't start now. From what I can tell there are a few main system groups, which are all independent standalone systems right now. Is that a safe assumption?

Here's the systems I see at current in the adafruit guides that would line up as their own schematic/layout sets. Feel free to correct me at any point. I'd like to start doing up schematics for these individual units.

A main hat, which includes the external power controller, temp controller, and level controller. This should be very easy to incorporate into a single dedicated PCB.

A PH monitor controller. This looks like just a basic breakout board.

A dosing controller. Very simple looking single board system to be honest.

A lighting controller system. Most of this could also be put onto a single board, with the PCA9685 as a daughter-board stacked on top. Once the design is nailed down, given that adafrut stuff is typically also open-source based, it could potentially be integrated into a single board system, but I'd probably not do that yet. I also don't have lights that are compatible, but doesn't mean boards can't be made. Just that I can't test here.

On spot with your assessment

 

[DawnfromPa]

Let me know when you master this and it comes to market. I will buy for sure just price it under 99 so I can afford it. I know this will be huge especially for all in ones. I use apps that were set it and forget it...and barely do much now.

 

[burningbaal]

So now iam also finished my build

With 3.0 and pca9685 as outputs i integrated 5 12v outputs thanks for that

i designed 3 pcbs, two of them are stackable

a jebao dp4s can be directly connected.

thank you ranjib for reef pi :-D

wanna send one to me!?

 

[Ranjib]

Hive minds,
I have just cut another release candidate, reef-pi 3.0-rc0 . This is the first release candidate for 3.0, coming two weeks after 3.0 beta release. This means we have fixed all the critical bugs. And unless some new blockers found (serious bugs ) we are on our course for final release. This is mostly a bugfix release since 3.0 beta with some ui improvements for the light module. Here is a summary of changes, taken straight from the release page,

• bugfix: reef-pi db command argument processing
• bugfix: jack remove ui was broken
• feat: expose reversible option for macro (required for control target)
• feat: light ui support for 3.0 api
• feat: dutch translation
• bugfix: temperature and ph chart current reading not showing
• chore: log driver initialization errors
• chore: improve internationalization coverage (connectors ui)
• feat: use selectable pins for connectors based on driver types
• bugfix: email to ui field under telemetry section does not accept user input

packages can be found in usual location: https://github.com/reef-pi/reef-pi/releases/tag/3.0.0-rc0

 

[Des Westcott]

Is one wire enabled and the sensors are detected by raspbian. Reef-pi can only show them and use them if its detected. If its not,, this is beyond reef-pi, and we have to look are circuit/raspbian configuration

They were working for about 5 days then stopped. So 1-wire must be enabled or they wouldn't have been detected and installed in the beginning. Not so?
I'll run Raspi-Config now and check though.

 

[Des Westcott]

They were working for about 5 days then stopped. So 1-wire must be enabled or they wouldn't have been detected and installed in the beginning. Not so?
I'll run Raspi-Config now and check though.

Update

Before checking Raspi-Config, I physically unplugged my sensors from the board. Rebooted and plugged back in. Reef-Pi now sees the sensors = win.

BUT because I deleted my "Tank Temp" from the Temperature tab, I needed to set that up again with Temp control. I can't save the settings because the UI tells me "The temperature settings cannot be saved due to validation errors. Please correct the errors and try again. " in the pink box. I also get the "hysteresis_less_than" notification in red under the Hysteresis box. I have tried various numbers for hysteresis.

Reef-Pi is measuring temp and plotting the graph now, but I can't add control.

Solution?

 

[PaulJ2303]

Hi @Ranjib
just tried rc0 everything looks good apart from my first temp sensor disappears when you reboot the pi ... i've attached ac couple of screen shots to help with any debugging as i know you've had this from a couple of others:



these screen shots are both from after a reboot and the tank temp disappears from my dashboard.

Edit:
if i then delete "tank temp" from reef pi and re make it ... i can then re initialise it in the dashboard and it's fine again until the next reboot
hope that helps.

 

[Mike Kearns]

So now iam also finished my build

With 3.0 and pca9685 as outputs i integrated 5 12v outputs thanks for that

i designed 3 pcbs, two of them are stackable

a jebao dp4s can be directly connected.

thank you ranjib for reef pi :-D

That looks awesome. Did you 3D print the box?

 

[Blaxkin]

That looks awesome. Did you 3D print the box?

