Oscilloscope Timebase Calibration

Overview

The oscilloscope's timebase determines how accurately it measures time and frequency. This guide shows how to calibrate using GPS atomic clock accuracy.

Key Insight: Same GPS methods used for TinySA/NanoVNA work perfectly for oscilloscopes!

Quick Method: GPS 1PPS

If You Already Built GPS Setup

From TinySA/NanoVNA calibration: - GPS module with 1PPS output: Ready ✓ - Arduino (optional): Not needed for scope!

New procedure: 1. Connect GPS 1PPS to oscilloscope input 2. Measure pulse characteristics 3. Calculate timebase error 4. Document or apply correction

Time: 15-30 minutes Cost: $0 (reuse existing GPS)

Step-by-Step Procedure

Step 1: GPS Setup (if not already built)

See: gps_calibration.md for detailed GPS module setup

Quick version: 1. GPS module (NEO-6M/7M/8M): $10-20 2. Power: 3.3V or 5V 3. Antenna: Included ceramic patch 4. Wait for lock: 30-120 seconds 5. 1PPS output: Blinks once per second

1PPS signal characteristics: - Frequency: 1 Hz (period = 1.000000 seconds) - Pulse width: Typically 100-200 ms - Voltage: 3.3V or 5V TTL - Accuracy: ±50 nanoseconds (±0.00005 ppm!)

Step 2: Connect to Oscilloscope

Connection:

GPS Module          Oscilloscope
                ──────────────────────────────────
                1PPS output    →    CH1 input
                GND            →    GND (scope ground clip)
                VCC (5V)       →    USB power or bench supply

Scope settings: 1. Channel 1: - Coupling: DC - V/div: 1V or 2V (to see ~3-5V signal clearly) - Position: Center

  1. Timebase:
  2. Time/div: 200 ms/div (to see ~1 second period)
  3. Trigger: CH1, rising edge

  4. Trigger level: ~1.5V (mid-level)

  5. Acquisition:

  6. Mode: Normal or Auto
  7. Average: 16 or 32 (reduces jitter)

Step 3: Measure Period

Method A: Cursor Measurement

  1. Enable cursors:
  2. Press CURSOR button

  3. Select TIME cursors

  4. Place cursors on rising edges: Cursor 1 → First rising edge Cursor 2 → Second rising edge (one pulse later)

  5. Read delta time (ΔT): Display shows: ΔT = 1.0023 s (example - your scope will differ) Expected: 1.000000 s Error: +2.3 ms = +2300 ppm!

Method B: Frequency Measurement

  1. Enable frequency counter (if scope has one):
  2. Press MEASURE
  3. Select Frequency

  4. Source: CH1

  5. Read frequency: Display shows: Freq = 0.9977 Hz (example) Expected: 1.0000 Hz Error: -0.0023 Hz = -2300 ppm

Method C: Pulse Width

  1. Measure pulse width:
  2. MEASURE → Width+ (positive pulse width)

  3. Should be stable value

  4. Calculate period: If pulse width = 100.0 ms And duty cycle is known (typically 10% for GPS 1PPS) Period = width / duty_cycle

Note: Period measurement is better than width.

Step 4: Calculate Timebase Error

From period measurement:

Measured period: T_measured (from scope)
                Actual period: T_actual = 1.000000 s (GPS is atomic clock)
                
                Error (seconds) = T_measured - T_actual
                Error (ppm) = (Error / T_actual) × 10^6
                
                Example:
                T_measured = 1.000025 s
                T_actual = 1.000000 s
                Error = +0.000025 s = +25 μs
                Error (ppm) = 0.000025 / 1.0 × 10^6 = +25 ppm
                
                Interpretation: Scope timebase is 25 ppm FAST

From frequency measurement:

Measured frequency: F_measured
                Actual frequency: F_actual = 1.000000 Hz
                
                Error (ppm) = (F_actual - F_measured) / F_actual × 10^6
                
                Example:
                F_measured = 0.999975 Hz
                F_actual = 1.000000 Hz
                Error = (1.0 - 0.999975) / 1.0 × 10^6 = +25 ppm
                
                Same result: Scope is 25 ppm fast

Alternative: Mains Frequency

If no GPS available:

Using 50/60 Hz Mains

Warning: This is less accurate but free!

Mains frequency accuracy: - Short term (seconds): ±0.1 Hz - Long term (hours): ±0.01 Hz - Utility companies maintain accurate frequency - Not as good as GPS, but usable

Procedure:

  1. Build isolation circuit (IMPORTANT - SAFETY!): ``` NEVER connect scope directly to mains!

Instead: Use transformer Mains → Small transformer (12V or 9V output) → Scope input

Or: Use phone charger Mains → USB charger → Monitor 5V ripple (120Hz in US) ```

  1. Measure frequency:
  2. Expected: 60.000 Hz (US) or 50.000 Hz (EU)

  3. Scope reading: Compare

  4. Calculate error (same as GPS method)

Accuracy: ±100 ppm (okay for rough check)

Applying Corrections

Method 1: Document the Error

Simplest approach:

