I hooked up my KX3s IQ port to an EMU0202 sound card and used NaP3 to display the spectrum.
With no preamp, the display floor is fairly flat, as Scott found. With preamp, the floor is humped in the middle, as apparently the bandwidth of the preamp rolls off smoothly at frequencies further removed from the tuned frequency. In any event the display is useful for finding signals +/- 96khz from center frequency. A signal blip can be clicked to tune the KX3. A small IF offset is required to make the displayed spectrum agree with the KX3's dial. I have not determined yet what value of offset is best. Monty K2DLJ In the last two days we've seen a number of questions raised regarding the KX3's I/Q output and its frequency response characteristics. In an effort to provide some information that others may find helpful, this evening I made several measurements of my KX3's I/Q output (early 700 S/N) using my H/P 89441 Vector Signal Analyzer operating in the baseband mode. For my initial tests, an H/P 8640B signal generator provided a CW signal at -70 dBm on 14.100 MHz. The KX3's preamplifier was switched off and the radio was tuned to zero-beat. When tuned to zero beat in the CW mode, the carrier appears at an offset equal to the sidetone frequency (700 Hz in my KX3). The power of the baseband CW tone was measured at 700 Hz with the receiver at zero-beat, then the receiver was tuned to move the received carrier to baseband frequencies of 2000 Hz, 9000 Hz, 25 kHz and 48 kHz. The measurement results are shown below: For -70 dBm CW @ 14.1 MHz: Baseband Freq. I Power Q Power 2000 Hz -76.0 dBm -76.2 dBm 9000 Hz -76.2 dBm -76.2 dBm 25 kHz -76.5 dBm -76.7 dBm 48 kHz -78.5 dBm -78.3 dBm For signals within about 10 kHz of the carrier frequency, the KX3 displays a signal gain to the I/Q port of very close to -6 dB when the preamplifier is disabled. The power balance between the I and Q port is within 0.2 dB at all tested frequencies. Next, instead of a CW tone I injected wideband Gaussian noise from a GR 1383 Noise Generator. The noise power at 14 MHz was set to -90 dBm/Hz. With this input stimulus, the measured noise power on the I or the Q port was -94 dBm/Hz, and the noise displays a 2.5 dB slope from 700 Hz to 50 kHz, just as noted above when CW was the input stimulus. I will spend time looking at some of the other baseband aspects of the KX3's I/Q output as time allows later this week. 73, Scott N7NB ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[email protected] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
