UV-K5’s HF full-band reception (LSB, USB) circuit (confirmed version)

UV-K5’s HF full-band reception (LSB, USB) circuit (confirmed version)

After releasing the full band reception firmware of UV-K5 using SI4732-A10, various experiments were conducted. This post shows the experiment results and the final circuit diagram.

1.About Receiving all HF bands on UV-K5

Most shortwave radios sold recently that support AM, LSB, and USB reception use SI3732 or SI4735.  In a typical circuit, approximately 15 to 20 additional components are required to operate the SI4732-A10.

The goal is to attach the SI4732-A10 with the minimum number of parts by utilizing the parts already installed on the UV-K5.

2.Experiment

Below is the first released version.

http://www.hamskey.com/2024/01/modification-of-si4732-a10-for-full-hf.html

2.1 Generate clock from UV-K5'MCU(CPU) and input directly to SI4732-A10

SI4732-A10 requires a 32.768Khz clock.

I generated that clock from the UV-K5's MCU (CPU) and connected it to the SI4732-A10.

The result is a complete failure.

FM broadcasts are received normally. Although the frequency of AM shortwave broadcasting fluctuates, it can be heard. USB and LSB cannot be heard at all. The frequency is very unstable.

UV-K5 does not use a separate crystal for the MCU (CPU) but uses an internal oscillator circuit. Because it is an LC oscillation, stability is low and this phenomenon occurs.

Although it failed, I would like to thank those who looked at the circuit together on github and gave various opinions.

2.2 Using 20Mhz X-tal used in BK4819 of UV-K5

I connected the 26Mhz signal connected to BK1080 to SI4732-A10, but it did not work.

As a result of several experiments, it was confirmed that, unlike BK-1080, SI4732-A10 maintains the bias at 1.5Volt.

Additionally, the amplitude of the 26Mhz signal is quite small. Therefore, I used a fairly high bias resistance and the result was successful in operating the SI4732-A10.

As in the circuit above, a 1.6Volt bias was created with 200k * 2, and SI4732-A10 started operating.

Below is a video of the above circuit in operation.

But a problem occurred. No matter how I modified the firmware, the USB and LSB were not properly separated. At first, SSB mode did not work, but I tried various methods and succeeded in receiving SSB mode as shown in the video above, but the single sideband (USB, LSB) could not be separated properly.

It looks like the PATCH file for using SSB (USB/LSB) is calculated at 32.768Khz. (This is just my guess) 

The most important thing when DSP processing a signal is to separate I and Q. Once I and Q separation is successful, LSB, USB, AM, and FM signals can be extracted using several simple algorithms. It is assumed that 32.768Khz was used as the reference frequency when extracting the 90-degree phase signal when separating I and Q.

This modification can be considered a 30% success.

2.3 Removing the resistor and capacitor used on the RESET pin

Below is the circuit diagram used in the existing H/W Mod.

This is a modification to remove the capacitor and resistor in the red square in the circuit diagram below.

I succeeded in removing all of those parts, and just connected pin 9 (RST) directly to point A on the board below. (The existing firmware does not work and operates normally from 0.4HF)

3. Final Schematic

Removing the 32.768Khz crystal failed and I finalized the circuit diagram below.

When connecting the above circuit diagram to the board, refer to the picture below.

Point A in the picture below is connected to pin 9 (RST) of SI47320-A10.

Pin 16 of the SI-4732-A10 is connected to pins 3 and 4 where the BK-1080 was, but if you want to increase the volume, connect it to point B. (I can't guarantee the stability of this, but I connect it to point B and am quite satisfied)

C points will be used in the next reception rate improvement chapter.

3.1 Recommended 32.768Khz crystal attachment location

I placed the crystal on the UV-K5's board at a point where no components were attached and it gave quite satisfactory results. It was quite easy to perform hardware modifications as it did not interfere with the speakers.

As shown in the picture below, a 32.768Khz crystal and two capacitors were attached just below the LCD. In particular, the part where the capacitor is attached is easy to solder because there are no parts.

4. Modification to improve reception performance

This is the part where the circuit diagram was explained in the first release.

Remove the two parts in the area marked MOD.S1 below. The reception rate will improve considerably.

To further improve the reception rate, replace MOD.S2 with a capacitor of 1uF or higher.

(But I didn't want to use additional parts, so I connected pin 8 (AIN) of SI4732-A10 to the left side of the capacitor of MOD.S2.

This omits DC-BLOCK and is not recommended by SI4732-A10. I have been using it like this for several weeks and am feeling excellent reception performance without any major problems.)

5.Firmware download

There are no major improvements in this firmware.

Just a few minor things and the ability to use RST directly connected to point A were added.

https://github.com/phdlee/uvk5cec/releases/tag/v_01HF