I’m a sucker for continually re-building, I’ll have you believe *improving*, my 4m transverter, the first I built in 2010. The Nacton variant is the Mk5 iteration in as many years! The transverter is essentially the same, 28MHz IF and > 50w output (300w peak in later versions) on 70MHz and all in the same box, well sort of. Only the innards and complexity changed… Here is a brief history.
Mk1 – 2009
The Mk1 version used an OZ2M 70MHz Transverter in a box of its own (the bottom thin box in the image below) with a modified Bremi BRL-200 valve amplifier (top). What is not shown in the image below is a Spectrum Communication TA4S2 250mW > 45w 12v solid state amplifier, which was used, with a fair bit of input attenuation (30dB or so), to increase the 150mW output from the OZ2M transverter to the required 12w to drive the Bremi-200 to full output.
All this meant there were 2 PAs in series, a loose coax relay was used for the RX/TX switching and all all needed the associated cabling (for control and coax), this all means it was not very pretty or simple to setup.
I also could never get the Bremi BRL-200 to put out any more than 50w so I gave up on that so I built a Mk2 which used an MRF-151G RF pallet as the PA.
Mk2 – 2010
I had an old Bremi-BRL-200 box, in addition to the modified amplifier, which I used to house the MRF-151G pallet (for consistency), so the Mk2 version used the OZ2M transverter, still in its own box, the Spectrum Communication TA4S2 and an MRG-151G housed in, well on, an old Bremi box. The MRF-151G only needed 3w drive vs the 12w of the Bremi BRL-200 so more attenuation was needed.
The new MRF-151G RF pallet needs 48v at around 20A, the Bremi box was just a little too small to house the 48v PSU I had so in the Mk2 version the box ended up being an empty space.
Mk3 – 2011
The Mk3 version brought the OZ2M and RX/TX coax relay into the old Bremi box. In addition the MRF-151G pallet was rotated so the amp was internal. With this there was little space for the Spectrum Communication amplifier so a homebrew equivalent was made.
The homebrew pre-drive amplifier was based on a design by GI0GDP. As I didn’t need as much gain, efficiency was not pushed to maximum performance.
Also in the Mk3 design there were lots of crossband operation between 6m and 4m so I added a new coax relay brining the number to 2 coax relays, this was to allow 4m/6m crossband when using a dual band antenna. As the complexity was rising, a new control circuit to handle this increase, including the additional relay to allow 4m RX/4m TX, 4m RX/6m TX, 4m TX/6m RX was created;
Mk4 – 2013
The Mk4 was an improvements on the Mk3 in terms of a large tidy-up and application of best practices.
This included boxing up the pre-drive PA, bringing the Low Pass Filter into the old Bremi box, this LPF (Low Pass Filter) was a new design and has built in forward and reverse power detection outputs which I feed into the controller, the controller also drives a 4×20 LCD on the front panel for displaying real-time transverter parameters and is based on a PICAXE-20X2 chip. Additionally a 48v > 12v SMPS (switched mode power supply) was added allowing the whole transverter to be powered from a single 48v PSU .
The LPF has good performance, as measured using my VNWA 2;
The LPF forward and revers power detections outputs are then used for a ‘high SWR’ detection circuit (reverse power) and a [backlit] analogue power meter (forward power). The PICAXE-20X2 controller circuit also updates the LCD and shuts down the PTT if a poor load on the RF output is detected while on TX.
The Watt indicator here (94W) is a rough calculation of DC watts on the 48v line (with an efficiency correction) and not RF Watts, although I could calculate it.
The a PICAXE-20X2 controller also does the following;
- Fan control with temperature based hysteresis
- Display the current draw on the +48v input using a ‘hall device’
- Temperature display
- Voltage display of the +12v and +48v lines
- DC Watts on the +48v with a correction factor allowing for PA efficiency
- RX/TX state
- Protection State
- Protection override by enabling PTT on power-on
- Dropped support for 6m/4m crossband as separate antennas are now used and not dual-band ones
Cosmetically the improvements have been in making the control lines all joined together and of the correct length. The OZ2M transverter is still used and has been mounted on it’s side using some rivet nuts.
Mk5 – 2014 Nacton
In late 2013 / early 2014 Sam G4DDK released a new 70MHz transverter known as the ‘Nacton’. The Nacton has some notable performance improvements over the OZ2M transverter, these are only slight but as stated on the entry statement of this article, I’m a sucker for continually re-building my 4m transverter and this seemed a good excuse to build a Mk5.
The *only* change therefor in the Mk5 was swapping the OZ2M transverter for the Nacton and the GI0GDP pre-drive PA for the 7w Mitsubishi version.
The 7W PA is a little more permanent, it needs to be for heat transfer to the heatsink. Here is the PA PCB fitted before the Mitsubishi RA07H0608M module was mounted.
The Mitsubishi RA07H0608M module is 12v Vcc and 5v on the bias to enable the PA. In my implementation both PAs (pre-drive and final) are mounted on the heatsink on the top cover. When I firs done this I only provided a switched +12v on PTT cable to the PA section (no 5v one). I therefore added an 78L05 voltage regulator to the 7w pre-drive board the 5v Bias was applied at the same time as the 12v Vcc which was on the +12v switched line. The mod is shown in red (click to enlarge).
A 1N4148 diode was used on the 5v out of the 78L05 to reduce the regulated 5v line down by 0.7v to 4.3v, this is a recommendation to keep the bias below 5v just as Sam originally done using R1 and R2. After fitting it I realised I could have just used R1 and R2 instead of the diode.
Using the Nacton my RX noise floor is now much lower than the OZ2M, this could be down to less RX gain on the 28MHz IF path. However signals from generator down to –140dBm can still be heard.
UPDATE on Mk5 (18/05/14)
The Nacton got it’s first real outing at the CRG rally in April which coincided with a 70Mhz RSGB Cumulatives contest. It worked great there but it was no real test.
It was then used properly during the Camb-Hams Lewis 2014 DX’Pedition where it worked far better than I hoped.
However we has some real issues with HF braking through to 4m, which we believe was down to “rusty bold effect” as can be seen in this video. If you listen carefully you can hear about 3 stations at different points.
When I had no breakthrough from HF the received was far better than previous years, and even with the HF breakthrough I still managed 38 4m QSOs in a week from IO68 I declare the Mk5 Nacton the best performing 70MHz transverter so far.