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

Yeah, I did try that and couldn't make it work either. I finally gave up and, after a great deal of searching, found an old copy of raspberry pi os from 4 years ago that worked with reef-pi.
Dang, glad you found something that worked for you !
 
Do you have an isolated carrier board? I've got an isolated carrier board, and a pH chip installed in my reef-pi circuit, but need a probe still to test it.
Yes, I bought their isolated carrier board, I went from getting +/- 1pH swings to +/- 0.2pH but according to EZO its still too much. Im talking to their tech support to figure this out
 
Curious, though it seems that if you get those fluctuations even with the isolated carrier board your whole setup or tank might have an issue. You could create a separate thread for the troubleshooting and I can try to see if I can help you out.

If you don't find any solution I could write you a file driver that averages a few readings together and see if that helps removing the fluctuations.
Sure I'll create a new thread on this
 
And there's a Sainsmart relay switch to turn it on (off), but the only way to do this is with an ATO. In this case, I've set the alarm to go if it is on for longer than 100s (check frequency is 60s). So, if it's off for longer than one check cycle, there's a blockage problem and I'll get an email (my tank is not at home, but at my school where I work.
I've made something similar for my Skimmer with a custom script (made by AI). It controls skimmer pump based on cup level and also on sump level. With macro it does work but I have several errors in the log...
 
that is very strange. let me check
@Ranjib with help of AI I've found the problem, it's in the Reef-Pi Home Assistant Integration. Here is the report:

Root cause​

The issue is in the integration’s __init__.py file, specifically in the update_atos() function.

The code currently parses the timestamp returned by Reef-Pi using:


ato_states[id]["ts"] = datetime.strptime(
ato_states[id]["time"], REEFPI_DATETIME_FORMAT
).replace(tzinfo=dt_util.UTC)

The problem is that Reef-Pi appears to return timestamps in local time, but the integration assigns the UTC timezone directly using .replace(tzinfo=dt_util.UTC).

This does not convert the timestamp to UTC; it simply marks the local time as UTC.

As a result, Home Assistant applies the local timezone offset again, causing the displayed time to be incorrect.

I replaced the timestamp conversion with:


ato_states[id]["ts"] = dt_util.as_utc(
datetime.strptime(
ato_states[id]["time"], REEFPI_DATETIME_FORMAT
).replace(tzinfo=dt_util.get_time_zone("Europe/Rome"))
)
This change needs to be applied in both timestamp conversion branches of update_atos().

I also noticed the same issue in update_pumps(), where the timestamp is parsed using the same UTC assignment.

I applied the equivalent fix there:


time = dt_util.as_utc(
datetime.strptime(
current["time"], REEFPI_DATETIME_FORMAT
).replace(tzinfo=dt_util.get_time_zone("Europe/Rome"))
)
After replacing the code and restarting Home Assistant, the timestamps are now displayed correctly.

For example:

Before the fix:

  • Actual event time: 20:25
  • Displayed event time: 21:25
After the fix:

  • Actual event time: 20:42
  • Displayed event time: 20:42
The fix has been successfully tested on my installation running integration version 4.0.6.
 

TOP 10 Trending Threads

HOW WOULD YOU FEEL ABOUT RECEIVING A FISH OR CORAL AS A HOLIDAY GIFT?

  • I’d love one, as long as it’s a species I chose

    Votes: 19 26.0%
  • I’d trust someone who knows my tank to choose

    Votes: 4 5.5%
  • I’d rather get a gift card and pick it myself

    Votes: 28 38.4%
  • I’d prefer gear or supplies instead

    Votes: 2 2.7%
  • Please don’t surprise me with livestock

    Votes: 20 27.4%
  • Other (Post in the comments)

    Votes: 0 0.0%
Back
Top
Home
Post thread…
Market
What's new