Lab 1.1 — Scope Probe Compensation
← Course 2 syllabus · Module 1 · Prev: « Lab 0.3 · Next: Lab 1.2 »
Goal
Power on the Siglent SDS1104X-E for the first time and make it trustworthy: attach a 10× passive probe, compensate it against the front-panel probe-comp square wave, and learn the timebase/trigger controls that every later scope lab depends on. An uncompensated probe silently distorts every edge you look at — over- or under-shooting square waves, wrong rise times, wrong amplitudes on fast signals. Getting the probe flat, and setting the channel’s probe ratio to match, is the price of admission to believing anything the scope tells you. For DSP/firmware work this is the instrument you will use to check clocks, PWM, DAC output, and filter step responses, so the habit is non-negotiable.
Recommended reading
- O-S&S Ch. 1 — signals, especially the square wave as a sum of harmonics; a probe’s compensation is really about passing all those harmonics with the same gain. Light read. → Oppenheim, Signals and Systems
- Your Siglent SDS1104X-E quick-start / user manual: the sections on the vertical (V/div, coupling, probe ratio), horizontal (timebase), and trigger (edge, level) controls, plus the probe-compensation output. Read the probe-ratio and trigger sections in full.
- No Course 1 dependency — this is pure bench skill.
Equipment & parts
- Siglent SDS1104X-E oscilloscope + one 10× passive probe (as shipped) + its trimmer adjustment tool (small screwdriver in the probe kit).
- The scope’s front-panel probe-compensation terminal (a small square-wave output, ≈1 kHz, ~3 Vpp) and the adjacent ground tab.
- Nothing else is powered — this lab uses only the scope’s own reference output.
Safety & don’t-break-it
- This lab is low-risk, but form habits now. The probe-comp output is a few volts and current-limited; you will not hurt yourself or the scope with it.
- Match the ground. The probe’s ground clip is tied to the scope chassis and mains earth. On the probe-comp terminal, clip it to the labeled ground tab — never to a node that could be at another potential (a rule that will matter the moment you probe a live circuit).
- Set the channel probe ratio to 10× to match the probe. If the scope thinks the probe is 1× while a 10× probe is attached, every voltage reading is off by 10×. This is a measurement error, not a hardware risk, but it invalidates the whole lab.
- Adjust the trimmer gently. The compensation trimmer is a small variable capacitor near the BNC or in the probe body; turn it slowly with the plastic tool, not a metal screwdriver you jam in.
- Do not connect the 10× probe to a high-voltage source expecting 10× headroom until you have read the probe’s voltage rating — this lab never leaves the probe-comp terminal, so it is safe here.
Background
A 10× passive probe is a deliberate frequency-compensated attenuator. The probe puts a 9 MΩ resistor in series ahead of the scope’s 1 MΩ input, giving a 10:1 DC divider. But the scope input also has capacitance \(C_\text{in}\) (input C plus cable C, tens of pF), and the probe tip has a small adjustable capacitor \(C_\text{p}\) across its 9 MΩ. Without \(C_\text{p}\), the divider is purely resistive at DC but capacitive at high frequency, so the division ratio changes with frequency and edges get distorted.
The probe is compensated when the RC time constants of the two legs match:
\[R_\text{p} C_\text{p} = R_\text{in} C_\text{in},\]
with \(R_\text{p} = 9\ \text{M}\Omega\) and \(R_\text{in} = 1\ \text{M}\Omega\). When this holds, the 10:1 ratio is flat from DC through the probe’s bandwidth, and a square wave passes through as a clean square. When \(C_\text{p}\) is too small the tops sag/round (under-compensated); too large and the edges overshoot/peak (over-compensated). The probe-comp square wave is the standard test signal because a square wave is rich in harmonics — a flat top confirms uniform gain across frequency.
\[\text{under-comp: rounded tops} \quad\longleftrightarrow\quad \text{correct: flat} \quad\longleftrightarrow\quad \text{over-comp: peaked tops}\]
Procedure
Part A — First power-on and self-cal.
- Power on the SDS1104X-E with nothing connected. Let it boot to the live grid display.
- (Optional but recommended for a first power-on) run the built-in self-calibration from the Utility menu with all probes disconnected, and let it finish. This zeroes internal offsets.
Part B — Attach the probe and set the ratio.
- Connect the 10× probe BNC to CH1. Set the probe’s own slide switch (if present) to 10×.
- Open the CH1 channel menu (press the CH1 button). Find Probe and set the attenuation to 10×. Confirm the on-screen V/div now reflects the 10× scaling.
- Turn CH1 on, set coupling to DC, and set V/div to about 1 V/div.
Part C — Hook onto the probe-comp signal.
- Clip the probe tip to the probe-compensation terminal and the ground clip to the adjacent ground tab.
- Set the timebase to about 500 µs/div (a 1 kHz square has a 1 ms period, ~2 divisions).
- Press Auto Setup if you want the scope to frame it quickly, then fine-tune: you should see a square wave, roughly 3 Vpp, a couple of cycles across the screen.
Part D — Set the trigger.
- Set Trigger type to Edge, source CH1, slope rising. Adjust the trigger level knob to about mid-amplitude until the waveform is stationary (“Trig’d” indicator). A stable trace is required to judge the tops.
Part E — Compensate the probe.
- Look at the top and bottom of the square wave. If the corners overshoot (a spike then settle), the probe is over-compensated; if they round off / sag, it is under-compensated.
- With the plastic trimmer tool, slowly turn the probe’s compensation adjustment until the tops are flat — square corners, no overshoot, no rounding. Nudge back and forth to bracket the flat setting.
- Bump the timebase faster (e.g. 100 µs/div) to inspect the leading edge closely, then back to 500 µs/div to confirm the flat top. Re-check after any adjustment.
Deliverable & expected results
Three captures (screenshots or Siglent CSV/PNG to USB) at the same settings, labeled:
- Under-compensated (deliberately mis-set: rounded tops),
- Over-compensated (deliberately mis-set: peaked/overshooting tops),
- Correct (flat tops).
Plus the measured amplitude and period of the comp signal once compensated.
| Quantity | Predicted | Measured |
|---|---|---|
| Comp-signal frequency | ≈ 1 kHz (≈ 1.0 ms period) | … |
| Comp-signal amplitude | ≈ 3 Vpp | … |
| Probe ratio set in CH1 menu | 10× | … |
| Top of square when correct | flat (no over/undershoot) | … |
Analysis & reconciliation
The measured period should be within a percent or two of 1 ms; the amplitude depends on your unit’s comp output (commonly ~3 Vpp) and on the probe ratio being correctly set to 10× — if you read ~0.3 Vpp you left the channel at 1×, and if you read ~30 Vpp the probe switch and menu disagree. Explain in one line why the shape, not just the amplitude, is the point: an uncompensated probe would still show ~3 Vpp but with distorted edges, which would corrupt every rise-time and overshoot measurement in later labs. Note which way you had to turn the trimmer and relate it back to \(R_\text{p}C_\text{p} = R_\text{in}C_\text{in}\).
Going further
- Repeat the compensation on CH2 with a second probe so both channels are trusted for the two-channel labs (Lab 1.3 uses both).
- Compare the same probe-comp signal with the probe set to 1× (if switchable): note the reduced bandwidth and loading, and why 10× is the default for anything with fast edges.
- Use the scope’s automatic measurements (Vpp, Freq, Rise) on the comp signal now as a warm-up for Lab 1.2, where they carry the measurement.