Skip to content
View in the app

A better way to browse. Learn more.

Ham Radio Base -Powered By Ham CQ DX

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.
Solar
SFI 128
SN 113
A 18
K 2 Quiet
X-Ray C1.2
Wind 554.7 km/s
Aurora 3
Updated 22:30 UTC HamQSL · N0NBH
Day 80/40m Fair 30/20m Good 17/15m Good 12/10m Fair
Night 80/40m Good 30/20m Good 17/15m Good 12/10m Poor

Callsign Lookup
_
Vanity Call Signs Available
Enter filters above and click Search.
ⓘ Callsign lookups are in real time via the FCC database. Vanity callsign availability is refreshed daily at 6:00 AM CST. The vanity search may be unavailable for a few minutes during this update.
Live DX spots
Live DX Spots — 70cm via PSKReporter · scroll or pinch to zoom
Band
Mode
Time
Loading map data…
MHz DX Spotter Info
Recent spots
Select a band above to load spots
Ready — select a band to fetch live spots

Neutralization and Parasitic Suppression

An amplifier that oscillates is not an amplifier — it is a transmitter. And a transmitter that oscillates on an unwanted frequency is a source of interference that can disrupt other operators, attract regulatory attention, and in the case of a power amplifier, destroy itself in seconds when the oscillation drives it out of its safe operating area. Parasitic oscillation and RF instability are among the most common problems encountered when building or troubleshooting RF amplifiers for ham radio.

The root cause of RF amplifier instability is almost always the same: internal feedback from output to input through capacitances inside the transistor itself. In a BJT, the base-collector capacitance Cbc (also called Ccb or Cre) couples output signal back to the input. In a MOSFET, the gate-drain capacitance Cgd does the same. Because the common-emitter (or common-source) amplifier inverts the signal, this feedback is 180° out of phase from the output — and 180° phase shift through the feedback path combined with the amplifier's own inversion makes the total loop phase shift 360°, potentially satisfying the Barkhausen criterion for oscillation.

Understanding how this feedback causes instability — and how to prevent it through circuit design, neutralization, and parasitic suppression — is essential knowledge for anyone building RF circuits above a few megahertz.

What you will learn: Internal feedback capacitance and the Miller effect; how feedback capacitance causes instability and oscillation; neutralization techniques (Hazeltine, Rice, bridge methods); the Rollett stability factor K; parasitic oscillation — causes, symptoms, and diagnosis; parasitic suppression techniques (ferrite beads, resistors, layout); and how common-base and cascode topologies avoid instability.
Circuit diagrams showing: left, a common-emitter amplifier with Cbc feedback path causing instability; centre, Hazeltine neutralization with external capacitor Cn cancelling feedback; right, Rice neutralization using the output coil centre-tap
Three related circuits: the unstable common-emitter with Cbc feedback (left); Hazeltine neutralization using an external capacitor Cn to cancel Cbc (centre); Rice neutralization using the centre-tap of the output coil (right).

Internal Feedback Capacitance — The Root Cause

Every transistor contains internal capacitances between its terminals. The most important for RF stability is the reverse feedback capacitance:

  • BJT: Cbc (base-to-collector capacitance, also listed as Crb or Cre in datasheets, typically 1–10 pF for RF transistors)
  • JFET: Cgd (gate-to-drain capacitance, typically 1–5 pF)
  • MOSFET: Cgd (gate-to-drain, also called the Miller capacitance, typically 10–200 pF for power MOSFETs, 1–5 pF for RF MOSFETs)

These capacitances are parasitic elements that come from the physics of semiconductor junctions and the physical proximity of the transistor's internal structures. They are not listed on component symbols but they are always present, and at radio frequencies where capacitive reactance XC = 1/(2πfC) is small, they provide a significant feedback path from output to input.

For a 2N3904 BJT with Cbc = 4 pF at 7 MHz:

XCbc = 1 / (2π × 7×10⁶ × 4×10⁻¹²) = 5684 Ω

This might seem large compared to the transistor's other impedances. But with a collector load of 2.7 kΩ and a base input impedance of perhaps 1 kΩ, a 5684 Ω feedback path is not negligible at all. At 70 MHz, XCbc drops to 568 Ω — comparable to circuit impedances and a serious feedback problem.

