Project #1 Grey Cube DRO LO

This DRO oscillator is part of a Grey Cube from DMC.  I would like to use this oscillator as a locked 24GHz source as a beacon or receive LO .Original it's  TX frequency was 22.974 GHz according to the sticker. This oscillator has a mechanical tuning possibility which moves a piston towards the DRO. The oscillator has 2  SMA outputs, a male and a female one. The female one originally fed the RX mixer, while the male one drives an amplifier modulator.

Since I have no equipment for reasonable accurately measuring power at this  frequency I experimented with a 8472A HP diode detector at hand.

Other experiments using a P481, 12-18 GHz waveguide sensor on my 432A power meter, with a P281B waveguide to APC7 and a Suhner APC7 20 dB attenuator into a APC7 SMA adapter failed.

This failure is due to observed resonances (very sharp notches, which change with tightening and loosening connectors a bit or changing frequency)

I started of using a known power at 100 MHz and taking readings using 1 dB steps from -10 up to 15 dBm, the maximum power being 20 dBm for less than 1 minute according to my HP1980 catalogue. Originally this detector can be used up to 18 GHz. The figure below shows the measurement result using a semi-log display for the detected voltage. The 2nd figure uses a linear voltage axis. The semi-log plot shows the diode behaves nicely as a quadratic detector.

Figure 1

Figure 2

Figure 3

The 3rd figure show both o/p powers of the DRO LO as a function of mechanical tuning.

23616 MHz = 24048 MHz- 432 MHz

23760 MHz = 24192 MHz - 432 MHz

The lowest frequency is obtained with the mechanical tuning piston removed (or at it's top position).  Lowering the piston increases frequency. Measurement made around every 50 MHz. Around 23.9 GHz oscillation stops. Measurements were calculated into powers based on the 'calibration' at 100 MHz. Frequency was measured using a waveguide mixer on my HP8555 spectrum analyzer an measuring it's 1st LO at zero IF span. Accurate enough for the purpose.Conclusion : This oscillator is not useable as beacon at 24048 or 24192 MHz but can used as a locked LO.

According to latest IARU recommendations 24192 MHz should be dropped in favour of 24048 MHz

for details please have a look at Arie PA0EZ's http://home.hccnet.nl/a.dogterom/

 I didn't see any effect by pumping a sub-harmonic into one of the o/p ports while observing the other port with my spectrum analyzer to see if injection lock might be useful.

The supply voltage normally is 5 V. To see frequency over voltage change I measured frequency at .5 below and above 5V.

Figure 4

The 4th figure shows the output of my 8690B sweeper with 8696A plug-in. Attenuators used are Suhner SMA type DC-18 GHz (6810.19.A)

The notch using 2 attenuators is not due to measurement error, it reproduces over and over again. Attenuators seem to function reasonably well at 24 GHz though.

Figure 5 showing the differences of the traces shown in figure 4

Figure 6