Pioreactor1 going offline when ending experiment

Using the Pioreactor Troubleshooting
21

I tried cd /var/www/pioreactorui/ && sudo pip install -r requirements.txt anyways and I think there was one thing that it ended up installing. That actually ended up fixing it and now it is showing up on the web page. Unfortunately I tried the cat command and now I don’t know how to look at everything before the cat output.

Thanks for the fix!

Edit:

I did notice these two error messages pop up in the event log.

17:20:58 pioreactor1 get job contrib Yaml error in /var/www/pioreactorui/contrib/jobs/08_self_test.yaml: Object missing required field type - at $.published_settings[0]
17:20:58 pioreactor1 get job contrib Yaml error in /var/www/pioreactorui/contrib/jobs/07_od_blank.yaml: Object missing required field type - at $.published_settings[0]
22

YES!

Okay the problem is how I deploy code. I deployed a small update to the UI this weekend and was like “oh this will be fine expect for that one edge case” and wow we hit that one case.

I’ll be adding more checks and safeguards to my deployments so problems like this happen less and less.

Thank you for your assistance and patience!

BTW, I fixed your other problem in the meantime, too. Run pio update app -b develop to get the latest fix.

23

Thanks a lot! I appreciate the help.

24

Yaml error

Right, that’s annoying. Run this:

sudo cp -R /var/www/pioreactorui/{jobs,charts,automations} /var/www/pioreactorui/contrib/

The fix is in a future version.

Edit: That command might fail, if so, try:

`

sudo wget -O /usr/local/bin/update_ui.sh https://raw.githubusercontent.com/Pioreactor/CustoPiZer/pioreactor/workspace/scripts/files/bash/update_ui.sh

followed by

pio update ui -b develop
25

Thanks. The first command failed, so I tried the second two commands. I’ll let you know if I continue encountering the test errors.

One last issue I’ve noticed is that the temperature automation does not appear to be working currently. The temperature automation appears to be starting correctly, and to be measuring temperature correctly, but it does not appear to be heating up. For example, I’ve set it to 30C, but currently it is reading a temperature of 18C. I did take it apart the other day, and I replaced the flex cable with a longer flex cable (7 pins, 1mm pitch, A-Type), so I can’t be 100% sure it isn’t something like a loose cable or bad connection. Is there a way to check whether it is a software or hardware issue?

26

Hm, in your logs (pio logs from the command line), does it saying something like “Starting thermostat(target_temperature=30)” after starting it? That would tell me that the thermostat automation is starting correctly.

Next, in your logs, do you see something like (every 4minutes or so):

{'previous_heater_dc': 32.35877951429195, 'room_temp': 22.0, 'time_series_of_temp': [38.1625, 37.0375, 36.025, 35.25, 34.5375, 33.9375, 33.375, 32.875, 32.425, 32.1125, 31.7625, 31.45, 31.1875, 30.9875, 30.75, 30.5375, 30.3875, 30.225, 30.1, 29.975, 29.825, 29.6875, 29.5875, 29.5, 29.4, 29.35, 29.2625, 29.1375, 29.175]}

If so, what does the sequence of previous_heater_dc look like? Is it increasing towards 100?

27

Here’s the log.

2023-02-17T11:08:28-0800 [temperature_control] DEBUG features={‘previous_heater_dc’: 100.0, ‘room_temp’: 22.0, ‘time_series_of_temp’: [20.1875, 20.0625, 19.9375, 19.875, 19.8125, 19.75, 19.6875, 19.65, 19.625, 19.6, 19.5625, 19.55, 19.5, 19.5, 19.5, 19.45, 19.4375, 19.425, 19.4375, 19.4125, 19.4125, 19.375, 19.375, 19.375, 19.375, 19.375, 19.375, 19.375, 19.375]}

I had one of my experiments overflow a little, so I took it apart to clean it up. I suspect the issue could be that one of the electrical contacts for the heater got coated in the nutrients and it isn’t getting a proper electrical contact. I just wanted to see if there was a way to confirm whether it was a software issue or not before I take it apart again.

28

So it’s sending a signal of 100, which is way more than needed. We can try to test it directly:

  1. Turn off the automation from the UI
  2. Once off, on the command line, try: pio run self_test -k temperature

This will run a quick self-test and will tell you if it is able to properly heat.

Liquid damage is possible. Definitely try inspecting it after the above finishes unsuccessfully. So it sounds like it did work at one point?

Edit: Actually you may have to turn off all the jobs to run that command, and maybe not something you want to do right now.

29

Yeah, it used to work. Then, it overflowed (I noticed it pretty quickly and shut it off), so I took it apart, I printed some new holders to separate the fan/heater/container from the Raspberry Pi, I bought a new flex cable, re-flashed the Pioreactor a couple times, and then had the issues in this thread.

It overflowed maybe two weeks ago, then I haven’t tried turning it on until the past day or two. I think it might be worth it to just try and fix it and restart the experiment later. I’ll try running the self test in a bit. I think I might also try and probe the different connections with a multimeter and double check the connections. Do you have any circuit drawings/schematics that would help with this?