  1. Measure timebase error: (e.g., +25 ppm)
  2. Create correction table: ``` Timebase Error: +25 ppm Scope displays 1.000 s → Actual is 0.999975 s Scope displays 1.000 ms → Actual is 0.999975 ms Scope displays 1.000 μs → Actual is 0.999975 μs

Correction factor: 0.999975 ```

  1. Label scope: Stick label on scope: "TIMEBASE: +25 ppm Multiply displayed time by 0.999975 for actual time"

Method 2: Internal Calibration (if accessible)

Some DSO1013D models have calibration menu:

  1. Enter cal mode:
  2. Power off
  3. Hold RUN/STOP while powering on

  4. Or: UTILITY → CAL (hidden menu)

  5. Find timebase cal:

  6. Look for "TIMEBASE CAL" or "FREQUENCY CAL"

  7. Adjust:

  8. Usually a numerical entry
  9. Enter PPM correction
  10. Save

Consult your specific firmware documentation!

Method 3: Mental Math

For quick measurements:

Error is +25 ppm
                
                If scope shows 1.000 ms:
                Actual = 1.000 × (1 - 25/10^6) = 0.999975 ms
                
                Usually ignore for rough work
                Apply correction for precision measurements

Verification Methods

Cross-Check 1: Known Frequency

Use calibrated signal generator (if available): - Set to 1.000 MHz - Measure on scope - Should read 1.000 MHz ± your ppm error

Or use calibrated TinySA: - Set TinySA to CW mode, 1 MHz - Feed to scope - Measure frequency

Cross-Check 2: Crystal Oscillator

Build simple crystal oscillator:

Materials:
                - 32.768 kHz watch crystal ($0.50)
                - CD4060 divider IC ($0.50)
                - Resistor, capacitors
                
                Output: Exact 1 Hz from watch crystal
                Accuracy: ±20 ppm (watch crystal)
                
                Measure on scope, should match GPS within ±50 ppm

Cross-Check 3: Multiple GPS Modules

If you have two GPS modules: - Both produce 1 Hz - Should agree to within ±0.1 ppm - If scope shows difference → scope error

Temperature Effects

Characterizing Temperature Coefficient

Same procedure as TinySA:

  1. Cold soak: Refrigerator, 30 min, ~5°C
  2. Measure period at cold temperature
  3. Warm up naturally to room temp (~22°C)
  4. Measure periodically
  5. Heat gently to ~40°C
  6. Plot error vs. temperature

Typical results:

Temp (°C)   Period (s)   Error (ppm)
                5           1.000035     +35
                15          1.000028     +28
                25          1.000025     +25
                35          1.000030     +30
                45          1.000038     +38

Conclusion: Scope timebase drifts with temperature. Calibrate after warmup!

Advanced: 10 MHz Reference Input

Some scopes have external reference input:

If your scope has 10 MHz REF IN: 1. Build GPS-locked 10 MHz reference 2. Connect to REF IN 3. Scope locks to GPS 4. Continuous GPS accuracy!

Building GPS-locked 10 MHz: - PLL locks 10 MHz VCXO to GPS 1PPS - See advanced projects in gps_calibration.md - Cost: $50-100 - Result: Continuous atomic clock lock

Real-World Example

Calibrating DSO1013D

Equipment: - DSO1013D oscilloscope - NEO-6M GPS module (from TinySA project) - 5V USB power supply

Procedure:

  1. Connected GPS 1PPS to CH1

  2. Scope settings: CH1: 1V/div, DC coupling Timebase: 200 ms/div Trigger: CH1, rising, 1.5V level Average: 16

  3. Enabled cursors, measured period: ΔT = 1.000027 s

  4. Calculated error: Error = 1.000027 - 1.000000 = +27 μs PPM = 27/10^6 = +27 ppm

  5. Interpretation:

  6. Scope timebase runs 27 ppm FAST
  7. At 1 second: reads 1.000027 s (27 μs error)
  8. At 1 ms: reads 1.000027 ms (27 ns error)

  9. At 1 MHz: reads 1.000027 MHz (27 Hz error)

  10. Applied correction: Correction factor: 1.000000 / 1.000027 = 0.999973 Multiply all scope time readings by 0.999973

  11. Created label: "TIMEBASE: +27 ppm FAST Correction: × 0.999973 Cal date: 2026-01-02"

Result: Know scope accuracy to atomic clock standards!

Summary

What We Achieved

✓ Measured scope timebase error using GPS ✓ Accuracy limited by GPS: ±0.01 ppm ✓ Documented correction factor ✓ Can now make accurate time measurements

Key Points

  1. GPS 1PPS is perfect reference - atomic clock accuracy
  2. Scope measures period - compare to exact 1.000000 s
  3. Calculate PPM error - quantify timebase accuracy
  4. Apply correction - mental math or label on scope
  5. Verify regularly - monthly check, annual full cal

Next Step

Voltage calibration: oscilloscope_voltage_cal.md

Timebase now calibrated to GPS atomic clock accuracy!