The Miller Effect

In a common-emitter amplifier with voltage gain Av, the feedback capacitance Cbc appears much larger than its actual physical value as seen from the input. This is the Miller effect: the input sees Cbc multiplied by (1 + |Av|).

CMiller = Cbc × (1 + |Av|)

Example: Cbc = 4 pF, Av = −50 (gain of 50)
C_Miller = 4 × (1 + 50) = 4 × 51 = 204 pF

A 4 pF internal capacitor appears as a 204 pF capacitor at the base input. This dramatically lowers the input impedance at high frequencies, reducing gain and potentially causing the input matching network to resonate at an unexpected frequency. More seriously, the Miller capacitance interacts with the source impedance to create frequency-dependent phase shifts that can turn negative feedback into positive feedback at some frequencies — leading to oscillation.

The Miller effect explains several important practical observations:

  • High-gain common-emitter amplifiers have lower bandwidth than low-gain stages
  • Higher supply voltage (which increases gain) also increases Miller capacitance and reduces stability
  • Common-base and common-gate amplifiers do not suffer from Miller effect (output and input are different terminals, with no feedback capacitance multiplication)
  • The cascode amplifier (common-emitter followed immediately by common-base) has very low Miller capacitance because the CE stage has very low gain (the CB stage is the load)

How Feedback Causes Instability

The Barkhausen criterion for oscillation requires two conditions to be met simultaneously:

  1. The loop gain must be ≥ 1 (the signal must be amplified around the feedback loop)
  2. The total phase shift around the loop must be exactly 0° (or 360°) — positive feedback

In a common-emitter amplifier, the transistor itself provides 180° of phase shift (phase inversion). For oscillation to occur, the feedback path (Cbc + source impedance network) must provide another 180° of phase shift. At DC and low frequencies, Cbc is effectively open and provides no feedback. At high frequencies, Cbc's reactance decreases and allows significant signal to flow from collector to base — and depending on the source impedance's phase relationship, this feedback can be regenerative (positive) at certain frequencies.

The frequency at which instability can occur is approximately where the product of the transistor's gain and the feedback fraction through Cbc equals or exceeds unity — the transistor's maximum oscillation frequency f_max. For an RF transistor at frequencies well below f_max, neutralization is required to prevent oscillation. At frequencies approaching f_max, even a perfectly neutralized amplifier will struggle to remain stable.

Neutralization — Cancelling the Feedback

Neutralization is the technique of connecting an external capacitor (or inductor) from the output to the input of the amplifier in such a way that the feedback through this external path is exactly equal in magnitude and 180° opposite in phase to the feedback through Cbc — so the two cancel.

Hazeltine Neutralization

In the Hazeltine method (the most common for solid-state HF amplifiers), a small capacitor Cn is connected from the collector to the base, but through a phase-inverting tap on the output circuit. The output transformer or LC circuit provides an antiphase voltage at the tap; this antiphase voltage, coupled through Cn, cancels the in-phase feedback through Cbc.

Neutralization condition: Cn = Cbc
(Assuming the neutralizing voltage is equal in magnitude and 180° out of phase with the collector voltage — achieved by taking the signal from the other half of a centre-tapped coil or from a phase-inverting tap)

In practice, Cn is made adjustable (a trimmer capacitor) and set to the value that minimises the amplifier's output when the input is driven at the operating frequency with zero supply voltage on the transistor — the "zero-bias neutralisation" procedure. When properly adjusted, the amplifier is stable at all normal operating levels.

Rice Neutralization

Rice neutralization uses the centre-tap of the output inductor to derive the antiphase feedback voltage without a separate phasing winding. It is simpler in construction and is commonly used in single-ended vacuum-tube HF power amplifiers, though less common in solid-state designs where push-pull topology is more typical.

Bridge Neutralization

In push-pull amplifiers, self-neutralization is achieved by the symmetry of the circuit: Cbc of Q1 feeds back to Q2's base and vice versa, and the two feedback paths cancel in the balanced circuit. This is why neutralization is rarely needed in push-pull RF amplifiers — the balanced topology provides inherent neutralization as long as the circuit is truly balanced and the transistors are matched.

