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

Have you seen the A+? I’m sure you have. I was debating on ordering one to play with.
I have not used one. But I think it better suited for my use case, A+. I dont need LAN connection or so many USB port the Pi3 B+ gives, and I could use some extra space. May be next build. I have lost only one Pi in past two years. So, I really dont need any more .. which is another problem. I have to give away at least 4 Pi 2 or older model B :-/
 
What did you use to diagram all of this? It looks incredibly pristine. Very neat and tidy. And don’t tell me it’s freehand...

It has the color scheme of Eagle CAD.

(You can usually tell the various EDA tools apart based upon their default colors in schematics [emoji854])
 
I have not used one. But I think it better suited for my use case, A+. I dont need LAN connection or so many USB port the Pi3 B+ gives, and I could use some extra space. May be next build. I have lost only one Pi in past two years. So, I really dont need any more .. which is another problem. I have to give away at least 4 Pi 2 or older model B :-/

Waiting for the A+ to be in stock somewhere, but it’s a nice form factor as long as it’s not restricted to the one per person like the Pi Zero.
 
So, its working as expected now? I'd love to know your experience with reef-pi. If I recall correctly, you have been running it for few months now

Well, it was working & I managed to break it again, lol. I'm attempting to run a pi hole ad-blocker on the same RBPi, but it also uses port 80. I attempted changing the end of the address field in the setting for reef pi from 80 to 8093, but it seems to have broken it. It still shows it as running, but I can't access any of the UI. It simply shows a blank dashboard when I go to http://localhost.

I tried stopping & starting reef pi, as well as resetting its database, but no luck. Looks like I may have to do a fresh install, which I assume would reset that address to default.

Assuming it does reset everything I broke, how would you recommend changing the port so I don't kill everything again?

~`~

As far as my experience with reef pi, it's definitely been frustrating at times like this. But that's more due to me poking around & changing things. Overall though, it's been excellent. It's let me take multi-day trips without worrying at all about my tank, which is awesome. I managed to get a bit of saltwater in the RBPi housing & fried one, but once I replaced it, it's been smooth sailing since then. I'm definitely not using it to its fullest potential, just using the equipment control & timers at the moment. But it's been very reliable & an awesome project, especially due to all the excellent support & updates from you, @Ranjib !
 
@Ranjib minor issue with the dosing graph and possibly the other new ones as well. After rebooting the RPI I get 404 errors if the dashboard is configured to show the dosers. Because they have not put out an input yet the graph does not show up and the error pops up. The work around to get the errors to go away is to run the dosers quick via the calibrate button then the graph shows up on the dashboard and the error goes away. Otherwise the error persists and the graph is absent until the next scheduled run.
46023785032_32e82d440b_b.jpg
 
I think we all need to work on some sort of logo or design. My wife is an awesome artist. She’s designed a company logo for me? That is, once I have a company[emoji1]. I just have a FB group page for now.
But if you all would like to spitball some ideas on a logo for Ranjibs project, I’d be happy to have her try out some ideas from your ideas. Let me know.

I just put this little guy together in Paint as a logo proposal. Nothing fancy, but I really like it :D

Fish.png
 
@Ranjib minor issue with the dosing graph and possibly the other new ones as well. After rebooting the RPI I get 404 errors if the dashboard is configured to show the dosers. Because they have not put out an input yet the graph does not show up and the error pops up. The work around to get the errors to go away is to run the dosers quick via the calibrate button then the graph shows up on the dashboard and the error goes away. Otherwise the error persists and the graph is absent until the next scheduled run.
46023785032_32e82d440b_b.jpg
I fixed this issue for other graphs, doser chart was added at the last moment and I missed it for this. I’ll get it fixed by 2.1
 
Last edited:
What are the limits of the Raspberry pi B+? I know earlier @Ranjib showed usage for the Pi Zero and had stated that 60% CPU and 1% Memory was probably the limit. I am running the B+... VNC server is on, OS GUI is installed, I am powering my relays now, a pair of low current Computer fans from the PWM pins once per hour, temp checking every 10 seconds, and ATO every 10 seconds, two wavemakers on the power strip that alternate every hour, and my sump light schedule. I am still around 1% memory (showing between 10-14 on the graph) and CPU is really low around between .01-.04 with the occasional spike to 40%
 
What are the limits of the Raspberry pi B+? I know earlier @Ranjib showed usage for the Pi Zero and had stated that 60% CPU and 1% Memory was probably the limit. I am running the B+... VNC server is on, OS GUI is installed, I am powering my relays now, a pair of low current Computer fans from the PWM pins once per hour, temp checking every 10 seconds, and ATO every 10 seconds, two wavemakers on the power strip that alternate every hour, and my sump light schedule. I am still around 1% memory (showing between 10-14 on the graph) and CPU is really low around between .01-.04 with the occasional spike to 40%
Should you be running fans from the pwm pins, didn't think it would be capable tbh
 
Should you be running fans from the pwm pins, didn't think it would be capable tbh

@LionHeart2017 I have the doser circuit built using the TB6612 motor driver. I dont have the doser motors ready for reef-pi yet so I am simulating a doser motor with the fans. They are not running directly off the GPIO pins.
 
