PH probe issues

[Sral]

Curious. I don't quite know why the accuracy should be better, but maybe ReefPI doesn't use all three digits of resolution natively.

That errorData.dat reads: "1.711"
(ASCII Code 49 = 49-48 = 1)
(ASCII Code 46 = Comma)
(ASCII Code 55 = 55-48 = 7)
(ASCII Code 49 = 49-48 = 1)
(ASCII Code 49 = 49-48 = 1)

That means communication was successful, but the value was below 2.0, so it got dumped into the file. So this part of the code works, but I'm still stumped to why the dip occurs. Especially since it completely throws the axis off as well. If it were simply -0.15 it should just be a normal numerical axis. Very weird.

Can you try and keep an eye on the contents of that file ? That might inform us to what is happening here.

Ultimately we might need to try another thing: rejecting any value that deviates by more than 1.0 from the previous, as long as it isn't consistent. I'll think about it.

 

[iamdan]

Sorry for the late reply, been sick as a dog.

Just checked that error data file - this is the latest in it

 

[iamdan]

Before @robsworld78 went on his holidays I asked him some stuff about his pH probe and he says he has it in its code on the chip getting a reading every 3 seconds - would this be why it glitches out? If, regardless whatever time we set reef-pi to get its reading it glitches out because reef-pi is maybe asking for a reading when it’s already in the process of doing so already?

Now I could have misinterpreted what Rob had told me however if that’s the case, then to me that kinda makes sense because I setup another raspberry pi and hooked up a atlas ph ezo board with its power isolation board and in reef-pi if I set it to 3 seconds it’s leds status changes every 3 seconds or if I set it to 60 seconds the leds only change status every 60 seconds.

For giggles I set up the rpi and the atlas probe in the same cabinet and put a probe in the tank and no glitches so definitely nothing in the tank that’s causing it to glitch out or any rf/emi interference

Thoughts?

 

[Sral]

Before @robsworld78 went on his holidays I asked him some stuff about his pH probe and he says he has it in its code on the chip getting a reading every 3 seconds - would this be why it glitches out? If, regardless whatever time we set reef-pi to get its reading it glitches out because reef-pi is maybe asking for a reading when it’s already in the process of doing so already?

Now I could have misinterpreted what Rob had told me however if that’s the case, then to me that kinda makes sense because I setup another raspberry pi and hooked up a atlas ph ezo board with its power isolation board and in reef-pi if I set it to 3 seconds it’s leds status changes every 3 seconds or if I set it to 60 seconds the leds only change status every 60 seconds.

For giggles I set up the rpi and the atlas probe in the same cabinet and put a probe in the tank and no glitches so definitely nothing in the tank that’s causing it to glitch out or any rf/emi interference

Thoughts?

Well, I do know some of the code and the microcontroller makes a measurement every 3 seconds. The request from ReefPi should not lead to it glitching out, as far as I can see ... though I can't guarantee that it wont either. If it did, I would have expected more people to see the same behvior.

The problem with the test that you described is that it's not just the cabinet, but the whole setup. You would need to hook up that Atlas pH EZO board to the same PI as your Robo-Tank pH board.

Still strange though, that you even get a value of 1.7 and 1.5 pH. That sounds like a software artifact or an electrical issue, as that is roughly the very far end of the measurement range, which is somewhere between 0 and 2 pH, depending on resistor deviations and your probe.

 

[iamdan]

One thing I found is on my test pi with atlas ezo ph board and it’s isolater is as soon as I hooked up a ds3231 rtc it started glitching out. Didn’t matter how I connected it. Leaving the ph ezo device as the only thing on the i2c bus and it’s rock solid.

As for this issue, I have 2 Robo tank controllers on 2 separate tanks on opposite ends of the house, running totally different equipment/brands, on different house electrical circuits, and the ph module does the exact same glitch

I do have another ph ezo board and its isolater I can connect to either one of the Robo tanks as long as it doesn’t conflict with the Robo tank one

 

[Sral]

One thing I found is on my test pi with atlas ezo ph board and it’s isolater is as soon as I hooked up a ds3231 rtc it started glitching out. Didn’t matter how I connected it. Leaving the ph ezo device as the only thing on the i2c bus and it’s rock solid.

As for this issue, I have 2 Robo tank controllers on 2 separate tanks on opposite ends of the house, running totally different equipment/brands, on different house electrical circuits, and the ph module does the exact same glitch

I do have another ph ezo board and its isolater I can connect to either one of the Robo tanks as long as it doesn’t conflict with the Robo tank one

That would be a great test, although I think the RTC (or the presence of other I2C boards in general) seems to be to blame here.

I remember a few people having kind of hourly issues with different things and I would not be surprised if the RTC-handling causes this.

 

[biznatch]

Came across this thread as I just started having the same issue with 2 different robotank/reef pi 5.3 installs. Issue started yesterday, and the only change I made was setting up the reef-pi Home Assistant integration. Once it starting polling reef-pi, I started seeing consistent drops on the PH graph in the UI on both instances. One drops to .45, the other to .25 (I'm assuming it's the different calibrations) at almost predictable intervals.

As a test I removed one of the reef pi devices from the HA integration last night. Since removing, I've had no drops in the PH graph, while the other install continued the issue overnight. The HA integration just polls the API (@60s by default I believe), but seems to be what is causing this issue to manifest in reef-pi.

 

[Sral]

Came across this thread as I just started having the same issue with 2 different robotank/reef pi 5.3 installs. Issue started yesterday, and the only change I made was setting up the reef-pi Home Assistant integration. Once it starting polling reef-pi, I started seeing consistent drops on the PH graph in the UI on both instances. One drops to .45, the other to .25 (I'm assuming it's the different calibrations) at almost predictable intervals.

As a test I removed one of the reef pi devices from the HA integration last night. Since removing, I've had no drops in the PH graph, while the other install continued the issue overnight. The HA integration just polls the API (@60s by default I believe), but seems to be what is causing this issue to manifest in reef-pi.

You are a blast mate, thanks a lot !

@Ranjib read the above, I think that is important for a few people.

 

[iamdan]

Came across this thread as I just started having the same issue with 2 different robotank/reef pi 5.3 installs. Issue started yesterday, and the only change I made was setting up the reef-pi Home Assistant integration. Once it starting polling reef-pi, I started seeing consistent drops on the PH graph in the UI on both instances. One drops to .45, the other to .25 (I'm assuming it's the different calibrations) at almost predictable intervals.

As a test I removed one of the reef pi devices from the HA integration last night. Since removing, I've had no drops in the PH graph, while the other install continued the issue overnight. The HA integration just polls the API (@60s by default I believe), but seems to be what is causing this issue to manifest in reef-pi.

Yeah that’s interesting! I wonder what code on either side is causing this

 

[AbjectMaelstroM]

Same issue on my end for several months now after switching to HA integration. Glad I'm not the only one (sorry guys, haha).

I kinda learned to live with it, but it would be nice to find a fix.

Robotank running ReefPi 6.0.