A ProgRock and a very old MMT 144/28 Transverter

Having got the 10MHz side of the ProgRock project working (see here for details) the next stage was to look at the 116MHz output. If this was also acceptable, I would have an alternative means of generating the two frequencies (10MHz & 116MHz) necessary to ensure that my 2m and 23cms transverters were “on frequency”.

As with the 10MHz output the 116MHz output was a square-wave and a simple 7-element LPF (Low Pass Filter) was added. A quick look at the output on the SA (Spectrum Analyser) showed that the 3rd harmonic was ~ 38dB down on the wanted 116MHz output.

The output on an oscilloscope looked like this:

The final stages needed to complete the physical build of the ProgRock project was to build a small voltage regulator board and house everything in a standard “HSK project” die-cast box. The regulator board was needed to allow a 13.8V input and output of 5V for the QLG1 GPS receiver module and ProgRock module.

Once everything was fitted into the die-cast box I could start to use the QLG1 GPS receiver and check the frequency stability without fear of draughts or physical interaction / movement affecting the tests. The initial tests were done using SpectrumLab to view the 10MHz output. It can be seen from the next screen that the output showed no signs of drifting.

I was now at the stage where I could try to use the ProgRock to improve the frequency stability and accuracy of a very old MMT 144/28 2m transverter.

I purchased the Microwave Modules MMT-144/28 2m transverter new, way back in the early 1980s. It was originally used with a Yaesu FT101b and later with the ICOM IC-735.  When I acquired a Yaesu FT-225RD the transverter was “retired”and has basically been sat in a cupboard gathering dust since then. In its day the MMT 144/28 transverter was very popular, many are still in use today.

The local oscillator specification of the MMT 144/28 was quoted in the owner’s manual as being:

In its day that was acceptable for normal SSB and CW operation, but by today’s standards, especially for digital modes, it’s far from acceptable.

So the first test was to see if the transverter still worked, and then to see what the frequency stability and accuracy was like. I applied power to the MMT 144/28 and carefully monitored the current drawn and looked for any escaping magic smoke. It passed this test and seemed quite happy.  Tuning to GB3VHF proved that everything was working but the frequency was off by >1KHz and things were drifting.

A test on transmit, with the transverter running ~ 2W into a dummy load, showed that the transmit signal was drifting about 20Hz over a one minute period.

Clearly this would not work well with any of today’s digital modes. So the next step was then to try to either “reference lock” the original 116MHz oscillator or replace it with the 116MHz output from the ProgRock. For simplicity I chose to do the latter.

Fortunately the MMT 144/28 has a spare BNC socket on the back panel so it was easy to feed in the external 116MHz signal. Levels were adjusted using a 1dB stepped attenuator while monitoring the MMT 144/28 output on a spectrum analyser.

Here are the initial test results:

The 116MHz local oscillator (ProgRock) is ~60dB down and harmonics >55dB down.

Frequency accuracy and stability is vastly improved. The following screen grab shows GB3VHF (in-between keying sequences). The minor “glitches” are the 1PPS corrections taking place in the ProgRock.

The real test was to see if / how this 35+ year old transverter would work with today’s digital modes, in particular JT65b. To do this I used the MMT 144/28 and my TS-2000 (on 28MHz) in place of my normal 2m receive EME setup (K3 + Anglian transverter). The results were very encouraging, as the following screen grabs show:

Now I happen to know that Conrad, PA5Y uses a GPSDO for frequency control so the 8Hz difference between the DF and Dop: values is down to my TS-2000 which is not GPSDO disciplined, and not the ProgRock! 🙂

I continued to decode a number of stations including UR3EE and the Z66EME Dxpedition.

Arguably receiving these stations is hardly surprising given that the MMT 144/28 is being used with a masthead mounted LNA, but the key thing for me is the frequency accuracy and stability. This is especially important when operating EME, especially on the higher bands!

The next steps:

  • To check the MMT 144/28 alignment on both transmit and receive.
  • Run some further long-term tests and measurements on both receive and transmit.
  • Test the MMT 144/28 in conjunction with the SG-Lab 23cms transverter.



This entry was posted in 23cms, Blog, EME, GHz_Bands and tagged , . Bookmark the permalink.