Hi guys,
Almost ready and I will write a full report of what we did to get good temperature calibration (as well as OD, in fact). Only thing is that on one of the machine, there are 2 smart sleeves that, despite different calibrations, always end up significantly lower than the target temperature once we set it for an experiment. Always the same two sleeves. Any advice on what to change first ? Thermistors? Heating blocks ?
Thanks !
I’ve noticed that too sometimes on the newer version of the eVOLVERs. I am not sure why it’s changed since then. It probably means theres a software bug somewhere. Does the thermistor readout accurately indicate the temperature? In other words, if set at 30C, it’s reading properly at say 28.5C but not reaching 30C. If the readout is accurate, then it is probably the PID constants on the Arduino that need tuning.
Hi there!
Following up on this, temperature readings indicate the correct temperature that we set, but the actual temperature of the culture is almost always 0.2 - 0.3 ºC lower. The easy workaround is to compensate it by adding the offset to the temperature in the experiment settings.
I believe an easy way to get temperature calibration even more accurate is to measure each vial individually instead of measuring all of them and retrieving the raw values from eVOLVER later. Any idea on how to do this easily?
Sorry I guess im not following the logic on the 2nd point. Can you elaborate?
Sure! Right now, when we run a calibration, we measure the internal temperature of each culture. Since we only have two probes, we measure two cultures, change the probes to another two and wait a couple of minutes until the readings are stable. When we finish doing all the measurements of the 16 vial with the external probes, we write the values to the calibration interface and tell the eVOLVER to retrieve the raw measurements for all vials.
This way, the temperature of the vial may have oscillated between the moment the probe measurement was made and the moment the raw value was acquired. Thus, my suggestion is that acquiring the raw values right after measuring the temperature externally may improve the calibration.
Hope it is clearer now!
This would probably be difficult with the GUI but would be reasonable by modifying the python script (which is typically used for the DPU).
server_test.py is a script that we use for testing and is a simple example of how to send commands to the eVOLVER via python. If you run the script, it should send random commands to the eVOLVER (as a stress test) and receive the current state of the machine. I think the simplest thing you can do if you want to generate your own calibration programs, is to modify this script to do what you need.
Hope this helps!
Brandon