ATS-Mini V3(B)

[G8PTN]

Hi,
For general information, I recently purchased an ATS-Mini V3 (see below).

Internal view:

This is the version with the headphone amplifier on the display side of the PCB.

I have seen an issue matching the description from one of the reviews of this version.
“I also noticed occasional issues in the FM band: even on strong stations the audio sometimes fades and RDS data is lost, though tuning to another station usually fixes it. On some frequencies the strength indicator shows maximum, yet the sound feels slightly off‑tune.”

I believe this issue is likely to be caused by the headphone amplifier. The images below show the RSSI and SNR readings for a simple test case using the ats-mini v2.33 firmware which controls the amplifier(s) enable signals.

Example 1: Volume = 0, Amplifier(s) Enabled
In this case the SNR is poor and no RDS is decoded.

Example 2: Muted, Amplifier(s) Disabled
In this case the SNR is good and the RDS is decoded.

If the headphone amplifier is the same as that used on the ATS-Mini V2(B), the charge pump circuit may be the cause of the interference. The frequency of the charge pump oscillator probably drifts with temperature and supply voltage, perhaps explaining why the issue seems to come and go. At the moment I am not sure if the noise is radiated or induced, further investigations required.

I will probably disable and bypass the headphone amplifier, which should fix the issue.

The headphone amplifier used on the ATS-Mini V2(B) is an HT4832.

73, Dave

[G8PTN]

Additional observation using oscilloscope with a close coupled loop near the position of the headphone amplifier on the ATS-Mini V3(B).

This appears to confirm that the interfering noise is being generated by the headphone amplifier device.

73, Dave

Noise: Display Off, Mute Off

Noise: Display Off, Mute On

[EmEiBee]

Of course, if it's a (Chinese) Class-D amplifier. But purely analog amplifiers sometimes do this too, if an auxiliary voltage is generated internally (like the bigger of the TDA series).

[G8PTN]

Hi,
I don’t think the HT4832 is a Class-D output. The charge pump is used to create positive and negative supply rails, so that the output is biased at 0V, and does not require DC blocking capacitors.

Unfortunately, in this application, the hash generated by the charge pump is causing a bit of a problem.

For the cost of the radio (14 GBP), I am not particularly concerned, it is worth it for the components alone!

I intend to modify the radio, replacing the SI4732 with a BK4802 VHF/UHF transceiver chip (for RX only), which has its own on-chip audio amplifier, this should be adequate to drive headphones or the speaker directly.

73, Dave

[EmEiBee]

That's possible. Unfortunately, I can't find an English datasheet for the part.

But regardless, whether it's Class-D or analog with a charging pump... Such "hotspots" in a receiver circuit are always a problem, of course...

[G8PTN]

That's possible. Unfortunately, I can't find an English datasheet for the part.

Hi,
For reference, link to English datasheet below.
https://www.heroic.com.cn/public/uploads/file/20191114/1573715159633703.pdf
73, Dave

[Ilja-code-art]

https://www.rigpix.com/miscrx/amnvolt_atsminiv4.htm V4....

[G8PTN]

https://www.rigpix.com/miscrx/amnvolt_atsminiv4.htm V4....

Hi,
Detailed information on the V4 hardware can be found in the thread below.
#267
73, Dave