Table of Contents
Rigol DS1102D Oscilloscope, with Logic Analyser
Schematic
Log
2026-02-21
- Eugene: CH1 was misbehaving, so I started tracing the analog signal from the very input. The analog front-end itself checked out fine, but the bias voltages were clearly off. On the working channel (CH2) the position bias swept cleanly from 0V to −1V as expected. On the faulty CH1 it was stuck in a shifted range of −0.72V to −1.2V — enough to permanently overdrive the analog input stage and corrupt every measurement.
Traced the bias voltages back through the circuit, past the TLC274 buffer op-amp, all the way to the HC4051 analog multiplexer. The HC4051 is used here as a demultiplexer — a single DAC generates all the bias and position voltages and the mux time-shares it out to each consumer. The DAC signal looked correct for every other channel it served, which ruled out a hardware DAC fault and pointed squarely at the digital domain: a stored calibration value for CH1 position was simply wrong.
Running the built-in self-calibration routine fixed it immediately. Both channels now show identical, correct bias sweeps
Why was the calibration off in the first place? Hard to say for certain. One theory is a flaky inter-layer PCB contact that may have caused a mis-calibration at some point in the scope's past — and interestingly, disassembly and reassembly may have actually healed it. The self-cal then wrote correct values and the problem hasn't returned.
