What is mismatch uncertainty and how does it affect my power measurement accuracy?

Mismatch uncertainty is the uncertainty in a power measurement caused by the impedance mismatch between the source (device under test) and the measurement instrument (power sensor). It is often the largest source of error in RF power measurements: (1) Definition: when two mismatched devices are connected, the actual power transferred differs from the power that would be transferred with a perfect match. The mismatch uncertainty range: MU = ±20 × log10(1 ± Gamma_source × Gamma_load) (in dB). Where Gamma_source and Gamma_load are the reflection coefficients of the source and the power sensor. (2) Example: a signal source with VSWR = 1.5 (Gamma_S = 0.20) and a power sensor with VSWR = 1.15 (Gamma_L = 0.07): MU = ±20 × log10(1 ± 0.20 × 0.07) = ±20 × log10(1 ± 0.014) = ±0.121 dB. This means: the measured power could be up to 0.121 dB higher or lower than the true power, depending on the phase of the reflections. (3) Significance: mismatch uncertainty is: systematic (it is a fixed error for a given source-sensor pair), phase-dependent (the error depends on the phase of Gamma_source × Gamma_load, which is usually unknown), and treated as a worst-case bound (since the phase is unknown, we assume the worst case: maximum constructive and destructive interference). (4) Reduction techniques: use a high-quality power sensor with low VSWR (VSWR < 1.1 provides Gamma < 0.05). Add an attenuator pad between the source and sensor (the pad reduces the effective Gamma of the source as seen by the sensor). Use a calibrated matching pad specific to the source impedance. Perform a full error correction using a VNA measurement of both the source and sensor reflection coefficients (this measures the phase of Gamma, allowing exact correction instead of worst-case bounds).