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

[Ranjib]

Figured it out. I had an extra reef-pi.db in var/lib/reef-pi. I deleted it and then when I started the page again I entered reef-pi reef-pi for the id and password I now have a working 0.8.2

change your interface to "wlan0" if you are on wifi. .. then reef-pi will show IP correctly.

I am thinking if I should make that default

 

[Ryan115]

Ranjib,
What are your thoughts (likely in post 1.0 or 2.0) to have a guided setup process on first run or after a "factory reset"?
That would configure outlets, lights, username & password, timers, network connection, etc...

 

[Cary]

change your interface to "wlan0" if you are on wifi. .. then reef-pi will show IP correctly.

I am thinking if I should make that default

Thanks Hoss/Ranjib...;-)

That is probably the issue with my phone not connecting.

 

[Cary]

change your interface to "wlan0" if you are on wifi. .. then reef-pi will show IP correctly.

I am thinking if I should make that default

At the least we need to add this and the initial Auth I'd and password to the doc. I will give that try later this morning.

 

[Gareth elliott]

Is there a wire diagram for the ato?

 

[Somnifac]

@Ranjib

I'm seeing some odd things with the lighting controls in 8.2. If I create a jack, then add a lighting control on that jack things seem to work fine initially.

Eventually, I can slide things around, clicking update, and it doesn't set it to the set value, even though the Update button went green. Sometimes it seems that it is a setting behind. Example, eventually I will set the slider to 85 and click update, but the value won't update according to the status log. I can then change it to 50 and click update and the value will update to 84. Change the slider again to 25 and it will update to 50.

I haven't found a way to really get it back on track other than a reboot, which sets it back to zero, but that doesn't always get it to respond properly.

Edit:

While typing this, I found that clicking the slider and using the arrow keys to increase or decrease the value before clicking Update makes it respond properly every time. It seems that using the mouse to slide it up and down isn't being read properly?

 

[Ranjib]

Is there a wire diagram for the ato?

Dfrobot’s photo electric sensor, the one I am using has wiki on the wiring
https://www.dfrobot.com/product-1470.html
And https://www.dfrobot.com/wiki/index.php/Liquid_Level_Sensor-FS-IR02_SKU:_SEN0205

Let me know if this does not help ,
I’ll add these links and raspberry pi based wiring to reef pi docs soon

 

[Ranjib]

@Ranjib

I'm seeing some odd things with the lighting controls in 8.2. If I create a jack, then add a lighting control on that jack things seem to work fine initially.

Eventually, I can slide things around, clicking update, and it doesn't set it to the set value, even though the Update button went green. Sometimes it seems that it is a setting behind. Example, eventually I will set the slider to 85 and click update, but the value won't update according to the status log. I can then change it to 50 and click update and the value will update to 84. Change the slider again to 25 and it will update to 50.

I haven't found a way to really get it back on track other than a reboot, which sets it back to zero, but that doesn't always get it to respond properly.

Edit:

While typing this, I found that clicking the slider and using the arrow keys to increase or decrease the value before clicking Update makes it respond properly every time. It seems that using the mouse to slide it up and down isn't being read properly?

This is bug. I know someone else had mentioned earlier, but I didn’t get a chance to verify . I was able to detect it yesterday and fix it https://github.com/reef-pi/reef-pi/commit/1fea35fac1f64e932399d3713ac31a9f34120a50
I am also working on getting a full blown end to end automated test setup for the critical workflows, to ensure we don’t get regression like this

 

[Somnifac]

Ah, ok, great. Glad I'm not going crazy.

Another small question:

I wired up the 5v to 10v circuit on one of the pins, but it's not behaving as I would expect. When power is set to "0" I read ~12v. When power is set to "100" I read ~0v. With power set to "50" I read ~6v.

I expected "0" to be roughly 0v, 50 to be roughly 5v, and 100 to be roughly 10v.

Anyone have any thoughts?

 

[dburg30]

This is awesome! I've been following it for a while, just getting started in reefing. Actually tanks are still filling as we speak lol.. Fired up one of my pi's and got reef-pi installed. Went and picked up some stuff to start working on lights. Also picked up some of the other sensors. Only thing I didnt get was a temp probe. Microcenter showed having ONE in stock, but we couldnt find it. Super excited about this project!!!!!

 

[Ranjib]

Ah, ok, great. Glad I'm not going crazy.

Another small question:

I wired up the 5v to 10v circuit on one of the pins, but it's not behaving as I would expect. When power is set to "0" I read ~12v. When power is set to "100" I read ~0v. With power set to "50" I read ~6v.

I expected "0" to be roughly 0v, 50 to be roughly 5v, and 100 to be roughly 10v.

Anyone have any thoughts?

that is expected, this is result of how npn transistors work. Click the “reverse” checkbox next to you channel , then reef-pi will do the reverse logic for you

 

[Ranjib]

This is awesome! I've been following it for a while, just getting started in reefing. Actually tanks are still filling as we speak lol.. Fired up one of my pi's and got reef-pi installed. Went and picked up some stuff to start working on lights. Also picked up some of the other sensors. Only thing I didnt get was a temp probe. Microcenter showed having ONE in stock, but we couldnt find it. Super excited about this project!!!!!

