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

in this route, where you are primarily using the arduino for analog input, how you do galvanic isolation? reef-pi currently has a HAL Analog driver interface that we can use to implement an arduino driver. We can totally do it if it makes sense (i.e. the results we get are good enough for our use case, and theres enough appetite for such use case)

Ability to interface to Arduino from reef-pi over I2C would be tremendous. Especially for I/O control to start. Not to mention access to analog inputs that can be used for not only analog PH but many other applications like tuning optical sensor levels.

Another good option would be to do it over serial so this will provide ability to have Arduino at remote location if needed.

The simplest way of galvanic isolation is to have two independent power sources. so in this case if Arduino is fed by another power supply ( will need to be with transformer) then if another PH probe is fed from Reef-PI they will not interfere with each other.
 
Last edited:
Just a comment, nice power distribution but little scary if you need to load all on a single feed of power. Typical house wiring 14/2 with maximum 15A panel breaker. Since it is very likely that this is not going to be plugged into a dedicated circuit and will be shared with some of the existing receptacles in the house very easy for cabling to overheat and spark a house fire. Just an FYI


A buddy of mine is an electrician and he estimated that I'll need at least 4 separate circuits for what I plan to do. He said he'll run 6 just to be on the safe side. It's great having friends :D
 
I keep missing the invite link lol Mind linking it again and I’ll get into the discussion thread as well as follow up with GitHub and direct me to the priority coding / documentation , etc.
Start with the developer environment setup (docs in website, link is in slack header), check out 4.0 project in github under reef-pi repo for a tentative list of things we plan to work on 4.0 (more to be added). Let us know your strength , time availability and the things you want to learn and we can definitely find a suitable work item for you.
Godspeed
 
Ability to interface to Arduino from reef-pi over I2C would be tremendous. Especially for I/O control to start. Not to mention access to analog inputs that can be used for not only analog PH but many other applications like tuning optical sensor levels.

Another good option would be to do it over serial so this will provide ability to have Arduino at remote location if needed.

The simplest way of galvanic isolation is to have two independent power sources. so in this case if Arduino is fed by another power supply ( will need to be with transformer) then if another PH probe is fed from Reef-PI they will not interfere with each other.
In 3.0 we introduced HAL (hardware abstraction layer), using that it should be relatively easy to have a arduino driver that can do pwm, digital i/o, analog input etc.. if you are mimicing the existing protocols, then you should be able to just add it as an existing driver :-)
 
A buddy of mine is an electrician and he estimated that I'll need at least 4 separate circuits for what I plan to do. He said he'll run 6 just to be on the safe side. It's great having friends :D
+1000 ,,
 
I think what he's asking is where there's a config file on the filesystem that he can change that value in, because he can't get to that screen in the UI right now.

He could stop the other service on 80 for a few, as well, but maybe that's not possible or a good thing for some reason.

I can't check for the file right now, I don't have access to my pi from here.
the config and everything in reef-pi is stored in database, as binary file (/var/lib/reef-pi/reef-pi.db), in 3.0 we have introduced a command (reef-pi db) that allows you to introspect and update/delete stuff from database directly bypassing UI. reef-pi needs to stopped before doing this.
 
Hive mind
its been more than a month since 3.0 release. Thankfully we have not heard of any serious bugs except few edge cases (bugs induced by very specific type of inputs or usecases). I have taken time to go through those and fix them along with addressing several minor bugs (does not impact controller behavior/function critically but undesirable or confusing things). There are few minor but very useful feature that also i felt were missing after using 3.0 for a month. I have made all those changes and we are shipping reef-pi 3.1 today. This is a non invasive drop in upgrade. All 3.0 users should be able to upgrade without any modifications. Following are the key highlights from 3.1 release

Features
- Revert button in macro ui and api
- ATO, Temperature UI provides top-level enable/disable buttons
- Introduce pi version detection capability and expose it as part of summary information
- Macros can be triggered via ato
- Introduce Undervoltage detector (chart/alerting capability yet to be added)


