In the picture, the left is bad quality signal. The right is a good-quality signal. Notice the peaks are very well-defined. When we have good-quality signals, the bubble frequencies are accurate, while for bad-quality signals, it is difficult to describe a threshold to get proper bubbling frequency reading.
P.S.: It might look like it is a good idea to define a threshold for the trough to count signal, but the trough is also not well-defined; it is a coincidence that it looks good in the pictures above.
20:30
→
20:45
SBC code development (signal recognition)15m
Calibrated the output of ADC
binary output from ADC is converted to volts using ADC's internal calibration
testing board to take to KEK
trying to make the PGA inside ADC work
if it works, the voltage least count will be 1.25 mV instead of the default 6 mV