Objective and use case
What you’ll build: This guide demonstrates how to effectively reduce supply ripple at a load by adding a decoupling capacitor near it. You’ll set up a simple circuit using a 2N2222 transistor and a 100 nF ceramic capacitor.
Why it matters / Use cases
- Improving the stability of power supplies in microcontroller applications by minimizing voltage fluctuations during load changes.
- Enhancing the performance of RF circuits by reducing noise introduced by power supply variations.
- Ensuring reliable operation of sensors powered by unstable sources, such as battery-operated devices.
- Facilitating the design of audio equipment where power supply noise can lead to audible hum and distortion.
Expected outcome
- Reduction of supply ripple measured in millivolts (mV) on the oscilloscope.
- Improved voltage stability at the load, with latencies reduced to less than 10 ms during load transitions.
- Clear square wave output from the function generator with minimal distortion, indicating effective decoupling.
- Ability to observe a stable VCC_L on the oscilloscope, confirming the capacitor’s effectiveness.
Audience: Electronics enthusiasts; Level: Basic
Architecture/flow: Circuit includes a DC power supply, NPN transistor, resistors, and a decoupling capacitor connected on a breadboard.
Materials
- 1 DC power supply, +5 V (bench supply or USB 5 V source)
- 1 NPN transistor, 2N2222 (or 2N3904) — Q1
- 1 Resistor, 22 Ω — R1 (series supply impedance)
- 1 Resistor, 100 Ω — R2 (load to ground)
- 1 Resistor, 10 kΩ — R3 (base resistor)
- 1 Capacitor, 100 nF ceramic (X7R preferred) — C1 (decoupling)
- 1 Function generator (square wave 0–5 V, 1 kHz)
- 1 Oscilloscope (2 channels preferred), probes and ground clips
- 1 Solderless breadboard and jumper wires
Wiring guide
- Create a common ground rail on the breadboard for the DC supply, function generator ground, and all component grounds.
- Insert R1 between the +5 V supply and the “local VCC” node. R1 simulates supply/source impedance.
- Place C1 from the local VCC node to ground. Keep leads short; place C1 as close as possible to the load node.
- Build the pulsed load:
- Q1 collector to the local VCC node.
- Q1 emitter to one side of R2; the other side of R2 to ground.
- Drive Q1 base through R3: function generator OUT → R3 → Q1 base.
- Connect the function generator ground to the common ground.
- Set the function generator to square wave, 0–5 V amplitude, 1 kHz. Start at 50% duty cycle.
- Oscilloscope connections:
- CH1 tip to ● VCC_L; CH1 ground to GND.
- CH2 tip to ● V_SUP (optional comparison); CH2 ground to GND.
- You may also probe ● IN to verify the drive level.
- Abbreviations used:
- V_SUP: supply node before R1 (upstream of the series resistor).
- VCC_L: local VCC at the load (after R1; decoupled by C1).
- IN: base drive signal node (after R3, at Q1 base).
Schematic
+5 V
│
├───────────────┬───────────────────────────────┐
│ │ │
┌┴┐ ┌┴┐ │
│ │ │ │ │
│ │ │ │ │
└┬┘ └┬┘ │
│ │ │
C1 = 100 nF (desacoplo) C2 = 10 µF (bypass) │
│ │ │
GND GND │
│
● TP_VCC │
│ │
┌─────────────┴───┐
│ │
│ NE555 │
│ (astable) │
│ │
GND ─────────────────────────────────────────────────────┤1 GND 8 Vcc├─┘
│ │
+5 V ────────────────────────────────────────────────────┤4 RESET │
│ │
┌───────────┐ │ │
+5 V ───────────────────┤ │ │ │
│ ┌┴┐ R1 = 10 kΩ │ │
│ │ │ │
│ │ │ │
│ └┬┘ │
│ │ │
│ ┌┴┐ R2 = 100 kΩ │
│ │ │ │
│ │ │ │
│ └┬┘ │
│ │ │
└──────────┴───────┐ │
│ │
┌─● TP_RC │
│ │
┌┴┐ C_T = 10 nF │
│ │ │
│ │ │
└┬┘ │
│ │
GND ┌────┴────┐
│ │
│ 3 OUT ├───● TP_OUT───┬───┌┴┐ R_LED = 330 Ω ───┌┴┐ LED ─── GND
│ │ │ │ │ │ │
│ 2 TRIG ├──────────────┘ │ │ │ │
│ 6 THR ├──────────────────┘ └┘ └┬┘
│ 7 DIS ├────────────────────────────────────────┘
│ │
└─────────┘Measurements and tests
- Initial observation (with C1 installed):
- Set the generator to 1 kHz, 0–5 V, 50% duty.
- Observe CH1 at ● VCC_L. You should see small dips/spikes synchronized with IN edges, typically a few millivolts to tens of millivolts.
- Compare CH2 at ● V_SUP. This node should be cleaner than VCC_L or show smaller variations, depending on wiring and R1.
- Effect of removing decoupling:
- Temporarily lift one leg of C1 (or remove it).
- Observe CH1 at ● VCC_L again. The ripple and edge-related droops should increase visibly (often several times larger).
- Reinstall C1 and confirm the ripple reduction.
- Frequency dependence:
- Sweep the square-wave frequency from 500 Hz up to 100 kHz.
- Note that the 100 nF ceramic is most effective for fast edges/high-frequency content; at very low frequencies the voltage droop is dominated by R1 and average current, so consider a larger electrolytic if needed.
- Duty cycle and load step:
- Vary duty cycle from 10% to 90%. Larger average on-time increases droop across R1; C1 chiefly reduces the transient spikes at transitions.
- Abbreviation recap for measurements:
- V_SUP: upstream supply node; place probe at ● V_SUP.
- VCC_L: local decoupled supply node at the load; place probe at ● VCC_L.
- IN: base drive after R3; place probe at ● IN to verify 0–5 V toggling.
How it works
- R1 adds a small series impedance, mimicking real supply wiring/regulator output impedance.
- When Q1 switches on, current flows from the local VCC node through Q1 and R2 to ground, causing a momentary voltage drop across R1.
- C1 provides a low-impedance path to ground at high frequency, supplying instantaneous current to the load and keeping VCC_L steadier during edges.
Common mistakes
- Placing C1 far from the load node. Keep leads short and the capacitor close to the VCC_L node and ground.
- Forgetting the base resistor R3, which can overdrive Q1 and distort results.
- Using long ground leads on scope probes; use a spring ground to reduce measurement-induced ringing.
- Connecting the oscilloscope ground to a non-ground node; always clip to the common GND.
Safety and notes
- Ensure R2 is ≥100 Ω to keep Q1 current reasonable (≈30–50 mA at 5 V minus VCE(sat)).
- Q1 may get slightly warm; if so, reduce duty cycle or increase R2.
- Do not exceed the function generator’s output current; R3 protects the generator and Q1 base.
Improvements
- Add a 10 µF (or 22 µF) electrolytic in parallel with C1 to handle lower-frequency load changes.
- Try different decoupling values and dielectric types (10 nF, 100 nF, 1 µF; X7R vs. NP0) and compare ripple.
- Reduce R1 (e.g., to 10 Ω) to emulate a stiffer supply and observe how the need for local decoupling persists near fast loads.
More Practical Cases on Prometeo.blog
Find this product and/or books on this topic on Amazon
As an Amazon Associate, I earn from qualifying purchases. If you buy through this link, you help keep this project running.



