Troubleshooting and Debugging Common RF Problems Diagnostic

How do I troubleshoot a high noise figure measurement that does not match my cascade analysis?

Q: My cascade analysis matches to within 0.5 dB. Is that good enough?

Yes. A 0.5 dB agreement between cascade analysis and measurement is considered excellent. The cascade analysis uses typical (not measured) component noise figures, and the measurement itself has approximately 0.2-0.5 dB uncertainty. Agreement within 1 dB is considered acceptable for most applications. Discrepancies larger than 1.5-2 dB indicate a missing noise contribution or measurement error.

Q: Does cable loss after the LNA matter for noise figure?

Loss after the LNA has a greatly reduced effect on system noise figure, attenuated by the LNA's gain. For example, if the LNA has 20 dB gain, a 1 dB loss after the LNA contributes only 0.01 dB to the system noise figure (1 dB / 100 = 0.01 dB). The first element in the receive chain dominates the noise figure; elements after the first high-gain amplifier have negligible impact.

Q: Can a VNA measure noise figure?

Modern VNAs can measure noise figure using the cold-source method (no noise source required; the DUT's own thermal noise is measured along with its S-parameters). The accuracy is approximately 0.3-0.5 dB, slightly worse than a dedicated noise figure analyzer with a calibrated noise source. The VNA method is convenient because it measures gain and noise figure simultaneously without changing the test setup.

When a measured noise figure is higher than the cascade analysis prediction (Friis formula), the discrepancy is caused by contributions that were not included in the analysis or by measurement errors. The most common overlooked noise contributions are: input cable and connector losses before the LNA (any loss before the first amplifier adds directly to the system noise figure: a 0.5 dB cable loss adds approximately 0.5 dB to the noise figure), LNA noise figure under actual bias conditions (the datasheet NF is measured at optimal bias; actual bias may differ, especially if voltage drops across bias resistors are not accounted for), LNA input mismatch (mismatch between the source impedance and the LNA's optimal noise impedance increases the noise figure; the minimum noise figure is achieved only at the optimal source impedance), image noise in mixers (a mixer folds noise from the image frequency into the IF band, adding 3 dB of noise if no image-reject filter is present), and passive losses between stages (each loss between amplifier stages contributes additional noise). Common measurement errors include: incorrect noise source ENR (Excess Noise Ratio) calibration data, cold source temperature assumption errors (the actual physical temperature may not be exactly 290K), noise source mismatch with the DUT input, and instrument uncertainty (noise figure analyzers have their own uncertainty of 0.1-0.5 dB).

Category: Troubleshooting and Debugging

Updated: April 2026

Product Tie-In: Test Equipment, Components

Diagnosing Noise Figure Measurement Discrepancies

Noise figure discrepancies are among the most common and most frustrating measurement issues in RF engineering. The Friis cascade formula gives a theoretically correct result, but only if ALL noise contributions are accurately accounted for, which is difficult in practice.

Complete Noise Budget Checklist

Pre-LNA losses: Antenna feed loss, cable loss, connector loss, filter insertion loss, switch loss, and any other passive elements before the LNA. Each contributes directly to system NF
LNA noise figure: Verify at actual bias conditions (not datasheet typical). Check that the source impedance matches the LNA's optimum noise impedance (Gamma_opt from the datasheet). Check operating temperature versus datasheet conditions
Image noise: For superheterodyne receivers, noise at the image frequency (f_LO +/- f_IF) folds into the IF band. Without an image-reject filter, this adds 3 dB to the effective mixer noise figure. Include the image noise contribution in the cascade analysis
Inter-stage losses: PCB trace losses, filter losses, and pad attenuators between amplifier stages. Each loss degrades the noise figure by an amount that depends on its position in the cascade
Mixer noise: Mixer noise figure includes both conversion loss and excess noise. Double-balanced mixers have NF approximately equal to conversion loss (7-10 dB typical). Active mixers may have lower NF but check at actual LO power

Measurement Error Sources

The Y-factor method using a calibrated noise source assumes the cold temperature is exactly T_cold = 290K. If the actual ambient temperature is 300K, this introduces approximately 0.15 dB error. Noise source ENR calibration data must be entered correctly (frequency-by-frequency values, not just the nominal ENR). The noise source output impedance mismatch with the DUT input creates measurement error proportional to the mismatch and the DUT's noise sensitivity to source impedance.

Noise Figure Diagnostic Equations

Friis formula: NF_system = NF1 + (NF2-1)/G1 + (NF3-1)/(G1xG2) + ...
Loss before LNA: NF_total = L [dB] + NF_LNA [approximately, for L < 3 dB]
Image noise: NF_mixer_effective = NF_mixer + 3 dB (if no image filter)
Y-factor: NF = ENR / (Y - 1), where Y = P_hot / P_cold
Measurement uncertainty: typically +/- 0.2-0.5 dB for calibrated NF analyzer

Common Questions

Frequently Asked Questions

My cascade analysis matches to within 0.5 dB. Is that good enough?

Yes. A 0.5 dB agreement between cascade analysis and measurement is considered excellent. The cascade analysis uses typical (not measured) component noise figures, and the measurement itself has approximately 0.2-0.5 dB uncertainty. Agreement within 1 dB is considered acceptable for most applications. Discrepancies larger than 1.5-2 dB indicate a missing noise contribution or measurement error.

Does cable loss after the LNA matter for noise figure?

Loss after the LNA has a greatly reduced effect on system noise figure, attenuated by the LNA's gain. For example, if the LNA has 20 dB gain, a 1 dB loss after the LNA contributes only 0.01 dB to the system noise figure (1 dB / 100 = 0.01 dB). The first element in the receive chain dominates the noise figure; elements after the first high-gain amplifier have negligible impact.

Can a VNA measure noise figure?

Modern VNAs can measure noise figure using the cold-source method (no noise source required; the DUT's own thermal noise is measured along with its S-parameters). The accuracy is approximately 0.3-0.5 dB, slightly worse than a dedicated noise figure analyzer with a calibrated noise source. The VNA method is convenient because it measures gain and noise figure simultaneously without changing the test setup.

Keep Reading