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

[DirtDiggler2823]

"Tipping" them?

My apologies. I'm an old jarhead, and for us that means adding the connector to the cable. In this case, the "tip" is either a 3.5mm headphone jack, or RJ-45 connector.

 

[dmolavi]

Gotcha, thanks :)

 

[Ranjib]

This weekend I was busy with analog input connector and ph module integration with it.. the basic integration is now done, but there’s more work needed for making it useful with the ph board. This includes converting raw analog voltage values to ph and incorporating the calibration logic. That will be my next focus...
while not coding , I got busy with building (more like rebuilding) a controller for raffle for our local club in upcoming local coral farmers market . The idea is to change one of my old build with simplified circuit and more probes. Aiming for a pi 3 based build, with three temperature sensor, two ato sensors, one db9 connector for power strip, two spst buttons (one for power and the other is unused) and plenty of empty room in the enclosure for future improvements (ph board or display etc). This time I measured electricity consumption very carefully in each step , with individual components. Here is a rundown of the observations for the curious minds:
My measurements are taken using kill a watt, connected to an AC 110v, 60Hz house hold power line. The reef-pi build is powered by a 12 v , 1a dc wall wart (ul listed, and regulated ). I ran various equipment against a 5.5g tank.
- just the controller :0.03 A
- Power head (12w) : 0.02 A
- light (12w): 0.03A
- heater (25w): 0.17A
Tank water was at garage temperature at 64F, and I preset it to 79F. The heater was able to increase 2F per hour in the first 4 hours , followed by 1.5 F per hour and then plagues at 1.3/0.7 F per hour. I was very excited to see the graphs and note down the numbers , I have many ideas after runnng this. I think having an ability to specify the expected usage and monitor the ongoing usage will be tremendously useful. I am also think probably it’s not required to have multiple independent monitoring of current to get the bulk of benefits ...thoughts ?

 

[theatrus]

Thanks, yeah I didn't see any followup from @theatrus on his tests.

The big question is the sealing of the probes. Ranco just used heat shrink on the one they sold to me for the temp controller I use, and it's been just fine, reliability-wise. They specifically call out PVC shrink tube on their site...I guess because it probably doesn't leech like the normal PO shrink tubing, and has thinner walls which makes for better thermal conductivity.

Both of my dipped units that had epoxy fills in the capsule are still working fine. The unit without the epoxy fill and just a dip failed quickly.

 

[theatrus]

This weekend I was busy with analog input connector and ph module integration with it.. the basic integration is now done, but there’s more work needed for making it useful with the ph board. This includes converting raw analog voltage values to ph and incorporating the calibration logic. That will be my next focus...
while not coding , I got busy with building (more like rebuilding) a controller for raffle for our local club in upcoming local coral farmers market . The idea is to change one of my old build with simplified circuit and more probes. Aiming for a pi 3 based build, with three temperature sensor, two ato sensors, one db9 connector for power strip, two spst buttons (one for power and the other is unused) and plenty of empty room in the enclosure for future improvements (ph board or display etc). This time I measured electricity consumption very carefully in each step , with individual components. Here is a rundown of the observations for the curious minds:
My measurements are taken using kill a watt, connected to an AC 110v, 60Hz house hold power line. The reef-pi build is powered by a 12 v , 1a dc wall wart (ul listed, and regulated ). I ran various equipment against a 5.5g tank.
- just the controller :0.03 A
- Power head (12w) : 0.02 A
- light (12w): 0.03A
- heater (25w): 0.17A
Tank water was at garage temperature at 64F, and I preset it to 79F. The heater was able to increase 2F per hour in the first 4 hours , followed by 1.5 F per hour and then plagues at 1.3/0.7 F per hour. I was very excited to see the graphs and note down the numbers , I have many ideas after runnng this. I think having an ability to specify the expected usage and monitor the ongoing usage will be tremendously useful. I am also think probably it’s not required to have multiple independent monitoring of current to get the bulk of benefits ...thoughts ?

Current monitoring at least at the strip level is helpful to know some failure modes, like “hey your lights didn’t turn on” or “your heater is jammed”. For bigger tanks also helps budget breaker loading.

Back when I ran T5s and halides I paid attention to the power factor as well, since some of those were **** on the ballasts, and really loaded the circuit down in current but not in real power. An accurate real power meter is slightly more expensive and requires voltage taps and watching the integration output as well, as well as some real time math from outputs in the cheap units.

 

[daveCB]

I agree. If we could see simply overall draw as a monitoring tool, that would be sufficient.

