1
\$\begingroup\$

I have a wireless Qi charging system for my mobile phone. When I put the phone into the cradle it starts getting charged and I can hear a very quiet noise coming from the arrangement. Other questions (like What might cause audible noise in an inductive charging system?) are dealing with why there is noise at all, and that's understood.

I am curious why the noise (at least in my case) is so strangely rhythmed. It is not a constant squeal as the author of the other Q described. It is rather a repeating rhythmic tic pattern with a period of just under two seconds. I can reproduce the pattern by saying something like

chip-chip-chip-chip-chip-chip-tatata

and repeat this right away (so it goes ...-chip-tatatachip-chip-chip-...). The delay after each chip is around 0.27s, the delay after a ta is around 0.09s, so that as a whole the pattern has a period of 6×0.27s + 3×0.09s = 1.89s.

To the musicians among us: The rhythm sounds like six 8th notes and then a triplet of the length of another 8th, so it is a 7/8th rhythm.

Can anybody explain what't going on in these circuits? I have been searching for an answer for quite some time now but I found no mentionings of any rhythmic aspect in combination with wireless charging in general or Qi specifically.

It might be a special property of my charger of course, so I am not sure that this is reproducible for others.

EDIT: I managed to create a recording of the sound. It is very silent, so there is a lot of noise captured as well. Especially a wall clock in ~1m distance is clearly audible over the signal. Here is a spectrogram of the recording over several seconds in the range 4kHz - 20kHz. The pattern is clearly visible (and on the bottom you see the ticking of the wall clock).

I would love to upload an mp3 of the recording but this site doesn't make it as easy for sounds as for images. Does anybody know of a equivalent of Imgur for sounds?

EDIT 2: I uploaded a recording to YouTube as a commenter suggested: https://youtu.be/xKPyiHGkoks

Spectrogram of a recording of the sound

\$\endgroup\$
  • 3
    \$\begingroup\$ ”Nureek.” So the next one will be a “rotut” and the one after that will be a “hanunga.” \$\endgroup\$ – winny Nov 21 '18 at 17:31
  • 2
    \$\begingroup\$ Definitely aliens trying to communicate. \$\endgroup\$ – Andy aka Nov 21 '18 at 18:03
  • \$\begingroup\$ the charger circuit consists of two coils of wire in close proximity to each other ..... the coil in the charger base is being driven by AC ..... as a result, any ferrous object that is brought into close proximity of the charger base coil will vibrate like a speaker ..... i think that you can fill in the rest ..... also read this en.wikipedia.org/wiki/Magnetostriction \$\endgroup\$ – jsotola Nov 21 '18 at 20:16
  • \$\begingroup\$ @jsotola I guess you didn't read very much of my Q otherwise you would have found that I already linked to another Q which covers the topic of magnetostriction. My Q is about the rhythm of the audible ticking while charging. \$\endgroup\$ – Alfe Nov 21 '18 at 21:27
  • 3
    \$\begingroup\$ I'm curious to hear the sound, can you record and upload it? \$\endgroup\$ – peufeu Nov 21 '18 at 22:03
3
\$\begingroup\$

To keep the received voltage where the receiver wants it, the receiver sends data packets back to the transmitter. When the voltage is close, the packets are sent about every 250ms. When the voltage is not close, the packets are sent about every 32ms. If the transmitter does not detect these packets, it will shut off the power and assume the device is gone.

Several factors may be involved in the audio effects you observed. One is the transmitter changing its modulation in response to these requests it is receiving from the receiver. Another is the data itself, which is transmitted by modulating the load the receiver applies.

The transmitter uses a PID algorithm to adjust its output power to provide the receiver the voltage it requires. The load drawn by the receiver varies with this voltage, creating a system that is a bit difficult to stabilize. Apparently, this algorithm is dithering in a loop that includes a portion that is somewhat far off from the desired voltage.

Getting PID algorithms to smooth out is notoriously difficult. It's somewhat analogous to a bad driver who notices he's too far from the car in front of him and so starts pressing on the accelerator but then doesn't notice he's too close until he needs to tap on the brake, repeating the cycle.

Qi's need to support a variety of chargers and devices probably made it difficult to get it to perfectly zero in on a stable setting.

The "chip-chip-chip" you are hearing is likely these coarse adjustments. The faster "tatata" sound is likely when it gets far enough off that more frequent corrections are required. As expected, the "chip"s are roughly 250ms apart. The "tatata" is likely three fast adjustments, which we would expect to be 3x32ms or .092 seconds. This fits your measurements nicely.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.