Bugfixes
- Fix grid alignment under dashboard configuration UI
- Fix interval profile out of index error
- Fix inlet UI components alignment
- Hysteresis validation is applied even when the chiller threshold is not defined
- Fix data race issues in unit tests

Chore
- i18n loader improvement
- dependency upgrades (frontend and backend)
- Improve Chinese translation


Release builds can be found in usual location


cheers,
 
Hive mind
its been more than a month since 3.0 release. Thankfully we have not heard of any serious bugs except few edge cases (bugs induced by very specific type of inputs or usecases). I have taken time to go through those and fix them along with addressing several minor bugs (does not impact controller behavior/function critically but undesirable or confusing things). There are few minor but very useful feature that also i felt were missing after using 3.0 for a month. I have made all those changes and we are shipping reef-pi 3.1 today. This is a non invasive drop in upgrade. All 3.0 users should be able to upgrade without any modifications. Following are the key highlights from 3.1 release

Features
- Revert button in macro ui and api
- ATO, Temperature UI provides top-level enable/disable buttons
- Introduce pi version detection capability and expose it as part of summary information
- Macros can be triggered via ato
- Introduce Undervoltage detector (chart/alerting capability yet to be added)


Bugfixes
- Fix grid alignment under dashboard configuration UI
- Fix interval profile out of index error
- Fix inlet UI components alignment
- Hysteresis validation is applied even when the chiller threshold is not defined
- Fix data race issues in unit tests

Chore
- i18n loader improvement
- dependency upgrades (frontend and backend)
- Improve Chinese translation


Release builds can be found in usual location


cheers,

Awesome!
 
In 3.0 we introduced HAL (hardware abstraction layer), using that it should be relatively easy to have a arduino driver that can do pwm, digital i/o, analog input etc.. if you are mimicing the existing protocols, then you should be able to just add it as an existing driver :)

Would you have an example on how to use it? say how would I create digital I/o from Arduino to reef-pi.
 
A buddy of mine is an electrician and he estimated that I'll need at least 4 separate circuits for what I plan to do. He said he'll run 6 just to be on the safe side. It's great having friends :D

Hope you live near me in Chicago! I need an elechicken to check my work, but they charge wayyyyyy too much by the hour around here. Seriously called a guy, and for a run of 12' linear, I was quoted 650
 
Would you have an example on how to use it? say how would I create digital I/o from Arduino to reef-pi.
checkout reef-pi/drivers repository it has all the drivers implementations. The file driver will give some idea on what a minimal implementation looks like. The actual interfaces that needs to be implemented are defined under reef-pi/hal
 
@Ranjib , @crusso1993 .... if one of you could take a peek at my DLI driver and point me to what needs fixing, i'd like to get that into at least my branch and then put in a pull request. The reason I like the DLI switch so much is that it isn't reliant on the cloud for connectivity and functionality.

Separately, I've been pondering how to do an auto shut-off on my skimmer, so that when the cup is nearly full, it turns off. I know that I could use this: https://www.bulkreefsupply.com/smart-level-security-autoaqua.html , but I would have no reef-pi feedback that the skimmer was turned off. I suppose the first discussion would be a reliable way to detect it - optical sensors could easily give false positives if they get "crudded up" by the skimmate. Float switches could get locked up by the stuff too...this particular aspect of the discussion I'll post elsewhere, as it's not directly related to reef-pi, but I'd like to be able to incorporate that functionality into reef-pi eventually. (Same for detecting when my ATO is low, and shut it off).
 
Just a comment, nice power distribution but little scary if you need to load all on a single feed of power. Typical house wiring 14/2 with maximum 15A panel breaker. Since it is very likely that this is not going to be plugged into a dedicated circuit and will be shared with some of the existing receptacles in the house very easy for cabling to overheat and spark a house fire. Just an FYI

Could you expand on that? 14 gauge house wiring is rated to safely carry a 15A load. When you're pulling power through that 15A panel breaker, there shouldn't be a way for it to draw more current than the wire can handle. Once it hits the threshold of that 15A breaker, it will trip, preventing you from overloading what the wire is rated for.

