Y-Factor Method
Understanding the Y-Factor Method
The Y-factor method is the most widely used noise figure measurement technique. It requires only a calibrated noise source and a power detector (or noise figure analyzer). The measurement is straightforward: compare the DUT output noise with a known noise input (hot) vs thermal noise (cold).
| Amplifier Type | Noise Figure | Gain | Output Power |
|---|---|---|---|
| LNA | 0.3-2 dB | 10-25 dB | -10 to +10 dBm |
| Driver | 3-8 dB | 10-20 dB | +15 to +25 dBm |
| Power Amp | N/A | 8-15 dB | +30 to +50 dBm |
| Distributed | 3-6 dB | 5-12 dB | +10 to +20 dBm |
Measurement Procedure
- Connect calibrated noise source to DUT input.
- Measure DUT output noise with noise source OFF (cold): P_cold.
- Measure DUT output noise with noise source ON (hot): P_hot.
- Calculate Y = P_hot / P_cold (linear ratio).
- Calculate NF = ENR - 10 log10(Y - 1).
Accuracy Considerations
- ENR accuracy: Noise source ENR uncertainty directly adds to NF uncertainty.
- Mismatch: Source mismatch during hot and cold states introduces error.
- Second-stage correction: The measurement includes noise from equipment after the DUT. Second-stage correction removes this contribution.
Y = P_hot / P_cold (linear ratio)
NF (dB) = ENR (dB) - 10 log10(Y - 1)
Or in noise temperature:
Te = (Th - Y x Tc) / (Y - 1)
NF = 10 log10(1 + Te/290)
Where Th = ENR_linear x T0 + T0
and Tc = T0 = 290K (typically)
Frequently Asked Questions
What is the Y-factor method?
The Y-factor method measures noise figure by comparing DUT output noise with a noise source ON (hot) vs OFF (cold). Y = P_hot/P_cold. Then NF = ENR - 10log(Y-1). It is the standard method used by noise figure analyzers.
What equipment is needed for Y-factor measurement?
A calibrated noise source with known ENR, a noise figure analyzer (or spectrum analyzer/power meter), and appropriate cables and adapters. Modern noise figure analyzers automate the measurement, including second-stage correction.
What limits Y-factor measurement accuracy?
Primary error sources: ENR calibration uncertainty, mismatch between noise source and DUT, instrument noise figure (second-stage contribution), and environmental temperature variation. Combined uncertainty is typically 0.1-0.3 dB for NF < 3 dB.