A varicap diode (also known as a varactor or tuning diode) acts as a voltage-controlled capacitor. By varying the reverse-bias DC voltage across its PN junction, you change the width of the depletion region, which directly alters its junction capacitance. For most hobbyist RF, audio filter, and VCO (Voltage-Controlled Oscillator) projects, the BBY58 (SOT-23 surface mount) or the through-hole NTE618 are the safest default part numbers. They offer a reliable tuning range from roughly 2pF to 20pF with a reverse bias between 1V and 20V, and they cost less than $0.50 each in single quantities.

This guide skips the abstract semiconductor physics and goes straight to the bench: how to read the symbol, how to bias the diode without destroying your signal, how to wire a complete VCO, and how to diagnose a failed part with a standard multimeter.

Symbol, Pinout, and Safe Default Part Numbers

On a schematic, the varicap diode symbol looks like a standard rectifier diode (a triangle pointing at a vertical bar), but the cathode bar is drawn with a second, parallel line attached to it. This extra line represents the capacitor plates formed by the PN junction.

Physically, varactors come in standard diode packages. For the common SOT-23 surface-mount package (like the BBY58), the pinout is typically:

  • Pin 1: Anode
  • Pin 2: No Connection (or second anode in dual-diode variants)
  • Pin 3: Cathode (marked by a small dot or band on the silkscreen)

For through-hole glass packages (like the DO-214 or NTE618), the cathode is simply marked with a thick black or colored band on one end of the glass body.

Table 1: Safe Default Varicap Diodes for Hobbyist RF and Audio
Part Number Package Cj at 1V (pF) Cj at 10V (pF) Max Reverse Voltage Typical Price
BBY58 SOT-23 ~18 pF ~4 pF 30V $0.25
MV2101 DO-214 ~15 pF ~5 pF 32V $0.80
NTE618 DO-7 ~22 pF ~6 pF 25V $1.10

Sources: NXP BBY58 Datasheet, Electronics Tutorials - Varactor Diodes.

Operation Regions and Biasing Rules

The core operating principle of a varicap is that capacitance is inversely proportional to the reverse-bias voltage. As you increase the reverse voltage, the depletion region widens, pushing the 'plates' of the capacitor further apart and lowering the capacitance.

Table 2: Typical Operation Regions (Based on BBY58 Characteristic Curve)
Reverse Bias Voltage (V_R) Junction Capacitance (C_j) Typical Use Case
1.0V 18.0 pF Low-end frequency tuning (maximum C)
3.0V 10.5 pF Mid-band VCO centering
10.0V 4.2 pF High-end frequency tuning
20.0V 2.1 pF Maximum reverse swing (minimum C)

The Golden Rule of Biasing

A varicap must always be reverse-biased. If the voltage across the diode swings into forward bias (typically > 0.4V for silicon), the diode begins to conduct heavily. This destroys the high-Q (quality factor) of your LC tank circuit, introduces massive harmonic distortion, and can overheat the junction.

Bench Tip: The AC Swing Constraint
Your DC reverse bias voltage must be strictly greater than the peak-to-peak AC voltage of the RF signal in your circuit. If your tank circuit has a 4V peak-to-peak RF swing (±2V), your minimum DC bias must be at least 2.5V to 3V. If you bias it at 1V, the negative half-cycle of the RF signal will push the diode into forward conduction, clipping your waveform and generating spurious emissions.

Application Circuit: VHF Voltage-Controlled Oscillator (VCO)

Let us build a practical VHF VCO tuning network. This circuit isolates the RF tank from the DC tuning supply while allowing the DC voltage to control the varactor. This is a standard sub-circuit used in FM transmitters, PLL synthesizers, and sweep generators.

Component List and Values

  • D1 (Varicap): BBY58 (Cathode connected to the tank, Anode to ground via RF choke)
  • L1 (Tank Inductor): 120nH (Air-core, 5 turns of 22 AWG enameled copper, 6mm ID)
  • C1 (DC Blocking Capacitor): 47pF (NP0/C0G ceramic, 50V rating)
  • R1 (Bias Isolation Resistor): 10kΩ (0805 SMD or 1/4W through-hole)
  • V_tune: 0V to 15V DC variable supply (filtered with a 100nF bypass cap to ground)

Circuit Netlist / Wiring Sequence

  1. Connect the Cathode of D1 to the RF tank node (where L1 meets the active device's collector/drain).
  2. Connect C1 (47pF) in series between the tank node and the rest of the RF load. This prevents the DC tuning voltage from being shorted to ground by the load, and stops load impedance from detuning the VCO.
  3. Connect the Anode of D1 to one end of R1 (10kΩ).
  4. Connect the other end of R1 to the V_tune DC voltage source.
  5. Place a 100nF ceramic bypass capacitor directly from the V_tune node to ground to shunt any RF noise riding on your DC supply line.
Why 10kΩ for R1? The resistor must be high enough to prevent the RF signal from leaking into the DC power supply (which would cause frequency modulation from power rail noise), but low enough to supply the tiny reverse-leakage current (typically < 20nA) without creating a significant voltage drop. 10kΩ to 47kΩ is the standard sweet spot for VHF applications.