Theoretically, a power strip with 100 outlets would be perfectly safe, as long as you're pulling less than 15A through it. Of course, at those extremes, you'd probably have to start seriously considering the resistance introduced by the outlets themselves.

I'm not sure why you'd consider it scary, unless you've done something to your panel to bypass the circuit breaker. Drawing 10 A through a single power enclosure like this is the same as drawing 1A through 10 different outlets on the same circuit as far as power draw & heat concerns go.
 
A buddy of mine is an electrician and he estimated that I'll need at least 4 separate circuits for what I plan to do. He said he'll run 6 just to be on the safe side. It's great having friends :D

That's wild! What all are you running?? I'm going to guess there are either multiple metal halides or seriously heavy duty pumps/heaters involved. My setup at peak (all pumps, powerheads, lights, heaters, etc. at their absolute highest power draw they're rated for) is less than 7 amps. Not sure what the actual current draw is though. Sounds like you're about to have 90 amps of circuitry installed!!

When I was designing my enclosure, I did some research to find what people with smart plugs typically see as their actual current draw for their tanks. The typical current draw for an average reef tank seems to sit in the 2.5-3A range. I saw only 2 responses using above 5A, both were using multiple metal halide lights.

Absolute highest response I found was a guy running about 25 amps, with 3 250W metal halides, 4 60" T5s, huge 3.5A return pump, 800W 9A heater, & more. This was for a 375 gallon SPS tank.

I guess what I'm getting at, is unless you're running a fish store, or the equivalent, in your house.... 90A is *major* overkill. The 4 circuits at 60A miiiight be necessary if you've got a crazy setup & a lot of other power-hungry electronics on the same circuit as well. I'd be curious to hear what all you've got on that one circuit & why he thinks 60A would be the minimum. I'm not a professional electrician, so I'm not saying he's wrong... but for a figure like that, there'd better be a good reason :oops:
 
Last edited:
Hive mind
its been more than a month since 3.0 release. Thankfully we have not heard of any serious bugs except few edge cases (bugs induced by very specific type of inputs or usecases). I have taken time to go through those and fix them along with addressing several minor bugs (does not impact controller behavior/function critically but undesirable or confusing things). There are few minor but very useful feature that also i felt were missing after using 3.0 for a month. I have made all those changes and we are shipping reef-pi 3.1 today.

Update went perfectly smooth - have seen no issues as of yet on my small (currently desktop monitoring - LOL) instance.

I'll play around and report if I do manage to find anything but so far looks good!
 
@Ranjib , @crusso1993 .... if one of you could take a peek at my DLI driver and point me to what needs fixing, i'd like to get that into at least my branch and then put in a pull request. The reason I like the DLI switch so much is that it isn't reliant on the cloud for connectivity and functionality.

Separately, I've been pondering how to do an auto shut-off on my skimmer, so that when the cup is nearly full, it turns off. I know that I could use this: https://www.bulkreefsupply.com/smart-level-security-autoaqua.html , but I would have no reef-pi feedback that the skimmer was turned off. I suppose the first discussion would be a reliable way to detect it - optical sensors could easily give false positives if they get "crudded up" by the skimmate. Float switches could get locked up by the stuff too...this particular aspect of the discussion I'll post elsewhere, as it's not directly related to reef-pi, but I'd like to be able to incorporate that functionality into reef-pi eventually. (Same for detecting when my ATO is low, and shut it off).

Are you looking for something like the below link?

 
Are you looking for something like the below link?


almost yeah. I guess that isn't reliant on Arduino (just give it +5V VCC from a pin)...not a fan that it needs to be power cycled to re-zero it, though (in looking at the docs on the DFRobot pages). I suppose I could have a quick connect that i unplug each time...hmm...
 

TOP 10 Trending Threads

Back
Top
Home
Post thread…
Market
What's new