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

Does the cpu and memory graph looks normal?
Not really, CPU went randomly up, but still only to 8%, not sure exactly what the memory number is supposed to tell me tbh but it didn't change much

Screenshot_20211204-113643.png

The left third of the current graph is old data from october, there's a cut sometime yesterday, some data went missing on reboot

Actually, now that I am looking, what also went missing was a new doser entity I created and doser settings I changed that were showing up before the reboot. Something really weird is going on, not sure if reefpi is holding data in ram but not writing to card properly or something entirely different, idk
 
What do I put in upgrade text box EXACTLY to upgrade inside reef pi to version 5.0?

Thanks
 
Wave maker electronics are not yet sorted. Because most of them ships with their own controller and very few allows external control, we have been mostly delayes the support for the wavemaker. With 3.0 we shipped pwm profiles that will make things easy, and probably from software side we have the bare minimal things, but the electronics needs to be sorted out. My gut feeling is we can take the pwm output from pca9685 or pi and then use a BLDC/BEMF controller for 24V (drok has one) and then test some pumps (mp10s, gyres etc).
I'm VERY interested in this. I purchased some RC ESCs and found them unsuitable - they require complicated arming procedures that in that application are fail-safe, but in this application are fail-deadly. I'm not worried about crashing my pump into a tree, I want it to turn on even if it's the wrong speed, y'know?

I've got a JYQD_V6.31 board arriving Tuesday, hoping that exactly that approach works out for me. I've read other threads that seem to imply that there's some loss in power, but I'm not super worried about that, I already run pretty low and plan on adding more wavemakers anyway.

I want to basically emulate the features of the Hydros 'WaveEngine', not the smallest of which is getting rid of the power bricks and controller boxes. Jebao apparently told CoralVue they couldn't add compatibility for their pumps to the WaveEngine, which... I really don't care if I void the warranty on an 80 dollar pump, I can buy another one - why is not the consumer's choice?
 
I'm VERY interested in this. I purchased some RC ESCs and found them unsuitable - they require complicated arming procedures that in that application are fail-safe, but in this application are fail-deadly. I'm not worried about crashing my pump into a tree, I want it to turn on even if it's the wrong speed, y'know?

I've got a JYQD_V6.31 board arriving Tuesday, hoping that exactly that approach works out for me. I've read other threads that seem to imply that there's some loss in power, but I'm not super worried about that, I already run pretty low and plan on adding more wavemakers anyway.

I want to basically emulate the features of the Hydros 'WaveEngine', not the smallest of which is getting rid of the power bricks and controller boxes. Jebao apparently told CoralVue they couldn't add compatibility for their pumps to the WaveEngine, which... I really don't care if I void the warranty on an 80 dollar pump, I can buy another one - why is not the consumer's choice?
Keep us posted on what you find. I think @robsworld78 has tinkered with the jebao pumps, .. there were some posts on robotank forum related to this.
Long ago , I think @theatrus also shared some insights (my BEMF/BLDC suggestions were based on those).
 
Not really, CPU went randomly up, but still only to 8%, not sure exactly what the memory number is supposed to tell me tbh but it didn't change much

Screenshot_20211204-113643.png

The left third of the current graph is old data from october, there's a cut sometime yesterday, some data went missing on reboot

Actually, now that I am looking, what also went missing was a new doser entity I created and doser settings I changed that were showing up before the reboot. Something really weird is going on, not sure if reefpi is holding data in ram but not writing to card properly or something entirely different, idk
only usage data is kept in memory (and written back to disk after every hour, or during planned reload/restart). Rest every thing , all CRUD operations (create, update, retrieve, delete ) of domain objects (ato, connectors, light, ph, macro .. pretty much all things in reef-pi) is always done with database (sd card/disk).
 
I think @robsworld78 has tinkered with the jebao pumps, .. there were some posts on robotank forum related to this.
Just bought a board from Rob last night. I was just going to buy a pH board to wire into my DIY controller, but they're clearing out the v1 boards which include a pH sensor as well... I can use the prototype on my garden or something lol.

@theatrus runs BlueAcro, yeah? I saw that they had been working on a prototype driver, but the threads all seem to just end without a satisfying conclusion.
 
Just bought a board from Rob last night. I was just going to buy a pH board to wire into my DIY controller, but they're clearing out the v1 boards which include a pH sensor as well... I can use the prototype on my garden or something lol.