Bench Scenario: Tuning a 100MHz FM Exciter

Theory is clean; the workbench is messy. Here is a real-world scenario demonstrating what happens when you ignore the AC swing constraint.

The Setup

I was building a low-power, 100MHz FM exciter for a local community radio project. The goal was to sweep the full 88MHz to 108MHz commercial broadcast band using a varactor-tuned Colpitts oscillator. Based on the resonant frequency formula f = 1 / (2π√LC), I selected a 56nH tank inductor. To hit 88MHz, I needed roughly 58pF of total capacitance; to hit 108MHz, I needed 39pF.

I chose an MV2101 varactor, which offered a 15pF to 5pF swing. I padded it with fixed 40pF of NP0 ceramic capacitors in parallel to hit my target range. The math checked out perfectly.

The Numbers and The Outcome

I applied a 0V to 12V tuning voltage via a 10-turn potentiometer. The frequency swept beautifully from 87.5MHz to 109MHz on my spectrum analyzer. However, when I injected a 1kHz test tone into the audio input, the spectrum analyzer lit up with massive harmonic sidebands. The audio on my receiver was horribly distorted, specifically on the bass notes and negative peaks of the audio waveform.

What Went Wrong

The oscillator was generating a robust 6V peak-to-peak RF swing across the tank circuit (±3V from center). To tune the oscillator down to 88MHz, my DC bias voltage was set to just 1.5V to maximize the varactor's capacitance.

During the negative half-cycle of the RF oscillation, the instantaneous voltage at the cathode dropped by 3V. Because the DC bias was only 1.5V, the net voltage across the diode swung to +1.5V (forward bias). The varactor turned into a standard rectifier for a fraction of the RF cycle, clipping the sine wave. This clipping generated the 3rd, 5th, and 7th harmonics I was seeing on the analyzer, and the audio clipping caused the severe distortion.

The Fix

I could not simply increase the DC bias, because that would lower the capacitance and push the minimum frequency above 88MHz. Instead, I wired two MV2101 varactors in series, anode-to-anode (a back-to-back configuration). This halved the effective capacitance (requiring me to add more fixed padding caps), but it doubled the AC voltage handling capability. The RF swing was now divided across two junctions, keeping both safely in reverse bias even at the lowest tuning voltages. The harmonics vanished, and the audio was crystal clear.

Failure Modes and Multimeter Testing

Varicap diodes are relatively fragile compared to heavy-duty rectifiers like the 1N4007. Understanding how they fail and how to test them will save you hours of troubleshooting dead VCOs.

How Varactors Fail

  • ESD Punch-Through: At low reverse-bias voltages, the depletion layer is very thin. A static shock can easily punch through this layer, creating a permanent short circuit.
  • Thermal Runaway (Forward Bias): If the AC swing constraint is violated (as in the bench scenario above) and the diode conducts, it dissipates power. Varactors are not rated for forward current; they will overheat and short out internally.
  • Avalanche Breakdown: Applying a DC tuning voltage higher than the rated maximum reverse voltage (e.g., >30V on a BBY58) will cause avalanche breakdown, usually destroying the junction.

Note: Varactors almost always fail shorted, not open. If your VCO is stuck at the lowest possible frequency and won't tune upward, suspect a shorted varactor.

Testing with a Digital Multimeter (DMM)

You cannot measure the picofarad capacitance variance with a standard $20 multimeter, but you can verify the health of the PN junction using the Diode Test mode.

  1. Isolate the Component: Desolder at least one leg of the varactor. In-circuit testing is unreliable because parallel inductors and capacitors will skew the readings.
  2. Set DMM to Diode Mode: Look for the diode symbol on the dial.
  3. Forward Bias Test: Place the Red probe on the Anode and the Black probe on the Cathode. A healthy silicon varactor will read between 0.400V and 0.650V.
  4. Reverse Bias Test: Swap the probes (Black to Anode, Red to Cathode). The meter should read 'OL' (Overload / Open Loop).
  5. Verdict: If the meter reads '0.000V' or beeps continuously in both directions, the junction is shorted (dead). If it reads 'OL' in both directions, the junction is open (dead, though rare).
Proving the 'Varactor' Action
To actually verify that the capacitance changes with voltage, you need an LCR meter. If you do not own one, build a simple 555-timer astable multivibrator circuit on a breadboard. Wire the varactor into the timing network via a 10k resistor, apply a variable 0-12V DC bias, and measure the output frequency with your multimeter's frequency counter or an oscilloscope. If the frequency shifts smoothly as you turn the voltage knob, the varactor is functioning perfectly.

By respecting the reverse-bias constraints, choosing the right part for your voltage range, and isolating your DC tuning lines properly, varicap diodes offer an incredibly elegant, solid-state alternative to bulky mechanical variable capacitors in modern RF and audio design.