The Rollett Stability Factor K

For a more rigorous assessment of amplifier stability, particularly at VHF/UHF and microwave frequencies, the Rollett stability factor K is used. K is calculated from the transistor's S-parameters (scattering parameters measured on a network analyser) and indicates whether the amplifier is unconditionally stable or only conditionally stable:

K = (1 − |S11|² − |S22|² + |Δ|²) / (2 × |S12| × |S21|)

where Δ = S11×S22 − S12×S21

Unconditionally stable (stable for any passive source and load): K > 1 AND |Δ| < 1
Conditionally stable (stable only for certain source/load impedances): K < 1

An amplifier with K < 1 can oscillate if the source or load impedance is outside a certain region of the Smith chart. Adding resistive loading (at the input or output) reduces gain but increases K, making the amplifier unconditionally stable — this is why a small resistor (10–50 Ω) in series with the base or gate is often the first fix for an oscillating RF amplifier.

Four parasitic suppression techniques illustrated: ferrite bead on transistor base lead, resistor in base lead, short direct PCB traces, and decoupling capacitor directly at supply pin
Common parasitic suppression techniques. Short leads and direct grounding (bottom right) are the most fundamental — no suppression component can overcome a circuit built with long inductively-coupled leads.

Parasitic Oscillations

A parasitic oscillation is an unwanted oscillation in an amplifier that occurs at a frequency determined not by the intended circuit but by stray inductances and capacitances in the wiring, PCB traces, and transistor package. Parasitic oscillations can occur at:

  • VHF/UHF frequencies (most common): The transistor's internal capacitances (Cbc, Cce) resonate with the stray inductances of the circuit board traces, bond wires, and component leads. A 2 nH trace inductance and a 10 pF capacitance resonate at 2π×f = 1/√(LC), giving f = 1/(2π×√(2×10⁻⁹ × 10×10⁻¹²)) ≈ 1.1 GHz. This means a transistor intended for HF use can oscillate at gigahertz frequencies if the circuit is not carefully laid out.
  • Audio/low frequencies: The supply decoupling network (electrolytic capacitors and PCB trace inductances) can form a resonant circuit at audio frequencies (50 Hz–10 kHz). RF power can be converted to audio-frequency oscillation in poorly decoupled stages. This manifests as motorboating — a low-frequency putt-putt sound in the audio output, caused by audio-frequency feedback through the shared supply rail.
  • At the operating frequency: Under some bias conditions or load mismatches, the amplifier itself can oscillate at or near its intended operating frequency. This often occurs in Class C power stages when the tank circuit resonant frequency is slightly wrong, or in Class AB stages that are overdriven.

Symptoms of Parasitic Oscillation

  • Unexpectedly high DC current draw with no RF drive (transistor heating without input signal)
  • RF output on a frequency different from the intended operating frequency
  • Instability or jumping of output level when touching nearby components
  • Spurious outputs visible on a spectrum analyser at VHF/UHF while operating on HF
  • Transistor failure (immediate or progressive) with no apparent cause

Parasitic Suppression Techniques

1. Ferrite Beads

A ferrite bead threaded onto the base or gate lead of the transistor adds a few ohms of resistance and a few nanohenries of inductance at VHF/UHF frequencies, without significantly affecting the HF operating frequency. The ferrite's loss at high frequencies damps any VHF/UHF parasitic oscillation tendency. One ferrite bead (Fair-Rite type 43 or 61 material, or a Würth WE-CBF series bead) on each transistor lead is often sufficient to cure a marginal parasitic oscillation problem.

2. Base/Gate Resistors

A small resistance (10–100 Ω) in series with the base or gate lead adds resistive loss to the feedback loop that tends to cause VHF parasitic oscillation. At the HF operating frequency, a 47 Ω resistor in series with a 1 kΩ base impedance is a 4.7% attenuation — insignificant. But at 1 GHz where the transistor's internal capacitances have become low impedance, the 47 Ω resistor provides much more effective damping. This technique is used in virtually every well-designed RF power amplifier — you will find a small chip resistor (47–100 Ω) right at the gate of RF MOSFET finals in commercial transceivers.

