For using the Thales module as a LO source two sources would be needed in the stock configuration. 11-12GHz LO and something for the TX IF port.
The more I read about other peoples 76GHz transverters, the more I look at my junkbox, the more I begin to see a 76GHz transverter taking form.Proverbial "final nails" to make this project seem viable ware my aquisition of a WR-28 to male 2.4mm adapter, a 2.4mm --> 3.5mm adapter, cheap diodes from Skyworks and M/A-Com. That and recalling that the mixer in the TX chain in the Thales 38GHz transceiver module can be modified in to a DC variable attenuator.
In my tests with the Thales module I used 3.6GHz. Modifying the mixer in to a DC controlled attenuator removes the need for this second LO.
It also removes the need for an attenuator at the output so that the raw 300mW output does not burn mixer diodes.
An additional bonus is that the LO drive can be then easily switched between TX and RX. This is not possible in most traditional transverter designs.
This allows us to have our cake and eat it too. We can optimize for best NF/conversion loss on RX and most output power on TX.
Before realizing that things like connector frequency limits are merely suggestions I was wracking my brain with either designing a waveguide mixer with WR-28 LO input, SMA IF and WR-15 RF port or some slab/DB6NT style deal. Both require more fabrication than I want to do. But I do admit that the slabmixer published in DUBUS is tempting for future experiments, but one still needs a DB6NT mixer PCB for it.
For driving the 38GHz Thales module I'm using one of those Elcom synths with "Wyrm" controller. Initial tests will likely use 70cm FM as the test IF and then transition to 70cm or 2m SSB. Mixer will be an anti-parallel diode at the dish (a shined up empty butane container for initial tests) feedpoint.
Test source is a single shottkydiode at a similar "dish" fed with a 18GHz Mizar synth. With the idea set on expanding the Mizar system in to a similar transverter later on, but with 18GHz f/4 LO drive instead of f/2. I have some old Litton solidstate 18GHz link tranceivers with Gunn LO's but with power amplifiers, WR-42 waveguide output and output attenuator. The 18GHz Gunn LO is easy to remove and fitting a piece of semirigid should be possible. This way we can have plenty of LO drive. I also have commercial WR-42 to 3.5mm adapters so getting the RF in to the semirigid requires less work than with the WR-28 38GHz LO.
VK3DAG used a 26GHz Thales unit on 23Ghz as a 47GHz LO:
VK3XPD made a 76GHz & 122GHz dualband transverter with semirigid fed antiparallel diode mixer at antenna feedpoint. With 39Ghz LO.
OH2FTG Labs >