I’m glad you liked reef-pi, let us know if you need any help

 

[philshel]

This is bug. I know someone else had mentioned earlier, but I didn’t get a chance to verify . I was able to detect it yesterday and fix it https://github.com/reef-pi/reef-pi/commit/1fea35fac1f64e932399d3713ac31a9f34120a50
I am also working on getting a full blown end to end automated test setup for the critical workflows, to ensure we don’t get regression like this

That was me, I was just getting ready to reply. If you mouse click the slider track to make large changes it's a no-go. You can click and grab the slider button and that will work, or use the arrow keys.

 

[Gareth elliott]

Dfrobot’s photo electric sensor, the one I am using has wiki on the wiring
https://www.dfrobot.com/product-1470.html
And https://www.dfrobot.com/wiki/index.php/Liquid_Level_Sensor-FS-IR02_SKU:_SEN0205

Let me know if this does not help ,
I’ll add these links and raspberry pi based wiring to reef pi docs soon

Thank you,
Found the shipping a little cheaper on robotshop.com. At least on low weight items like this one. Came to a little over $10 shipped, in case anyone else is bargain shopping and getting components still.

 

[Somnifac]

that is expected, this is result of how npn transistors work. Click the “reverse” checkbox next to you channel , then reef-pi will do the reverse logic for you

Ah, ok. It was the 12v peak that was throwing me. It seems that work a peak of 12v, there would be less usable gradation, since in the conversion it would reach max brightness (10v) earlier than 100% power on the slider.

Is there some form of voltage regulator we could put inline before we bring the 12v into the circuit to step it down to 10v?

 

[Ryan115]

So, I opened the sBox 32 last night to survey before finalizing the design and plans for the dimmer circuit.
The drivers are using this dimmer circuit https://www.dhgate.com/product/getP...420958&pid=ff8080815003ad2b015012de52f762e2#1 which is being fed by a separate 12v supply, that is also running the fans.
It looks like it will be very simple to replace this with our custom circuit without having to cut any cables.
The drivers connect to the dimmer circuit with a 3-pin JST connector, so a matching header on our circuit is plug and play.
https://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=243697428&uq=636460675736317245
All of the internal wiring has disconnects using 2 pin JST connectors.
https://www.adafruit.com/product/2880
So the matching cable set will also plug right in.

Ah, ok. It was the 12v peak that was throwing me. It seems that work a peak of 12v, there would be less usable gradation, since in the conversion it would reach max brightness (10v) earlier than 100% power on the slider.

Is there some form of voltage regulator we could put inline before we bring the 12v into the circuit to step it down to 10v?

I plan on using a regulator, as without spec sheets, I have no idea what giving it 12v would do to the drivers.
I have some larger modules that could work for this (i.e. UBEC), but I think this will work for our needs.
https://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=243698739&uq=636460675736327246

 

[Somnifac]

I plan on using a regulator, as without spec sheets, I have no idea what giving it 12v would do to the drivers.
I have some larger modules that could work for this (i.e. UBEC), but I think this will work for our needs.
https://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=243698739&uq=636460675736327246

The link leads to search results, rather than a specific item. Do you have a part number handy? I'll definitely be investigating adding something like this. It's either that or the 8x channel converter linked above.

For $25, the 8x channel converter might be the simpler, all in one solution though... Would lead to an easier time if I needed to reconfigure lights in the future.

 

[Ranjib]

Ah, ok. It was the 12v peak that was throwing me. It seems that work a peak of 12v, there would be less usable gradation, since in the conversion it would reach max brightness (10v) earlier than 100% power on the slider.

Is there some form of voltage regulator we could put inline before we bring the 12v into the circuit to step it down to 10v?

I won’t be too worried about it. The kessil controller itself does not generate precise 10v peak (varies between 9-11 v). But if you want electronics safeguard (which is a good idea) , you can go with any 10v linear regulator like lm7810.
Last but not the least reef-pi has software configuration to enforce a upper limit on the pwm value, these are called min and max threshold , when configured reef-pi will ensure the pwm signal does not cross the specified range. I used it to ensure I don’t generate more than 10v or less than 3v (both me and @Aaron Smith experienced flickering of led lights at lower pwm values, which is why I had introduced this).
Unfortunately reef-pi ui still does not have support for this (min max threshold per channel), so you have to do it via command line , let me know if you take this route, I can share the specific curl command

 

[Ranjib]

Thank you,
Found the shipping a little cheaper on robotshop.com. At least on low weight items like this one. Came to a little over $10 shipped, in case anyone else is bargain shopping and getting components still.

Nice
I specifically mention adafruit whenever possible as it’s US made product with unmatched quality control, and they are very community focused, opensource /openhardware company. Most f my learning and reef-pi s underneath electronics is a direct derivative of their work/product/wisdom . So I use every opportunity to support them ;-)
Users should use whatever vendor they prefer though.

 

[Ryan115]

The link leads to search results, rather than a specific item. Do you have a part number handy? I'll definitely be investigating adding something like this. It's either that or the 8x channel converter linked above.

For $25, the 8x channel converter might be the simpler, all in one solution though... Would lead to an easier time if I needed to reconfigure lights in the future.

Sorry, the link for the regulator was a 7810, as ranjib suggested. The one I was looking at was 296-1996-ND, slightly different but should be the same result.