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

[burningbaal]

Is this an acceptable method of running my equipment every 2.5 hours ?

I can't speak to if the syntax is handled the way you think; haven't tried. But 180 would be every 3 hours, right? do you want */150 instead?

 

[Phildago]

I can't speak to if the syntax is handled the way you think; haven't tried. But 180 would be every 3 hours, right? do you want */150 instead?

lol yes, 150. but would it work?

 

[Ranjib]

Is this an acceptable method of running my equipment every 2.5 hours ?

Most likely not. You want the equipment to just turn on/off after every 2.5 hours?

 

[Phildago]

Most likely not. You want the equipment to just turn on/off after every 2.5 hours?

yes, its for dosing pumps that are run for 1 second. I'd like to just keep it simple and set the number of minutes. Between each dose. would there be another way to do this?

 

[Ranjib]

yes, its for dosing pumps that are run for 1 second. I'd like to just keep it simple and set the number of minutes. Between each dose. would there be another way to do this?

Im realizing we cant not do this easily, if at all. My first attempt is to come up with two different timers, one with
hour: 1,6,11,16,21 minute: 0, second:0, day/month:*
the other
hour: 3,8,13,18,23 minute 30, second:0, day/month:*
but this leaves the last dosing 1:30 hours apart, not 2:30 :-/

Let me think about this more. If i dont find any decent solution, I'll add a new feature. A simpler timer that just lets you define the periods in between action, instead of calendar time.

 

[Bigtrout]

Just a lil reef-pi humor !

 

[Phildago]

Im realizing we cant not do this easily, if at all. My first attempt is to come up with two different timers, one with
hour: 1,6,11,16,21 minute: 0, second:0, day/month:*
the other
hour: 3,8,13,18,23 minute 30, second:0, day/month:*
but this leaves the last dosing 1:30 hours apart, not 2:30 :-/

Let me think about this more. If i dont find any decent solution, I'll add a new feature. A simpler timer that just lets you define the periods in between action, instead of calendar time.

That sounds amazing. My tank is at that point where it is rapidly accelerating in alk and CA consumption and I have to often adjust the number, dose frequency would be awesome and so much easier

 

[burningbaal]

That sounds amazing. My tank is at that point where it is rapidly accelerating in alk and CA consumption and I have to often adjust the number, dose frequency would be awesome and so much easier

@Ranjib it seems a feature could be a simple entry to let you put in a modulo; it would still mean he'd have to get an even number of times per day, but what about a feature like this:

Enable macro/timer when:
[ dropdown for seconds/minutes/hours/days ]
modulo [ text entry for int value ] [dropdown for seconds/minutes/hours/days/months]
equals [ text entry for int value ]

then he could choose [minutes], type [144], choose [day], type [0].

Whenever the number of minutes in the day % 144 = 0, it runs the macro/timer.
He'd get exactly 10 doses per day

 

[Ranjib]

@Ranjib it seems a feature could be a simple entry to let you put in a modulo; it would still mean he'd have to get an even number of times per day, but what about a feature like this:

Enable macro/timer when:
[ dropdown for seconds/minutes/hours/days ]
modulo [ text entry for int value ] [dropdown for seconds/minutes/hours/days/months]
equals [ text entry for int value ]

then he could choose [minutes], type [144], choose [day], type [0].

Whenever the number of minutes in the day % 144 = 0, it runs the macro/timer.
He'd get exactly 10 doses per day

I was thinking more in the line of after every N seconds (or minutes or hours). Simple and intuitive.

 

[ummwhy]

Possibly. The error here indicates at that time the sensor returned corrupted data (checksum mismatched error). reef-pi will discard such result. But if the timeline matches (error logged and spike in the graph), it likely means during that time period there were few values which has checksum correct but incorrect reading.

Any ideas how to troubleshooting ???

 

[dmolavi]

I made a live stream of my tank...no audio. This will let me watch while I'm at work

 

[crusso1993]

I made a live stream of my tank...no audio. This will let me watch while I'm at work

Very nice!

FWIW - had to go to YouTube to watch the video due to your YouTube settings. I presume you already know this but, just in case you do not, figured I would let you know.

 

[dmolavi]

Very nice!

FWIW - had to go to YouTube to watch the video due to your YouTube settings. I presume you already know this but, just in case you do not, figured I would let you know.

Yeah I have embedding disabled, for now.

 

[Ranjib]

Yeah I have embedding disabled, for now.

I can see it fine.

 

[dmolavi]

I can see it fine.

I enabled embedding

 

[burningbaal]

I was thinking more in the line of after every N seconds (or minutes or hours). Simple and intuitive.

Certainly is straightforward that way. I was thinking it would be a little tricky to deal with that system across system reboots and things, but certainly a solvable problem by storing the last run time in the db

Speaking of, it looks like the 'seconds' field in a few places (like 'turn equipment back off after X seconds' in the timers) is supposed to be a dropdown to select other options (minutes/hours), but I can't seem to click it...is that supposed to be a dropdown? Perhaps that's a feature I could work on for v4

 

[Michael Lane]

Certainly is straightforward that way. I was thinking it would be a little tricky to deal with that system across system reboots and things, but certainly a solvable problem by storing the last run time in the db

Speaking of, it looks like the 'seconds' field in a few places (like 'turn equipment back off after X seconds' in the timers) is supposed to be a dropdown to select other options (minutes/hours), but I can't seem to click it...is that supposed to be a dropdown? Perhaps that's a feature I could work on for v4

The seconds label isn't a drop down yet, but the UI design was intended to eventually support a unit selection. It would be nice to create a time duration component.

 

[Schreiber]

So, I've got one of the @Michael Lane pH boards & I'm having an issue I can't seem to eliminate. Whenever my return pump kicks on or off, I see a huge spike in pH that drops back down over the span of a minute or so.

At first, I thought it may be an issue with my power supplies, like @Bigtrout was seeing in his build. So, I stopped using those buck converters all together & switched to powering the Pi through a quality USB plug. No luck, same issue.

Then, I got to thinking it may be an issue with my Leviathan board design. I had thought the ground planes would have prevented any interference, but to check, I took it out of the equation. At this point, all I have connected is the USB power & the pH board directly to the Pi. Same issue.

Then, I thought maybe since the USB brick & return pump are on the same power strip, there might be some weird coupling interference between the two, so I took the brick off that strip & simply plugged the USB into my laptop. So, I'm now completely isolated from the pump, no wires are anywhere near it. Same issue.

Well, now all I can think of is some sort of electrical interference being transferred over-air when the pump shuts off. So, I wrap the pH board in multiple layers of aluminum foil to shield it.... you guessed it, same issue.

My last resort was to wrap the cup of pH buffer solution I was calibrating the probe in as well, in case the probe itself was what was picking up the interference. No surprise, no dice.

I don't have this issue with *any* other equipment. What in the world could be causing this??

I've spent days troubleshooting this issue & am about to just give up, set the pH check interval to >1 minute, & just leave it be. But that still has the possibility of picking up a reading when it's artificially low... I'm stumped.

 

[Bigtrout]

The return pump could be causing intereference back thru the AC mains...that was my issue.

 

[Schreiber]

The return pump could be causing intereference back thru the AC mains...that was my issue.

How did you eliminate that?

Also, if that's the issue, shouldn't have my Pi & probe completely isolated (powered by my laptop running on battery) solve it?