I am encountering some problems with the OD calibration form standards in my leader and workers, with version 25.3.5.
I am trying to run the calibration with 10 standards (OD600 = 0.05 to OD600 =1.5). However, there is only registration of voltage on the first standard (0.05), and then with the next standards the voltage registration is zero. In the UI, there is the change of OD measurement when I put the vials with different OD for measuring. You can see in the picture the jump between every OD standard.
Hm, that’s very strange. Has your Pioreactor’s optical system been working so far? Were any changes made to it? Getting a recording of 0 suggests that very little light is hitting the OD sensor.
So far the Pioreactor optical system has been working fine. I ran three workers last week (including the one I am having trouble with) and they successfully recorded OD and nOD for more than 70 hours without issues. The UI records these values without any problem.
Issues arise when I try to perform the calibrations in the terminal. So far, no changes have been made to the system, besides updating to 25.3.5.
@jb1404 did you run a previous OD calibration on this Pioreactor? We may have found a bug where if you have an active OD calibration, and rerun an OD calibration, the active OD calibration will be invoked when it shouldn’t.
Yes, previous OD calibrations were run on the leader, however, they were deleted in the terminal before trying this new calibration. The workers have not been OD calibrated.
How did you delete them - with pio calibrations delete? If you visit your Calibrations page in the UI, do you see an “active” calibration for device od for the leader Pioreactor?
My cluster is composed of a Leader and 8 workers. I ran the self-test on all of them and all Pioreactors except 2 workers gave the error: X “Photodiodes are responsive to IR LED”.
Plus, some of the workers also have the error:
X “No ambient IR Light detected”
X “Reference photodiode is correct magnitude”
X “Photodiodes measures zero”
I checked the connections and everything seems to be fine.
The two workers that did not give any errors on the self-test were able to be OD calibrated properly. I think it’s worth mentioning that one of those workers that work was the one that you helped me reinstall a couple of days ago.
Might be worth it to redo those self-tests with slightly turbid water/media (ex: one of your standards would work). I’d like to see Photodiodes are responsive to IR LED pass for all the Pioreactors.
Maybe it’s a firmware issue. On a Pioreactor that’s having the OD problem, can you report back the output from: pio version -v? And also pio calibrations list --device od?
Hi, I reinstalled the Pioreactor firmware 25.3.5 in all of the workers (with the method suggested by you in my previous question), and now I can properly calibrate for OD in all of them in the terminal.
However, even trying with turbid water (OD=0.1), some of them still have the errors in the self-test for Reference photodiode is in correct magnitude and Reference photodiode is in correct position. The UI gave these two errors:
You can ignore those warnings I think. The 0.256 limit is arbitrary, we just don’t want anything too too high. The other warning is related to the high REF signal, and can be ignored. Sorry for the confusion!
I am running the OD calibrations again in all my workers, but I have encountered a problem in one of them.
I can run the calibration properly, and all my standards (ranging from OD=0.05 to OD=1.5) are being read and recorded properly, however, the problem arises when I want to read my blank. My blank is completely transparent, however, this specific worker records a higher voltage than the one recorded for OD 0.05 and 0.1 standards, making the calibration unusable. In the other workers, this blank is recorded with a voltage of zero or very close to zero, always lower than my first two OD standards.
I have encountered this problem before, and I solved it by reinstalling the software in the terminal; however, this time, it doesn’t seem to work. I ran the self-test and got an error in “Reference photodiode is correct magnitude”. Any suggestions?
Another thing I realized is that another worker is reading and recording the calibrations well, detecting the change of OD of every standard proportionally and detecting that same blank as zero.
However, in comparison with the other 8 Pios I have in my cluster, the voltage readings are too low. For example, for all the other pioreactors OD = 0.05 equals a voltage close to 0.013 and OD = 1.5 to a voltage close to 0.5, however, this worker I am mentioning does not go above 0.15 in voltage. I have noticed that its OD readings in the UI are just a bit off compared to the other Pios.
Does the low voltage have an impact in the precision of the OD readings? If it does, how can I solve the issue? I already checked the connections, did a self test and reinstalled the software.
@jb1404 It’s expected for there to be some variation between Pioreactors. There could be some variation due to the IR LED, or the IR photodiode(s), or both. A few things you can do:
try to wipe the PD photodiodes with some alcohol to remove any residues.
check the interior of the IR & 90 LED pocket for any 3D printing artifacts: imperfect prints in the LED pocket aperture might be blocking light.
after an experiment / test, investigate the dataset IR LED intensities from the Export Data page. This tracks the IR intensity that the REF photodiode sees over time. Does this worker look very different from the others? That might pinpoint part of the cause of variation.
