Jerry,
first, thanks. there is substantially your code in there.
second, onto the circuit. it uses two clocks. not three. the third is a spare output. more on that later.
the circuit here uses a resistive return loss bridge. the clock 1 drives the bridge through the R22 to a low level of -10dbm. If the bridge is perfectly balanced (that is, the antenna, R21, R29, R16, all the four are the same ohms), then, there will be no RF developed across pins 3 and 4 of the ADE mixer. Under ideal match conditions, there is no RF across the R26. As the mismatch increases, so does the RF across R26.
We could directly detect the voltage across the R26 with a diode detector. This is quite a popular configuration with most of the simple resistive kind of SWR bridges (like the one designed by Dan Tayloe). This simplicity comes at a cost. The problem is that the detector responds to all the RF between the arms. For instance, if another ham down the block starts to transmit, that energy will show up across the R26 and you will get crazy SWR. I had that problem with broadcast FM showing up on my 7 MHz dipole! Even if there was no RFI from elsewhere, harmonics and spurs from your own transmission can show false readings.
Here is an example: a 7 MHz transmitter with a 14 Mhz harmonic that is 20 db down is connected to a 7 MHz dipole. The dipole is perfectly tuned to show 1:1 SWR, hence, it should show no RF across R26. However, as the antenna is reflecting back the 14 MHz energy, the 14 MHz shows up across the R26.
What's the solution to get a clean dip?The solution is to substitute a simple detector like a diode detector with a simple receiver that is tuned exactly to the frequency that you want to measure the antenna at.
So, the ADE-1 mixer, Q2, Q1 together form a very simple superhet receiver with 25 MHz IF andCLK2 as the local oscillator. The RF at the IF is directly detected and converted to db range with the AD8307. This simple configuration makes this a very powerful instrument.
Here are things you can do with it:
1. Switch off the CLK1, now you have a receiver that can very accurately measure RF levels at any specific frequency in db range. For instance, you connected your transmitter with a suitable RF attenuator to P3, you can tune to various harmonics and measure them very accurately. If you inject a two tone signal into an amplifier, you could easily measure the IMD and IIP3.
2. With the CLK1 on, the instrument now measures the return loss. you can measure the SWR of an antenna, S11 parameters of an amplifier, filter, etc.
3. With CLK1 off, CLK 2 on, the CLK2 can now tune to the frequency tuned in by the receiver's LO (CLK0). By connecting a device/filter between P3 and P4, you can sweep it to measure the gain, frequency reponse.
4. As the diode mixer (ADE-1) has harmonic response, a local oscillator at 135 MHz, will also convert a 430 MHz signal into 25 MHz IF (430 - (135 x 3)). This is possible because we are driving the diode mixer with a square wave from the Si5351 and the local oscillator at 135 MHz also has a 405 Mhz harmonic in it. Hence, the range of this instrument extends to UHF.
The ADE-1 mixer is quite similar to the ubitx mixers. You could even use ubitx kind of discrete version of a diode mixer, it doesn't work too well beyond 50 MHz. The pins 4 and 3 of the ADE-1 are the primary winding of the RF-input side transformer. The documentation recommends that we must ground 4, but that is not essential. We need a differential drive between those two pins, that is what the bridge provides anyway.
73, f
jerry,
i built it so i could stuff it inside the cubesat to measure the antenna. an external spectrum analyzer and its cables were upsetting the RF model hence, i needed something that could read the return loss sitting inside the cubesat. then, i borrowed by daughter's DSLR with a monsterous tele lens and sat 100 meters away to read the the LCD display as it swept through the range.
the analyzer was removed once we knew the correct dimensions and the actual payload went inside the bird.
- f