 

[DirtDiggler2823]

This weekend I was busy with analog input connector and ph module integration with it.. the basic integration is now done, but there’s more work needed for making it useful with the ph board. This includes converting raw analog voltage values to ph and incorporating the calibration logic. That will be my next focus...
while not coding , I got busy with building (more like rebuilding) a controller for raffle for our local club in upcoming local coral farmers market . The idea is to change one of my old build with simplified circuit and more probes. Aiming for a pi 3 based build, with three temperature sensor, two ato sensors, one db9 connector for power strip, two spst buttons (one for power and the other is unused) and plenty of empty room in the enclosure for future improvements (ph board or display etc). This time I measured electricity consumption very carefully in each step , with individual components. Here is a rundown of the observations for the curious minds:
My measurements are taken using kill a watt, connected to an AC 110v, 60Hz house hold power line. The reef-pi build is powered by a 12 v , 1a dc wall wart (ul listed, and regulated ). I ran various equipment against a 5.5g tank.
- just the controller :0.03 A
- Power head (12w) : 0.02 A
- light (12w): 0.03A
- heater (25w): 0.17A
Tank water was at garage temperature at 64F, and I preset it to 79F. The heater was able to increase 2F per hour in the first 4 hours , followed by 1.5 F per hour and then plagues at 1.3/0.7 F per hour. I was very excited to see the graphs and note down the numbers , I have many ideas after runnng this. I think having an ability to specify the expected usage and monitor the ongoing usage will be tremendously useful. I am also think probably it’s not required to have multiple independent monitoring of current to get the bulk of benefits ...thoughts ?

What Kill A Watt unit did you use?

 

[theatrus]

For those looking at your own power strips, it looks like AliExpress has an interesting selection of metal chassis designed for HiFi DIY which look like they have a lot of internal space. I like metal since it nicely contains any burning bits :)

https://s.click.aliexpress.com/e/bW9bv4ta

There are some overbuilt aluminum ones for EU plugs but probably can find something cheaper there as well (using the Universal outlets common in some Chinese products)

 

[DirtDiggler2823]

Plugged my return pump and lights into the reef-pi power strip, so now I have full control over all electrical devices via the reef-pi. @Ranjib , I still can't get any feeds into adafruit, but I'm not sure it's even worth it at this point.

 

[Bigtrout]

I have my whole setup plugged into a wifi outlet that does power monitoring...I can look at the app and see volts and watts real time and it graphs kw usage by week month and year. And its accurate.
You can set a power threshold to turn off the outlet if power goes above a certain number of watts

https://www.amazon.com/Etekcity-Monitoring-Required-Warranty-Lifetime/dp/B06XSTJST6

 

[Ranjib]

Plugged my return pump and lights into the reef-pi power strip, so now I have full control over all electrical devices via the reef-pi. @Ranjib , I still can't get any feeds into adafruit, but I'm not sure it's even worth it at this point.

i'll follow up on your build thread. Its ok if its a bug, its important for us to know that its a bug, and not have too many gremlins

 

[DirtDiggler2823]

I have my whole setup plugged into a wifi outlet that does power monitoring...I can look at the app and see volts and watts real time and it graphs kw usage by week month and year. And its accurate.
You can set a power threshold to turn off the outlet if power goes above a certain number of watts

https://www.amazon.com/Etekcity-Monitoring-Required-Warranty-Lifetime/dp/B06XSTJST6

Nice. Gotta look into getting a pair of these.

 

[Ranjib]

I have my whole setup plugged into a wifi outlet that does power monitoring...I can look at the app and see volts and watts real time and it graphs kw usage by week month and year. And its accurate.
You can set a power threshold to turn off the outlet if power goes above a certain number of watts

https://www.amazon.com/Etekcity-Monitoring-Required-Warranty-Lifetime/dp/B06XSTJST6

my theory is that we should be able to use this data to determine if an equipment is malfunctioning, by comparing it with past data

 

[DirtDiggler2823]

i'll follow up on your build thread. Its ok if its a bug, its important for us to know that its a bug, and not have too many gremlins

Setting up the macro was pretty simple. I appreciate that. Thanks.

 

[Ranjib]

Current monitoring at least at the strip level is helpful to know some failure modes, like “hey your lights didn’t turn on” or “your heater is jammed”. For bigger tanks also helps budget breaker loading.

Back when I ran T5s and halides I paid attention to the power factor as well, since some of those were **** on the ballasts, and really loaded the circuit down in current but not in real power. An accurate real power meter is slightly more expensive and requires voltage taps and watching the integration output as well, as well as some real time math from outputs in the cheap units.

wow . .. so much to learn :-)

 

[Ranjib]

What Kill A Watt unit did you use?

https://www.amazon.com/dp/B00009MDBU/

 

[dmolavi]

Fabrication question - When drilling into the enclosure, would it be better to use a step drill bit or a standard fluted bit? Regardless, I will have a 1/32 pilot to prevent walking.

 

[16Reefer170]

There are some new “smart” power strips on the market like https://www.kasasmart.com/us/products/smart-plugs/kasa-smart-wi-fi-power-strip-hs300. If you could integrate something like this, it would provide 6 outlet control and power monitoring.

 

[dmolavi]

There are some new “smart” power strips on the market like https://www.kasasmart.com/us/products/smart-plugs/kasa-smart-wi-fi-power-strip-hs300. If you could integrate something like this, it would provide 6 outlet control and power monitoring.

www.digital-loggers.com has a few 'smart strips', including ones with 16 switchable outlets. I use one with 8 switchable ones. Interfacing is easy, and requires no additional hardware - it's all networked for outlet control and status.

 

[theatrus]

There are some new “smart” power strips on the market like https://www.kasasmart.com/us/products/smart-plugs/kasa-smart-wi-fi-power-strip-hs300. If you could integrate something like this, it would provide 6 outlet control and power monitoring.

Most of the IoT strips require the central server to be up, and in turn an internet connection.