Question about low-pass output filters for Bluetooth receivers

Could you be heaing other tones as the result of intermodulation between the signal and the spurious frequency?

Sure. BTW the square box has almost 2 Vpp audio out mixed with 2 Vpp 350 kHz! You can hardly see the audio in it on the scope !

The USB ones have about 200 mVpp 350 kHz mixed with almost 2 Vpp audio out.

I will have to write a review for it. It will not be positve. I am not sure it would not cause damage when put ino some good wideband audio amp and is amplified..

I wanted to sent some audio from some of my Raspberry Pis to the TV, audio power amplifier and elsewhere.. Now done with long cables all over the floor. The build in bluetooth does not work in the P4s I have, as those are in an alu housing with a fan.

pwm/class-d headphone output amplifer/ADC ?

No it is not PWM output The 350 kHz is a steady nice sinewave, unfortunalely by the square model about 2 V pp added with audio that is also 2 Vpp maximum. I am going to try some parallel LC tuned circuit at 350 kHz in series with the headphone, scope what is then left of the 350 kHz at the headphone side. I am also going to listen an my AM radio for the third harmonic of the 350 kHz, can use SSB to hear if it is frequency stable :-) But I noticed a 24 MHz (I think it was) crystal in the small bluetooth PCBs, some chip, and some Rs and Cs, so the 350 kHz is likely very stable, looks stable on the scope.

You have a buetooth receiver? Scope it! Who knows, you may see the same crap.

PS

No it is not PWM output

I just listened to the bluetooth output with my Tecson PL600 radio, it has long wave, no need to look for harmonics and it found that RF at 370 kHz, clearly detected, and rock solid on SSB, no AM, no FM.

it could still be pwm just lowpass filtered

Try a cap across the headphones.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

I tried some 370 kHz parallel LC circuits - so high impedance at that frequency - in series with the headphone today, not working as I was hoping for, testing not finished yet, but a bit of interruption happened, had to go to my mains power backup as the power company was replacing the electricity as well as the gas meters with new electronic ones that report back the use via some wireless link. They are changing everything as many people feed back solar power into the grid, but they no longer pay much for the feedback, claiming it 'overloads' the wires / cables. That leaves people the choice of using batteries to store solar power themselves.

Anyways the Bluetooth alternative is a low pass and a separate power chip, easy to do with the small boards I have some 5 V supply audio power amps?

I certainly do not want 370 kHz into my big 180 W audio power amp and maybe melt the tweeters?

So a simple filter board added to the small Bluetooth ones I have.

Might have designed the whole thing from scratch, Maybe look for a filter board on ebay..

I designed my own loudspeakers with a separate amplifier for each drive unit. The tweeters have a timing mechanism which protects them in the event of sustained overload.

A simple 1st order RC should be enough. -3 dB at 15K maybe.

Use an RLC to be compulsive.

I bet the gadget has all of the FCC/UL/CE/CSA stickers.

Yes, but then an output amplifier / headphone driver is needed after the filter. I just found a small ebay headphone ampolifier I already have and forgot about.

So RC lowpass, I also have a RC lowpass in the output of my Raspberry Pi audio too, build into the connector cable. It has a lot of RF on the 3.5 mm output jack. Did you check yout big Pi's audio out?

Many years ago I had some interaction with somebody who build and sold transistor power amps. Those were used on fairs etc.. IIRC we discussed adding an extra input stage transistor, free from the feedback circuit, that limited input signals and did we add a lowpass cap? Just to protect those amps. Easy to get some amplifier oscillating if your input is part of the feedback by connecting all sorts of s*it, even defective input cables can kill it.

I have been wondering the same.. No such thing found ! Does have a CE mark on the plastic bag it was shipped it:

I am sure the plastic bag is OK ;-)

No, just keep the impedances low. What's the headphone impedance?

We use the Pi400 keyboard thing for development but don't use the audio out. I personally design with the RP2040 chip.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

I melted a bunch of tweeters by failing to filter out a 31.25kHz PWM carrier (from a 16MHz ATMega328, guess where that came from). That was an expensive lesson!

Kragen

That is the point, could be my Sennheiser headphones, or Chinse earplugs or whatever, or my audio power amp,

I used the Pi4 today to make some test tones, some here may like this:

# sweep audio 20 Hz to 20 kHz fr 30 seconds: play -V -r 44100 -n synth 30 sin 20+20000

# single tones play -V -r 44100 -n synth 30 sin 20000 play -V -r 44100 -n synth 30 sin 2000 play -V -r 44100 -n synth 30 sin 10000 play -V -r 44100 -n synth 30 sin 12000 play -V -r 44100 -n synth 30 sin 15000 play -V -r 44100 -n synth 30 sin 20000

You can also use an audio USB stick like I have, it then gets a dfferent device number assigned by the Alsa sound system:

Generate a single 20 kHz tone for 30 seconds: AUDIODEV=hw:3,0 play -V -r 44100 -n synth 30 sin 20000

# Sweep USB stick with audio out / audio in: AUDIODEV=hw:3,0 play -V -r 44100 -n synth 30 sin 20+20000

# sweep goes 'over the top' AUDIODEV=hw:3,0 play -V -r 44100 -n synth 30 sin 20+30000 Interesting, it starts again after about 20 kHz with distortion at the sample speed point !

Also mixer to set the volume

for the intern audio volume: alsamixer for the extenal USB audio device: alsamixer -c 3

Just had a lot of fun with it today, playing with the good Trio analog scope and noticed the little audio amplifier I have does indeed some lowpass.

Like somebody else mentioned, the output of the Bluetooth modules may be PWM. I did see changes is the sine wave zero crossing when triggering the scope from the RF while there is some music playing. Next I will look up (try to read) the chip numbers in the small Bluetooth modules I bought.

The RP2040 has two 200 MHz ARM cores. One could be dedicated to being a DDS synthesizer, and that could drive a PWM. So we'd have a 75-cent precision function generator. Maybe 95 cents with the opuput filter and an opamp.

USB would add two resistors.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics