In 2021, due to the persistent inability to operate on shortwave frequencies, I decided to sell all my shortwave radios and began experimenting with more challenging frequency bands, focusing on building technical expertise. Consequently, I targeted a relatively accessible band within the domestic amateur radio spectrum that was largely unexplored: the 1.3 GHz band. My goal was to create my own 1.296 GHz radio station.
1.3 GHz amateur band is a secondary service within the domestic radio spectrum, with an operating range of 1.240 GHz to 1.300 GHz. Internationally, DX (long-distance communication) is typically conducted on 1.296 GHz.
1.3 GHz amateur band equipment currently has very few products that support multiple modes. The most common are the Kenwood TS-2000X and the Icom IC-9700. However, the TS-2000X is extremely rare in China, and the IC-9700 suffers from both a high price (well over $10,000) and instability in its frequency settings. Therefore, building my own equipment might be a more suitable option for me.
I looked at how foreigners were doing it, using pre-made 144/430 MHz radios and then using a downconverter, filter, and amplifier. However, in 2021, SDR technology had become very mature, so there was no need to use such a basic setup. It's perfectly feasible to directly connect an SDR receiver. Let's just do it!
As I have a background in software engineering, I haven't had any experience with circuit board design. Therefore, I decided to start by designing a 144MHz amplifier as a test project. I based my design on the G4DDK scheme and successfully created it. However, I haven't conducted any testing yet, so this article will not go into detail about the process.
Core
First, the core signal frequency source was chosen to be the HackRF One. It offers extensive GPIO control, a good community, and excellent development potential. I can directly implement a sequencer on the HackRF, and I can also introduce a high-precision clock through an interface, eliminating concerns about frequency drift. However, one drawback is that it doesn't support isolation between transmission and reception... This is a bit inconvenient, and I still need to add a relay.
Input amplifier
Before designing the front-end amplifier, I first needed to determine my target power output. My goal was to achieve 150 watts.
PS: Why is it 150 watts? Commonly used PCB materials are FR-4 (fiberglass reinforced epoxy laminate). Since 1.2 GHz already falls into the high-frequency range, the energy doesn't travel along the copper conductors but through the medium. At this point, the disadvantages of FR-4 become very apparent. Firstly, fiberglass is susceptible to moisture, which can change the dielectric constant and cause impedance changes. Secondly, the dielectric constant of FR-4 is large, resulting in some loss, and the Tg value is low, leading to unstable heat generation. Furthermore, when referring to MOSFET versions, P1dB is approximately 150 watts.
The output power of my HackRF at 1.296 GHz is only 11 dBm, which is a very low power level. To achieve the ideal P1dB of 52 dBm required for the MRF6S9160H, you would need at least 32.5 dBm. In this case, it's necessary to design a pre-amplifier.
MRF6S9160H datasheet
After reviewing the PGA103+ design for an LNA used by W1GHZ. W1GHZ Design LinkAfter reviewing the datasheet for the PGA103+, I decided to boldly use two PGA103+ chips as a front-end amplifier. The key advantages of this approach are that the chips are relatively immune to self-oscillation, have low noise, and can achieve approximately P1dB gain after one stage of amplification (around 11 dBm), which is close to the required gain for a second stage, ultimately reaching around 33 dBm. Therefore, I decided to start with a trial run.
Consequently, this circuit design was developed. (Note: Some of the LC circuits were based on the design of BG3MDO.)
After using my CMU200 for testing, the gain was excellent, achieving a gain of 23.6dB with no self-excitation. The results were very promising (no images available at this time).
Amplifier core amplification
The core amplifier in the power supply utilizes a circuit based on the MRF6S9160 from DF9IC. While directly copying it seemed like the best option, due to the complexity of matching and the difficulty of controlling impedance with the large solder pads, I haven't been able to find any suitable software for simulation. Therefore, I have had to proceed cautiously. DF9IC Design Link Prepare for purchasing and assembly after the Spring Festival.
Amplifier core control
The core control functions for the amplifier primarily involve monitoring and displaying PTT (Power Transmission Time) status, temperature, and power (TX/SWR), making the ATmega328P a suitable choice due to its rich ecosystem.
Regarding power detection, my current approach involves using a 30dB directional coupler to couple the forward and reflected power. After passing through an attenuator, an AD8313 is used to convert the RF power signal into a voltage signal.
For the AD8313, the maximum detectable power is 0dBm, and the output voltage amplitude is also relatively small. Therefore, an LM358 is used to amplify the voltage changes, and an INA226 is used for sampling. The I2C interface then sends the data to an ATMega328P microcontroller to calculate the final power.
Furthermore, the ATmega328P also needs to control the switching of the DPDT coaxial RF switch and the display of the LCD1206. I have currently designed a development board to verify the feasibility of this scheme.
However, the board was ultimately discarded due to excessive temperature variations in the amplifier design. Therefore, we decided to use the AD8362, which provides a voltage output directly corresponding to power, and simply use the microcontroller's built-in ADC for sampling.
Therefore, a second board was designed, using the STM32G030 microcontroller.
However, there were still issues with the SPI screen control, which led to the main controller component being abandoned.
At this point, we return to the core amplifier section.
To be continued...
We would like to express our sincere gratitude to: BG5TOX, BG3MDO, W1GHZ, and DF9IC.
闽ICP备2021006864号-7
闽公网安备35020602002794号