2017-03-04 08:10 AM
I'm using a simple capacitance to frequency converter with a ts555 to measure water level. After installation, I notice a large drift, so I returned the circuit to the lab to diagnose the problem. To eliminate the sensor as source of problem, I replaced it with a fixed (stable) value.
The drift seemed temperature related, so I tried to change each component's temperature separately, and was surprised that the power supply decoupling cap seemed the problem. I realized that probably the value was too low, so I changed with with several larger values, of different technologies, but the sensitivity didn't really go away.
I would like to simulate the circuit in Spice, but couldn't find a specific model for ST's TS555 version.
Questions:
Screenshot of the oscillator (C3 was changed to 22nF, then 100nF. T5/T6 are the terminals for the level sensor, replaced by a 560pF capacitor for the tests):