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

[Ranjib]

I tried running an extension cord into my kitchen to see if a running off a different circuit would fix the problem. Unfortunately, it did not:(



The thought has crossed my mind. As is, I can't use the pi as a heater controller, which is a big problem. But, I'd really like to make the halides work.



Thanks for sharing this! If I end up finding a solution, I'll be sure to share it here.

Curious, why you are not able to use reef-pi for heater control

 

[Ranjib]

I wonder, for those having issues with the halides and such, if a isolation transformer on the lamps may be what you need? Pretty common piece of equipment in older arcade games and pinball machines, they are used to isolate the control electronics from the CRT tubes of the TV's.

-Hans

Yeah, thats one option. I would also start researching what we can do from reef-pi circuit side to contain this risk

 

[Ranjib]

Happy Holidays everyone!!

I posted this on the slack channel, but there doesn't seem to be much (non-bot) traffic there.

I have a lot of time off and want to get the DLI Web Switch Pro working in ReefPi. The TPLink addition actually makes this a lot easier (DLIs authentication is simpler, and the interface is all RESTful, so it shouldn't be too hard).

But, I can't find any docs anywhere on how to get a dev environment setup that will allow for this. I see the standard dev environment docs, but doing a make/etc doesn't seem to pull in anything new that I do. What additional steps do I need to do?

Thanks :)

I responded in the slack channels. Treat these communication mediums as asynchronous :-) . I'll repeat my suggestion here,
The development guide is linked in the website, as well as in the github repo/readme. Its here: https://reef-pi.github.io/additional-documentation/development/
Take a look at the tplink driver for example, all you need to do is support hal.DigitalOutputDriver interface... https://github.com/reef-pi/drivers/tree/master/tplink

Best course is to write some go, put it in a github repo, and share it with us, so that we can help you by providing specific advice relevant to your code

 

[raketemensch]

I responded in the slack channels. Treat these communication mediums as asynchronous :) . I'll repeat my suggestion here,
The development guide is linked in the website, as well as in the github repo/readme. Its here: https://reef-pi.github.io/additional-documentation/development/
Take a look at the tplink driver for example, all you need to do is support hal.DigitalOutputDriver interface... https://github.com/reef-pi/drivers/tree/master/tplink

Best course is to write some go, put it in a github repo, and share it with us, so that we can help you by providing specific advice relevant to your code

Thanks for the additional links, I’m going to set up an environment tomorrow in hopes of getting more explicit logging.

it appears that the Slack invitation link in github has expired.

 

[Octocat]

Curious, why you are not able to use reef-pi for heater control

I should have been more clear. Reef-pi works fine as a heater controller with the halides off. But, with them on, there is no temperature input for it to use:p

 

[daveCB]

Link to my build thread - I will be posting photos shortly.

DaveBs Reef-Pi Build

 

[Bigtrout]

Yeah, thats one option. I would also start researching what we can do from reef-pi circuit side to contain this risk

One thing that may help is shielded cables on sensitive circuits. Unfortunately we dont know what type of interference is going on with the halides.
It could be emf interference which good grounding and shileding practices would help with. It could also be harmonic inteference which is being put back thru the AC mains which is harder to deal with...but isolation transformers as was mentioned a few posts back is a possibility.
When my Beamswork LED power supplies went south they were throwing both types of interference and because my temp probes use shielded cables it only affected them intermittently. The ph probe was affected big time because it is only reading millivolt level signals. My cable router was affected as well because it was plugged into the same circuit.
New quality power bricks for the beamswork led lights were an instant solution in my case.

Because interference can be so difficult to track down, we need to take a methodical approach to identify what we are dealing with.

So my first questions are:
What type of ballasts are the MH lights using?
Is it electronic or electromagnetic(more likely)?

Are the lights themselves grounded and shielded properly?

From what I read so far the lights cause issues when they turn on. Once the metal halide lamps are warmed up to full brightness does the interference go away?

I know that electronic ballasts made in china tend to bypass FCC testing and cause problems in commercial applications.

This may be a bit technical for most but here is a good article on the subject:

Grow Light RFI Kits - Palomar Engineers®

Grow Light RFI Kit for Digital Ballast To suppress interference to neighbors and nearby electronic devices, we suggest a RFI-BALLAST kit with 1 ring filter on the AC input and another on the ballast lamp output side for each grow light ballast used. (e.g. 20 ballasts = 20 RFI-BALLAST kits...

palomar-engineers.com

 

[Bzar]

How do we, or is there a way to make a macro/timer repeat itself? For example I want to turn an equipment on for 5 seconds, turn off for 120 seconds, on for 5 seconds, off for 120 seconds, and do that for 5 hours. Is there a way to do that with macro's and timers? I see in dosers there's duration and what not, but can't figure how to get it working outside dosing. I can't seem to make it work :/

 

[pickupman66]

HELP. been a while. I just re-installed the latest version and need to setup heater. it now asks for Hystersis but I cannot get anything I input to work. what is the Proper formatting for Hystersis???

 

[Michael Lane]

HELP. been a while. I just re-installed the latest version and need to setup heater. it now asks for Hystersis but I cannot get anything I input to work. what is the Proper formatting for Hystersis???

Hysteresis is how much wiggle room or dead band you want in the control. I use 0.2 since I'm running fahrenheit. Heater is set to 79.1, Chiller is at 80.0, and hysteresis is 0.2.