@theatrus runs BlueAcro, yeah? I saw that they had been working on a prototype driver, but the threads all seem to just end without a satisfying conclusion.
yes. i too started playing with it, but then felt it was not needed since i had to buy vortechs and they will ship with their own controller. It does not look like anyone except kessil (that too for their a80/160/360/150) has some known external control mechanism. Meanwhile most use cases that i had in mind involving controlling the pumps, is doable just by controlling the power source (like water change)..
 
yes. i too started playing with it, but then felt it was not needed since i had to buy vortechs and they will ship with their own controller. It does not look like anyone except kessil (that too for their a80/160/360/150) has some known external control mechanism. Meanwhile most use cases that i had in mind involving controlling the pumps, is doable just by controlling the power source (like water change)..
ReefOctopus pumps take in 0-10v, as do a couple models of the jebao controllers. I'm looking to bypass those entirely though and directly drive the pumps.

My ReefBreeders Photon lamp also does 0-10v control, although I had to buy a board for it.
 
Keep us posted on what you find. I think @robsworld78 has tinkered with the jebao pumps, .. there were some posts on robotank forum related to this.
Long ago , I think @theatrus also shared some insights (my BEMF/BLDC suggestions were based on those).

ReefOctopus pumps take in 0-10v, as do a couple models of the jebao controllers. I'm looking to bypass those entirely though and directly drive the pumps.

My ReefBreeders Photon lamp also does 0-10v control, although I had to buy a board for it.
I don't know much about them but it appears the wave makers are controlled using a 0-5v signal to control the speed of the pump. On the WP's I have I checked the voltage coming out of the pig tail from the Jebao controller, one pin was ground, another was 24v and the 3rd was 0-5v depending the speed of the controller. So I hooked up 24v to the respected pins and the signal I connected to a 0-5v PWM pin and it worked great. I removed the pig tail from the Jebao controller so I didn't have to cut the pump cable and marked where the wires went in the controller in case I ever want to put it back. I've mentioned this to others with Jebao wave makers and it was the same so my guess is they all work like that.

Forgot to mention, someone told me the other day @Ranjib added a couple extra light modes, random and sine, he thought they were going to be good for controller wave makers. Was this the intent and anyone else planning on trying them?
 
Last edited:
I don't know much about them but it appears the wave makers are controlled using a 0-5v signal to control the speed of the pump. On the WP's I have I checked the voltage coming out of the pig tail from the Jebao controller, one pin was ground, another was 24v and the 3rd was 0-5v depending the speed of the controller. So I hooked up 24v to the respected pins and the signal I connected to a 0-5v PWM pin and it worked great. I removed the pig tail from the Jebao controller so I didn't have to cut the pump cable and marked where the wires went in the controller in case I ever want to put it back. I've mentioned this to others with Jebao wave makers and it was the same so my guess is they all work like that.

Forgot to mention, someone told me the other day @Ranjib added a couple extra light modes, random and sine, he thought they were going to be good for controller wave makers. Was this the intent and anyone else planning on trying them?
The older jebao pumps have an internal controller that works that way but they kept overheating and dying. The new ones appear to be sensorless bldcs, at least by the board markings.
 
The older jebao pumps have an internal controller that works that way but they kept overheating and dying. The new ones appear to be sensorless bldcs, at least by the board markings.
The ones I have are old. It's probably better using BLDC controller, let us know how it works out. A lot of this is over my head, I'm not the expert some might think.
 
The ones I have are old. It's probably better using BLDC controller, let us know how it works out. A lot of this is over my head, I'm not the expert some might think.
Neither am I, heh - I just know enough to be dangerous. The only reason I know anything at all about motor controllers is because I built an ebike last year.
 
Going to start setting up my Raspberry Pi for Reef-Pi. Is there any specific SD card that is more reliable? Seems like the SD card is the biggest issue with Raspberry Pi reliability. I thought I had a spare but will need to buy a new card.
 
Had the same issue again. After a powercycle, all data after the same day in october as last time is gone again, a new doser i set up is gone, and what i can see now that i am looking on desktop and not mobile, reef-pi has reverted from 5.0 to 4.3

How can this be, could i have multiple copies of reef-pi running by accident? I remember that i tried to use the upgrade fucntion in the admin panel and it didnt work and locked up reefpi so it powercycled it then, could that have caused it? Or is a memory problem like i initially though more likely?
 
I remember that i tried to use the upgrade fucntion in the admin panel and it didnt work and locked up reefpi so it powercycled it then, could that have caused it?
I can't speak to reef-pi in particular, but as someone who works with these sort of systems for a living - in general, if you power cycle in the middle of an upgrade operation, you're gonna have a bad time.
 

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: 70 55.6%
  • Nah, I'll keep mine a secret...(Don't be like that, share with the class!)

    Votes: 56 44.4%
Back
Top
Home
Post thread…
Market
What's new