@LionHeart2017 I have the doser circuit built using the TB6612 motor driver. I dont have the doser motors ready for reef-pi yet so I am simulating a doser motor with the fans. They are not running directly off the GPIO pins.
Yeah I didn't think from the pins would work[emoji4]
 
I read this in another thread, but people here might want to consider it as I will be doing this for my ATO. I currently have a TUNZE ATO, my concern is with dosing Kalk,, if it fails my PH would rise enough to kill off everything even before the timeout feature kicked in. I was going to wait and have the PH probe shut off the power to the ATO if above X, but another solution would be to just plug my ATO into a reef-pi controlled outlet that only turns on every 30 minutes for 15 seconds. In the winter I have found my ATO runs about every 40 minutes for about 9 seconds. So this logic would work to keep me safe and note any slight high water conditions if they existed before a real problem as the extra time would raise the level. It is easy to see the level and fluctuation in my AIO tank. Hope this made sense and might help other with some extra peace of mind.
 
My gears are already turning for future add-on's or different build ideas. The thought of automatic water changes popped into my head this morning. So tell me if I am totally off base here or if my plan is flawed. I would like to change 10 gallons of water every 30 days. So that's approximately 50ml per hour of water to be removed and added back per day. In theory this could be done with just the regular doser circuit. one pump to remove 50ml and one pump to add 50ml from a source per hour. Float switches could easily be added in the ATO portion of reef-pi as fail safes for if either one of the fresh saltwater source or tank get out of whack for some reason. For instance if the water level from the source drops below set switch level turn off the drain pump and if tank water goes above set switch level turn off source pump. The question I have is this should work with regular peristaltic pump but would it be robust enough. I dont know enough about stepper motors but would that be better suited for something like this? I know that there are several H-Bridge drivers out there that could drive a stepper motor.
 
My gears are already turning for future add-on's or different build ideas. The thought of automatic water changes popped into my head this morning. So tell me if I am totally off base here or if my plan is flawed. I would like to change 10 gallons of water every 30 days. So that's approximately 50ml per hour of water to be removed and added back per day. In theory this could be done with just the regular doser circuit. one pump to remove 50ml and one pump to add 50ml from a source per hour. Float switches could easily be added in the ATO portion of reef-pi as fail safes for if either one of the fresh saltwater source or tank get out of whack for some reason. For instance if the water level from the source drops below set switch level turn off the drain pump and if tank water goes above set switch level turn off source pump. The question I have is this should work with regular peristaltic pump but would it be robust enough. I dont know enough about stepper motors but would that be better suited for something like this? I know that there are several H-Bridge drivers out there that could drive a stepper motor.
A failsafe that would not rely on consistent calibration of pumps would be to have an intermediate container that is the size of your water change, with a float switch/optical sensor.
Step 1) run pump A to fill container from tank water until float switch activates.
Step 2) run pump B for extended time to empty container to a drain.
Step 3) run pump C to fill container from reservoir until switch activates
Step 4) run pump A in reverse for extended time to return new water to tank.
 
I attempting reinstalling reef-pi, but am now getting this error in the journal:

>Nov 27 17:47:32 aquarium.local reef-pi[23709]: 2018/11/27 17:47:32 ERROR: Failed to run http server. Error: listen tcp 0.0.0.0:80: bind: address already in use

That port is being used by lighttpd at the moment & I've had no luck getting it to change. Whenever I try changing its config file, it won't let me save it. Any suggestions where to go from here?
 
A failsafe that would not rely on consistent calibration of pumps would be to have an intermediate container that is the size of your water change, with a float switch/optical sensor.
Step 1) run pump A to fill container from tank water until float switch activates.
Step 2) run pump B for extended time to empty container to a drain.
Step 3) run pump C to fill container from reservoir until switch activates
Step 4) run pump A in reverse for extended time to return new water to tank.

I thought this same thing as I almost bought the commercial system to do this a while back, you explained it much better than what was running through my head.
 
I attempting reinstalling reef-pi, but am now getting this error in the journal:

>Nov 27 17:47:32 aquarium.local reef-pi[23709]: 2018/11/27 17:47:32 ERROR: Failed to run http server. Error: listen tcp 0.0.0.0:80: bind: address already in use

That port is being used by lighttpd at the moment & I've had no luck getting it to change. Whenever I try changing its config file, it won't let me save it. Any suggestions where to go from here?
Something is already running on that rasoberry pi and using the port 80. You have to stop that and then start reef-pi. Disable the other things you have installed on the same pi
 

TOP 10 Trending Threads

ARE YOU READY TO CONFESS TO CRAZIEST, DUMBEST, FUNNIEST THING YOU’VE EVER DONE IN REEFING?

  • Yeah, I'll confess! (Share your story in the comments!)

    Votes: 30 55.6%
  • Nah, I'll keep mine a secret...(Don't be like that, share with the class!)

    Votes: 24 44.4%
Back
Top
Home
Post thread…
Market
What's new