3. Short, Direct Leads and Good Grounding

Stray lead inductance is the primary driver of parasitic resonance. Every 1 cm of wire has approximately 10 nH of inductance. The shorter the leads to the transistor, the higher the parasitic resonance frequency — and at frequencies far above the transistor's gain bandwidth, oscillation cannot be sustained. Surface-mount components have almost zero lead inductance compared to through-hole parts, which is a major reason why modern RF power amplifiers use SMD construction exclusively above 30 MHz.

4. Supply Decoupling

Every RF amplifier stage requires RF bypass capacitors as close as possible to the transistor's supply pin. Without adequate decoupling, RF current flows through the shared supply impedance, coupling energy between stages and creating the conditions for oscillation. Use a combination of a large electrolytic (10–100 μF for low-frequency bypass) and a small ceramic (100 pF–10 nF for RF bypass), placed within a few millimetres of the supply pin. The bypass capacitors must also have low inductance — a through-hole electrolytic is not an RF bypass capacitor on its own.

5. Input Resistive Loading

Adding a resistor (50–200 Ω) from the transistor's base or gate to ground provides a fixed, real (non-reactive) load at the input that tends to damp oscillation tendencies. It also helps set the input impedance to a known value (improving impedance matching). The cost is reduced gain (the input voltage is divided between the transistor's input impedance and the shunt resistor). This technique is widely used in MMIC (monolithic microwave integrated circuit) amplifiers, where a shunt resistor provides broadband impedance matching simultaneously with stability improvement.

Common-Base and Cascode — Inherently Stable

The common-base (BJT) and common-gate (FET) configurations do not suffer from the Miller effect because the input (emitter/source) and output (collector/drain) terminals are on opposite sides of the transistor from the base/gate, which is AC-grounded. The feedback capacitance from output to input in common-base is the emitter-collector capacitance Cce (BJT) or the drain-source capacitance Cds (FET), which is typically much smaller than Cbc or Cgd, and it is not multiplied by the gain of the stage.

The cascode amplifier (common-emitter/source followed immediately by common-base/gate) combines the high gain of the CE/CS stage with the low Miller effect of the CB/CG stage. The CE stage drives the emitter of the CB stage directly. The CB transistor's base is AC-grounded (bypassed to ground). The CE stage sees the CB transistor's input impedance (≈ 1/gm of the CB transistor, typically 20–50 Ω) as its collector load — a very low impedance. Since the CE stage has very low voltage gain (Av ≈ −RC/re where RC is the CB's input impedance), its Miller capacitance is negligible. The high gain comes from the CB stage, which does not suffer from Miller multiplication.

Cascode amplifiers are used extensively in HF and VHF receivers where low Miller capacitance is needed for high bandwidth and good stability, while maintaining useful gain — particularly in mixer-preamplifier combinations and in the front end of wideband receivers.

Miller Capacitance and Neutralisation Calculator

Miller Capacitance and Neutralisation Calculator

Calculate the effective Miller input capacitance for a common-emitter amplifier, and find the neutralisation capacitor value needed to cancel it. Also estimates the frequency where Cbc feedback becomes significant.

Results will appear here.

Experiment: Detect and Suppress a Parasitic Oscillation

Purpose: Experience how a VHF parasitic oscillation feels and find the cure using basic test equipment.

You will need: A simple one- or two-transistor RF amplifier circuit (e.g. a 7 MHz preamplifier built on a PCB), a multimeter, an RF probe (or a simple diode detector), and optionally a spectrum analyser app on a smartphone with an HF-capable SDR receiver.

Creating the Parasitic (for demonstration only):

  1. Build the amplifier on a PCB with intentionally long transistor leads (5–10 mm lead length rather than trimmed short).
  2. Remove any ferrite beads from the transistor leads.
  3. Power up the amplifier with NO RF input signal applied.
  4. Measure the DC supply current. Note the value.
  5. If the circuit is oscillating parasitically, the supply current will be significantly higher than the quiescent current calculated from the bias resistors (typically 2–5× higher). The transistor will also be warm to the touch with no input signal.

Detecting the Oscillation:

  1. Touch a short (5 cm) insulated wire to each transistor lead and note whether the current changes — this disturbs the stray capacitances and may start, stop, or shift a parasitic oscillation.
  2. Hold the RF probe (or SDR antenna) near the circuit and listen for signals on unexpected frequencies. Parasitics often appear at VHF (50–200 MHz) or UHF even in HF amplifiers.

