Hi there,
I recently bought 6 pioreactors to test my anaerobic culture. The first trial was very good, but in the second one all OD measurement remained 0 after I started experiment.
I left everything default in the setting. I thought the PDs broken, so I tested OD determination with my cultured vial (from 1st test), it jumped up, indicating PDs are okay.
I also thought it might be because of the connection as I used local_access_point (as I was struggling with connecting all 6 together), so now I decreased to 2 pioreactors only, and used ethernet cable to connect my laptop with leader. I also increased IR intensity to 80 and decrease measurement time to 0.0025 (3 reads/20 mins), but the problem still remains.
Could you please give me some suggestions for this? Thank you so much.
Kind regards,
Ezio.
Ps I saw a similar post but seems not my case. So I created one.
Hi @EzioPham,
Hm, that’s strange. Nothing obvious jumps out at me. Can you run a self-test on the Pioreactors have a problem? Running a self-test | Pioreactor Docs
Hi @CamDavidsonPilon ,
Thanks for your prompt reply. The issue happened in 5 pioreactors (except the one newly added into the cluster and run for first time). I did self-test before running experiments and passed all. I also thought I did something wrong, so I re installed the software (newly released software) to one of my pioreactors and connect it using cable, but the issue is still there. It also confused me.
hm! Do you know if these are v1.1 Pioreactors, or v1.5?
I left everything default in the setting. I thought the PDs broken, so I tested OD determination with my cultured vial (from 1st test), it jumped up, indicating PDs are okay.
Like, the OD was reporting near nil, and you removed the current vial (which has some turbidity?), put in an obviously turbid vial, and the OD shot up?
These are all v1.1 pioreactors.
the OD was reporting near nil, and you removed the current vial (which has some turbidity ?), put in an obviously turbid vial, and the OD shot up?
Yes, it is. I started experiment with 10 mL medium + 1 uL cell (it looked transparent at the moment). I set temp: 30 oC, stirring: 800 rpm, optical reading: as default (samples_per_second=0.2, ir_led_intensity=auto), and left for 3 hours, then turned on growth rate function. The next day I found growth rate is off as OD measurement remained 0 all the time (even logs said: OD variance is zero - this suggests that the OD sensor is not working properly, or a calibration is wrong), but when checked the vial, I saw cell growth. When took that vial out and put the other vial (obviously turbid vial), the OD went up.
Attached photo for your reference:
Note very clear, but you can also see on my screen when I reinstalled a software to one of my pioreactors and gave a try. Use vial with black arrow for self-test, put vial with blue arrow, OD jumped up, put vial with red arrow, OD went back to 0. In this set, samples_per_second=0.002, ir_led_intensity=80.
Thanks for your time.
Thanks @CamDavidsonPilon for your time on troubleshooting.
- Till now, I only run 2 experiments on these vials. I usually started with a vial with water and do self-test for each pioreactor. I didn’t run blank because when I used calibrate → blank record, it was running and never stop.
- I exported this file and it seems like the value is not in the range you mentioned. Only 0.0xx. Another thing is I didnt see value of pior01 till I started growth rate calculating function. (will also send to you email for your reference).
- Yeah. I’ll send it to you.
Should there be any other things, please let me know. Many thanks.
0 is a strange value to even achieve, and I so I’m trying to find places where 0 could occur:
I see that your Pioreactors are collecting signal from the PDs:
ADC offsets: {'2': 3932.9750003345453, '1': 2574.250001629091}, and in voltage: {'2': 0.0990447001, '1': 0.0648277244}
That’s good.
These values look kinda suspicious though, specifically signal 2 (the 90deg PD). These ADC offsets are the “dark” signal, and subtracted from the signal early on. The ~0.099 is higher than I expect (about 0.06 is normal). So if your signal is less than 0.099, then it would be truncated to 0.
I also see you have other LEDs active. My current guess is
- the other LEDs in channel C & D are not turning off during this “dark” signal recording period (I’ll try to repro). What wavelength are those LEDs?
- too much stray ambient light is getting into the Pioreactors somehow.
This would be consistent with putting in a very turbid sample, and the Pioreactor sensing it.
That’s my current guess. I’ll do some more digging though.
Hi @CamDavidsonPilon,
Wow, thank you so much for the brief explanation. Yes, you are right I have C and D channels: C is white LED (7000-8000K), and D is red LED (620-625 nm). I am not sure if the LED was off when the pioreactors recorded OD. However, when looked at the pioreactors, I can see the white and red LEDs were on and off.
Please let me know if there are any other things I can do to support. Like moving all pioreactors into dark side and run tests.
Many thanks.
So, normally, the LEDs turn off whenever OD reading is one (you can watch the “LED intensity” box on the Pioreactor card to confirm). I wondering if there is some light leakage from these LEDs though. Can you try removing the LEDs from their pockets (putting the LED cap back on), and starting OD reading? Then, you can add them back. I’m wondering if this makes a difference.
BTW, I think your cluster clocks are unaligned. Can you check visit Inventory → Leader (on the left-hand side), and see if the cluster clocks are different there? You can try to reset them there. Do this outside an experiment.
Hi @CamDavidsonPilon ,
Thanks for your suggestion. It did make different.
I turned on 1 pioreactor and used a water vial for self-test, then turned on temp., stirring, and OD reading. After a while, I change the water vial to the one with medium + cells, left it for a while, and then turned on LED. The OD seems to be well-functional. Attached for your reference (please look at the green line):
l
I will also test the others and get back to you. Regarding the alignment, I haven’t done it yet as one pioreactor is still running for my labmate’s test with algae culture + 100% white LED. Just wanna know if they can grow. If so, he is going to make a purchase.
Hi @CamDavidsonPilon ,
Thanks for your patience. I just tested the other 4, and your hypothesis is right. If I turn on the OD reading a while before turn on the LED, the problem is solve. Please take a look at the following photo for your reference.
- During this test, I notice another issue with temperature control in pior5. It seems to fluctuate around and is not unstable as the others. Do you have any suggestion on this?
(Sorry about splitting the post into 2. I couldn’t embed 2 photos in one post)
- Also, when I update the clock as your suggestion, only the time in leader is updated. The others remains unchanged. Could you also suggest on this matter?
Thanks.
w.r.t. the clocks, we may need to make a software update to fix that. Something isn’t being applied correctly!
You can try sshing into a worker and running:
sudo chronyc -a makestep
Let me know if that fails!
If I turn on the OD reading a while before turn on the LED, the problem is solve.
This might involve a hardware fix later. I’ll do some tests here to see if I can introduce a software fix though.
Thanks, @CamDavidsonPilon,
I run sudo with the pioreactor, but the clock remained unaligned. Are there any alternatives to do it?
Okay, try this:
sudo date -s YYYY-MM-DDTHH:MM:SSZ
to set the time (in UTC, adjust accordingly)
For example
sudo date -s 2025-12-06T02:21:00Z
is the current time as of this post.
I run the code, but unfortunately, only the leader is updated. Should I update the software? I am currently running version 25.11.12 in these pioreactors.
hm, no, I’ll spend time tomorrow debugging and preparing a solution for you.
But the following would help me:
- SSH into a worker,
- Provide the output of
sudo chronyc tracking and sudo chronyc sources -n
- do the same for your leader
I think I know the problem, and a bug fix for next release. For now, I think you can run the following on the workers to sync the clocks to your leader:
sudo systemctl stop chrony
sudo chronyd -q
sudo systemctl start chrony
Sorry @CamDavidsonPilon for the late reply. I am being on vacation. Thanks for your help. All the time is now right and aligned with the leader.
Thanks once again for your help.
