Bias Network
Understanding Bias Networks
Every active RF device needs DC bias to operate. The challenge is delivering DC power without disturbing the RF signal path. A poorly designed bias network can cause oscillation, degrade noise figure, or create unexpected frequency response problems.
Bias Network Components
- RF choke: High impedance at RF, low impedance at DC. Blocks RF from entering DC supply.
- DC block: Capacitor in series with RF path. Blocks DC, passes RF.
- Bypass/decoupling capacitors: Short RF to ground near the active device. Multiple values for broadband decoupling.
- Quarter-wave stub: At microwave frequencies, a quarter-wave shorted stub provides a high-impedance RF block.
Bias Network Design Rules
- Place decoupling capacitors as close to the device as possible.
- Use multiple bypass capacitor values (100 pF, 1 nF, 10 nF, 100 nF) for broadband decoupling.
- Check stability with the bias network connected (it adds impedance at low frequencies).
- Sequencing: many GaN devices require specific gate-before-drain power-up sequencing.
Frequently Asked Questions
What is a bias network?
A bias network delivers DC operating power to active RF devices while isolating the DC from the RF signal path. It uses RF chokes, DC blocking capacitors, and decoupling capacitors. Proper design prevents oscillation and ensures optimal performance.
Why can a bias network cause oscillation?
The bias network adds impedance in the supply path that varies with frequency. At low frequencies where the decoupling is insufficient, this impedance can create a resonance that provides positive feedback, causing the amplifier to oscillate. Multiple decoupling capacitor values mitigate this.
What is bias sequencing?
Some devices (especially GaN HEMTs) require specific power-up order: gate voltage applied before drain voltage. Applying drain first can cause destructive current surges. Bias sequencing circuits ensure safe and repeatable power-up.