How should RF circuits be decoupled?
RF supply decoupling aims to present a low impedance at RF to the device’s supply pins and to block RF and noise from traveling between circuits along supply lines. Typical practice combines a small capacitor with SRF near the operating frequency at the pin, larger capacitors for lower frequencies, and series ferrites or resistors between stages; low-noise LDOs supply oscillators and synthesizers.[1][2]
How are capacitors chosen?
A capacitor is most effective near its series self-resonance; above it, it is inductive. Choose a value whose SRF (including pad and via inductance) falls near the operating band for the closest capacitor, and add larger values for lower frequencies. Watch for anti-resonances between parallel capacitors of different values, which can create impedance peaks.[3]
When do ferrite beads help?
Ferrite beads are lossy above their crossover frequency and isolate supply branches. They saturate with DC current and can resonate with capacitors to create peaks; check impedance-vs-current curves and damp resonances.[1]
Why does supply noise matter for oscillators?
Supply noise modulates VCO frequency (pushing) and appears as phase noise and spurs; LDO output noise density in nV/√Hz across the PLL loop bandwidth and beyond is a key spec for synthesizer supplies.
Frequently asked questions
Is 100 nF enough for RF decoupling?
Usually not at GHz frequencies alone; add small, low-ESL capacitors at the pins.
Where should decoupling capacitors go?
As close to the supply pin as possible, with vias directly to ground.
Can decoupling cause oscillation?
Resonances in supply networks can contribute to instability; damping helps.
See also
References
Online sources accessed October 21, 2026; printed works cited by edition. See all sources.
- “Electromagnetic Compatibility Engineering”. H. W. Ott, Wiley. 2009.
- “RF Circuit Design, 2nd ed.”. C. Bowick, Newnes. 2008.
- “Signal and Power Integrity, Simplified, 3rd ed.”. E. Bogatin, Prentice Hall. 2018.