Signal Conditioning Pipeline

Trace a raw sensor voltage through gain, filtering, anti-alias protection, and ADC sampling

Animation Signal conditioning ADC readiness

Signal Conditioning Pipeline

Follow a weak, noisy sensor voltage as it is amplified, filtered, protected against aliasing, and sampled by an ADC. The goal is not a pretty waveform; the goal is data the microcontroller can trust.

Needs gain current pipeline condition

12% range ADC input range used

22% noise cut interference remaining

Nyquist ok filter cutoff vs sample rate

Signal Conditioning Workbench

What A stage-by-stage view of the analog front end before an IoT device stores a sensor value.

Why Bad conditioning can create clipping, noisy readings, or aliasing that software cannot reliably repair.

Try First Press Play, then switch between Good pipeline and Aliasing risk to see why filtering before the ADC matters.

Notice Gain improves usable range, but the filter and sample rate decide whether high-frequency noise reaches the ADC.

Controls

Analog Front End

Amplifier gain 2x

Low-pass cutoff 25 Hz

ADC sample rate 120 /s

ADC resolution 12-bit 8-bit ADC 10-bit ADC 12-bit ADC 16-bit ADC

Include high-frequency interference before filtering

Presets

Pipeline Workbench

A weak sensor output needs gain before the ADC can use enough of its input range.

Needs gain

Learning Support

Gain uses the ADC range

A sensor may produce only millivolts. Gain expands that signal so ADC codes change enough to measure it.

Bias centers the signal

Many ADCs cannot read negative voltage. A mid-supply bias lets an AC-like sensor swing above and below a center point.

Filter before sampling

The anti-alias filter must be analog and before the ADC. Digital filtering after sampling cannot undo aliasing.

Watch for clipping

Too much gain can push the signal outside 0 to Vref. Once clipped, the original peak is lost.

Quick Reference

Amplifier output

v_adc = bias + gain x v_sensor

For a single-ended ADC, bias keeps the conditioned waveform inside the ADC input range.

ADC range check

0 V <= v_adc <= Vref

If the conditioned signal exceeds this range, the ADC clips to its minimum or maximum code.

Low-pass filter

|H(f)| = 1 / sqrt(1 + (f / fc)^(2n))

This page uses a Butterworth-style teaching model with filter order n = 2.

Nyquist limit

f_nyquist = sample_rate / 2

Analog frequencies above this limit can fold into the sampled data as false lower frequencies.

Anti-alias target

filter cutoff < sample_rate / 2

Practical systems leave transition-band margin because real filters do not cut off instantly.

ADC step size

LSB = Vref / 2^bits

The smaller the LSB, the smaller the voltage change represented by one ADC code step.

Useful range

range_used = signal_span / Vref

Too little range wastes ADC resolution. Too much range clips the waveform.

Stage order

sensor -> gain/bias -> analog filter -> ADC

The exact circuit can vary, but alias prevention must happen before the ADC samples.

Guided Practice

Use more range

Start at Weak sensor and raise gain until the range metric becomes useful without clipping.

Remove interference

Lower the cutoff and watch the interference bar shrink. Keep the signal frequency below the cutoff.

Trigger clipping

Use Clipping and reduce gain until the warning clears. Notice that high resolution cannot fix clipping.

Find aliasing risk

Use Aliasing risk and raise the ADC sample rate or lower the cutoff until the Nyquist warning clears.

Sampling & Nyquist VisualizerSee how sample rate creates or avoids aliasing. ADC Resolution VisualizerConnect conditioned voltage to digital code steps. Low-Pass Filter EffectExplore cutoff frequency and noise reduction more directly. IoT Formula CalculatorUse ADC, sampling, and data-rate formulas together.