开云体育

Since I don't always have access to the nice test equipment from work I decided to come up with something else to generate the 5.34Ghz signal for calibration.
I use a cheap Chinese source that goes to 6Ghz (plus has a doubler to 12Ghz). My only beef is that there is ample of crud coming out of it besides the desired signal.
So I made a crude pipe cap filter as described by Paul Wade W1GHZ. I did not had the exact materials and I wanted connectors so I played around with different items I had at home. The result is a filter that has about 1db loss, is about 450Mhz wide to the 3db points but cuts the 2.5Ghz fundmental and other crud that is anywhere between -10dbm and -40dbm? totally out. I can not see it with my? Avantek analyzer.
One observation I made, since there is no filter in ultra mode there are some ghost signals. A 2.5Ghz -20dbm signal that is clean as a whistle and harmonics less than -80dbc at 5 and 10Ghz will display as a ~ -50dbm signal on 10Ghz. So people do need to be careful using harmonics to test. The signal you are seeing might not really be a harmonic generated by the external device but a signal internal generated in the Tiny.I.M.O. High Pass filters that suppress the fundamental are imperative for valid measurements if someone uses harmonics to test. Anyways, here is a picture of the pipe cap filter. The long interconnect cables are used as an attenuator to bring the signal at the end of the cable to -24dbm for calibration and of course for convenience.

After calibration the Tiny is within 1 db of the test signal. Astounding.