2010-02-01
AD8307 RF Power Meter
At the latest ARNSW Homebrew Group meeting I had the opportunity to finally calibrate my AD8307-based RF power measurement head.
Mark VK2XOF brought along his power and frequency reference equipment and gave me a bunch of calibration points.
Data points above 13 dBm were given using the generator out of levelling-loop calibration so the slight compression approaching 20 dBm is to be expected. The levelling off below -40 dBm however is not. I know the noise floor of the meter (dummy-load terminated) is about 500 mV DC output, -40 dBm is about 2.65 V out, suggesting wideband noise from the generator was swamping the lower level spot calibration signals.
Regardless in between the data is very consistent and lets me compute a "slope + intercept" calibration relying on the excellent linearity of the AD8307. The 0 dBm intercept is 5.09 Volts and the slope is 16.469 dB per Volt (60.7 mV per dB). Based on the 500 mV noise floor this equates to about -75.6 dBm, or about 80 dB dynamic range - as expected for an AD8307-based unit.
Frequency response wise, my use of leaded resistors means it maintains its accuracy to about 250 MHz, which is sufficient for my immediate purposes. The circuit design itself is the W7ZOI design from EMRFD, page 7.7, figure 7.13.
3 comments.