 

[Des Westcott]

You should be able to

How do we, or is there a way to make a macro/timer repeat itself? For example I want to turn an equipment on for 5 seconds, turn off for 120 seconds, on for 5 seconds, off for 120 seconds, and do that for 5 hours. Is there a way to do that with macro's and timers? I see in dosers there's duration and what not, but can't figure how to get it working outside dosing. I can't seem to make it work :/

You should be able to do that in timers. Check out options in Cron jobs cheat sheets

 

[Bzar]

You should be able to

You should be able to do that in timers. Check out options in Cron jobs cheat sheets

thanks....think I got it now. For anyone interested to use this timer repeat.
(1) make a macro to turn something on for however long you want. I needed 5 seconds to turn a motor on. Macro turns on a relay outlet that the motor is plugged into and waits 5 seconds, then turns off.
(2) Make a timer that triggers the macro instead of equipment. My timer has a range for the hours of 16-22 so it should run from 4:00pm to 10:00pm. Then it has a repeat for the minutes of */10 so it runs the 5 second macro once every 10 minutes.

Seems to work so far :)

This page was very helpful... https://reef-pi.github.io/build-guides/power/

 

[pickupman66]

Hysteresis is how much wiggle room or dead band you want in the control. I use 0.2 since I'm running fahrenheit. Heater is set to 79.1, Chiller is at 80.0, and hysteresis is 0.2.

That is my understanding of it. Back in 2018 i advocated on it in this thread.

thanks, Discovered I had to put in a Chiller temperature. Ugh. I dont own a chiller.

 

[b4tn]

For dosing. I’m using cheap jaebo dosers and may have come across the need to dose less than 1 ml. My pumps are calibrated so that 1 second at 85% duty cycle gets me 1 ml. Is it safe to say that if I run 42% duty cycle I should get close to .5ml?

 

[Urtoo]

Good evening,
I just reloaded version 3.0 tonight.
I have my HS300 working correctly! This is huge and such a great add.
I added my lighting 9685 driver and set to pin 13,14, I am using ML hat for Reef-Pi. Testing the voltage on the audio connector I am using is showing 0-3.3 volts, I thought it is supposed to be between 0-10v.
When trying to setup Telemetry, I am getting
21:03:57 ERROR: Failed to delete. Error: 535 5.7.8 Username and Password not accepted. Learn more at
Dec 22 21:03:57 ReefPi65g reef-pi[613]: 5.7.8 https://support.google.com/mail/?p=BadCredentials l39sm7477429ywk.36 - gsmtp

I have confirmed the password is correctly typed in and can login to my gmail account with the same password.
Anyone seen this?

 

[Urtoo]

Heh, clicked the HTTPS and can no longer access with the browser, when using https:// or :443.

Bugger

 

[Michael Lane]

Good evening,
I just reloaded version 3.0 tonight.
I have my HS300 working correctly! This is huge and such a great add.
I added my lighting 9685 driver and set to pin 13,14, I am using ML hat for Reef-Pi. Testing the voltage on the audio connector I am using is showing 0-3.3 volts, I thought it is supposed to be between 0-10v.
When trying to setup Telemetry, I am getting
21:03:57 ERROR: Failed to delete. Error: 535 5.7.8 Username and Password not accepted. Learn more at
Dec 22 21:03:57 ReefPi65g reef-pi[613]: 5.7.8 https://support.google.com/mail/?p=BadCredentials l39sm7477429ywk.36 - gsmtp

I have confirmed the password is correctly typed in and can login to my gmail account with the same password.
Anyone seen this?

I don't have an answer for telemetry, but the PWM output on that board is expected to be 3.3v. If you are using my Mars Aqua adapter, then that board is responsible for converting the 3.3v PWM to 10v analog.

 

[Urtoo]

I will be using your Adapter for my lights. Plan on getting that up and running tomorrow.

 

[BenB]

Heh, clicked the HTTPS and can no longer access with the browser, when using https:// or :443.

Bugger

Try https://XXX.XXX.XXX.XXX:80

 

[hhaase]

Because interference can be so difficult to track down, we need to take a methodical approach to identify what we are dealing with.

So my first questions are:
What type of ballasts are the MH lights using?
Is it electronic or electromagnetic(more likely)?

Are the lights themselves grounded and shielded properly?

From what I read so far the lights cause issues when they turn on. Once the metal halide lamps are warmed up to full brightness does the interference go away?

I know that electronic ballasts made in china tend to bypass FCC testing and cause problems in commercial applications.

This may be a bit technical for most but here is a good article on the subject:

Grow Light RFI Kits - Palomar Engineers®

Grow Light RFI Kit for Digital Ballast To suppress interference to neighbors and nearby electronic devices, we suggest a RFI-BALLAST kit with 1 ring filter on the AC input and another on the ballast lamp output side for each grow light ballast used. (e.g. 20 ballasts = 20 RFI-BALLAST kits...

palomar-engineers.com

Lighting ballasts are a constant source of RF noise. Unfortunately, the nature of a DIY system like this is going to make things a bit more problematic since every reef-pi and lighting installation is different.

There is also such a wide variety of MH/T5 lighting systems out there. Electronic and magnetic ballasts work at wildly different frequency ranges. Everybody installs them a bit differently, and I doubt that there is much attention given to grounding paths, shielding, or isolation by many reef owners.... mainly because it hasn't been an issue traditionally, and also because our installations are so far outside the industry norm for that type of lighting. Do people usually install MH/T5's with a functioning chassis ground on the ballast? That's the first thing I'd look at to be honest.

So would it be best to take the path of eliminating the source of the noise, which many people aren't familiar with? Or would it be better to try and shield/isolate the Pi from the noise? But even shielding could be tricky until it's identified if the noise is coming in through the lines, or over the air.

-Hans