Yes its 3d printed

 

[Des Westcott]

Update

Before checking Raspi-Config, I physically unplugged my sensors from the board. Rebooted and plugged back in. Reef-Pi now sees the sensors = win.

BUT because I deleted my "Tank Temp" from the Temperature tab, I needed to set that up again with Temp control. I can't save the settings because the UI tells me "The temperature settings cannot be saved due to validation errors. Please correct the errors and try again. " in the pink box. I also get the "hysteresis_less_than" notification in red under the Hysteresis box. I have tried various numbers for hysteresis.

Reef-Pi is measuring temp and plotting the graph now, but I can't add control.

Solution?

Discovered something.... When setting up Temperature control, you need to define Chiller Threshold as well as Chiller in addition to defining your heater, even if you have no chiller attached. If not, you get the "Hysteresis error" described above.

 

[mk1scott]

Hey guys & gals, first time poster, long time reader. First and foremost, a huge thank you to everyone involved in developing this software! This is a game changer for me (just getting back into reefing after a 10 year hiatus).

I've been running Reef Pi flawlessly for a little over 4 months now running 3.0 alpha 2 (Pi zero w/ Pico board). However, just this week, I have seen my system load spike (max out) and crash the web UI twice. Both times requiring a power cycle to get back online. When the system load spikes and crashes the UI, Reef Pi continues to control my system as it should, but stops sending data to the web UI as well as adafruit IO. It first occurred Sunday afternoon (2:56pm typical load of 0.06; 3:13pm load jumped to 1.01) and then again this morning (7:38am load 0.06; 8:09am load 1.00).

I'll try pulling the logs tonight but wasn't sure if anyone else was experiencing any recent unexpected spikes.

 

[Blaxkin]

Now here is my Problem. After Restart i get some Errors from Temp Probes. But i only have to delete and recreate the Probes because they are always aviable to choose.
I hope you know what i mean.

 

[Ranjib]

Now here is my Problem. After Restart i get some Errors from Temp Probes. But i only have to delete and recreate the Probes because they are always aviable to choose.
I hope you know what i mean.

Thank you for reporting this. I think I understand the problem a bit better now. Can you confirm following:
- Does this only happen if you reboot? What if you just reload reef-pi ?
- When you next encounter this, instead of deleting /adding them, can you just wait few seconds and reload reef-pi, and see if that resolves this.
- FYI: those "error: stats not found" issues are benign and just side effect sensor not being detected or enabled. they are used for charting only. I'll see if we can suppress those

 

[Blaxkin]

Thank you for reporting this. I think I understand the problem a bit better now. Can you confirm following:
- Does this only happen if you reboot? What if you just reload reef-pi ?
- When you next encounter this, instead of deleting /adding them, can you just wait few seconds and reload reef-pi, and see if that resolves this.
- FYI: those "error: stats not found" issues are benign and just side effect sensor not being detected or enabled. they are used for charting only. I'll see if we can suppress those

-Yes it only happens after a reboot

-I tried a reload and the problem is the same

-but now i see when i only click edit on the sensor and save. they came up after about 5 minutes

-and now i see the problem is not only on the temp probes. its the same problem with the pca9685

-the only i2c device that's always without an error is the ph ego board

 

[Ranjib]

-Yes it only happens after a reboot

-I tried a reload and the problem is the same

-but now i see when i only click edit on the sensor and save. they came up after about 5 minutes

-and now i see the problem is not only on the temp probes. its the same problem with the pca9685

-the only i2c device that's always without an error is the ph ego board

I am postulating a theory that sensors and other i2c devices are being initilized after reef-pi starts, as a result when reef-pi starts it does not find them.. I dont know how to verify this, as im never encountered it, but i am aware of some trick on systemd level we can do to make sure reef-pi starts after those things are up and running, i 'll cross check how to get it done and what implications of this.

 

[Ranjib]

I am postulating a theory that sensors and other i2c devices are being initilized after reef-pi starts, as a result when reef-pi starts it does not find them.. I dont know how to verify this, as im never encountered it, but i am aware of some trick on systemd level we can do to make sure reef-pi starts after those things are up and running, i 'll cross check how to get it done and what implications of this.

BTW, this is pi 4 or 3 or zero?

 

[Blaxkin]

BTW, this is pi 4 or 3 or zero?

Mine is a raspberry 3b