Suppression procedure:

  1. Trim all transistor leads short — within 2–3 mm of the PCB surface. Measure current; it should drop toward the quiescent value.
  2. Add one ferrite bead (type 43, size 0402 or 0603) to the base lead. Check current again.
  3. Add a 47 Ω chip resistor in series with the base lead. Check current.
  4. Add a 100 nF ceramic decoupling capacitor directly across the supply pins at the transistor, as close as possible to the transistor body. Check current.

Expected results: Each successive step should reduce supply current closer to the quiescent value. When all steps are complete, the circuit should draw correct quiescent current with no input, and gain should be maximum at 7 MHz. The SDR receiver should show no spurious signals from the circuit.

Frequently Asked Questions

Why does a parasitic oscillation draw more current than normal quiescent operation?

When a transistor oscillates — even on an unintended parasitic frequency — it is doing work: it is converting DC power from the supply into RF power. The oscillating signal drives the transistor through its full or partial operating range, just like a real input signal. This requires DC current from the supply, which is why the supply current rises significantly during a parasitic oscillation. In a Class C or class C-like parasitic, the transistor may be driven into and out of saturation on each cycle, which can be very efficient at converting DC to parasitic RF — sometimes so efficient that the transistor's average power dissipation increases dramatically, causing rapid heating. This is one of the main failure mechanisms for RF transistors in poorly designed or unshielded amplifiers.

Why does neutralization need to be readjusted when changing frequency?

Neutralization balances the feedback through Cbc against the feedback through Cn. For exact cancellation, the magnitude and phase of these two paths must match at the operating frequency. If the output circuit (tank circuit or transformer) has a frequency-dependent phase shift, the phase of the antiphase neutralizing voltage changes with frequency. A neutralizing capacitor set for one frequency may provide insufficient or excessive cancellation at another frequency. This is why single-band amplifiers are easier to neutralize than broadband designs, and why broadband amplifiers often use circuit topologies (cascode, common-base) that inherently minimise the feedback needing cancellation.

What is the difference between VHF parasitic oscillation and the intended operating frequency becoming unstable?

A VHF parasitic oscillation occurs at a frequency entirely outside the intended operating range, determined by the transistor's internal capacitances resonating with stray lead inductances — often hundreds of megahertz above the intended HF operating frequency. It is often present continuously (not triggered by input signal) and does not disappear when the tank circuit is detuned from the operating frequency. Instability at the operating frequency is different: it occurs when the loop gain at the operating frequency exceeds unity with the correct phase, and often manifests as gain peaking, ringing, or gain jumping. Both can be present simultaneously — a circuit can have a VHF parasitic and also be on the edge of instability at its operating frequency.

Does push-pull topology eliminate the need for neutralization?

Push-pull topology provides self-neutralization if the circuit is balanced: the feedback through Cbc of Q1 goes to Q2's base circuit, and the feedback through Cbc of Q2 goes to Q1's base circuit, with opposite phase. In a perfectly balanced circuit these two feedbacks cancel and the stage is inherently neutralized. In practice, the balance is never perfect — transistors are not perfectly matched, the transformer has slight imbalance — so some residual feedback remains. Push-pull stages operating below 30 MHz rarely need additional neutralization because the residual imbalance is small enough. At VHF and above, even small imbalances become significant and additional suppression (ferrite beads, resistors, careful layout) is needed.

What is motorboating and how is it cured?

Motorboating is a low-frequency oscillation (typically 1–100 Hz) in a multi-stage amplifier, named for the rhythmic putt-putt sound it produces in the audio output. It is caused by positive feedback through the shared power supply: output stage RF current flowing through the supply rail creates a voltage drop; this modulates the supply voltage of earlier stages; the earlier stages amplify the variation; it appears at the output; and the cycle repeats. The cure is adequate low-frequency supply decoupling at every stage — a large electrolytic capacitor (100–1000 μF) at the output stage and either individual regulation for each stage or a resistor-capacitor (RC) decoupling network (typically 47 Ω + 100 μF) between stages to present low supply impedance at all audio frequencies.

Check Your Understanding

Answer each question and click Check to see the explanation.

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.