Can something like this radiate on some unwanted wavelength?
I suppose the only way to check is to use a spectrum analyser…
As with switch-mode power supplies etc there are likely to be good designs and bad designs.
Cable charging is not ideal because all new phones use USB-C which – unlike micro-USB – has a vey high insertion force and it takes two hands to jam the connector together, which is why wireless charging is a good idea.
For the car mount, I normally discard the whole back portion (as shown in the above pic) and use a dedicated adaptor for the particular car.
As presumably it is an induction device and will produce a magnetic field so compass systems are likely to be effected.
It’s AC, it should not affect the magnetic compass. Qi works at up to 205kHz and could therefore directly interfere with an ADF. Apart from that it’s probably not far worse than the usual cheap switch mode stuff you find in every piece of consumer electronics. But let’s wait for the results from the spectrum analyser.
Indeed; it must be AC since “DC magnetics” can’t transfer power. The bigger concern is that cheap designs tend to use poorly shaped waveforms with loads of harmonics, so you will get say 200kHz, 400, 600, etc, etc.
I have a spectrum analyser at work and will report…
Even good designs will emit harmonics, and anything that deals with a weak signal can be affected (e.g. problems with GPS receivers and certain VHF transmission frequencies, all the way up at the 11th or 13th harmonic, if memory serves). The power on these chargers will be in the few watts range, and it wouldn’t surprise me if there were detectable harmonics all the way up into VHF. The odd harmonics (3rd, 5th etc) will likely be the strongest as I bet most of the cheap ones just use a square wave. Also it wouldn’t surprise me if some of the noise goes back up the device’s power cable and out into the aircraft’s wiring (some ferrites will stop that).
There is a lot of emission from the wireless charger I have.
This is the background, with it off
and this is with it on
The above is 100MHz centre frequency, and 195MHz span, so it goes from 2.5MHz to 197.5MHz. I will do more tests today. The probe was a piece of wire lying on top of the charger, which is obviously quite a critical test. But for sure it is a very good harmonic generator because even a square wave would not have such a spectrum 
It is CE marked but everything is CE marked; it means nothing.
@tomjnx may be interested.
Peter wrote:
It is CE marked but everything is CE marked;
And not every CE mark is a sign of EU conformance:
That’s very funny CE actually means almost nothing unless there is a referenced standard to which you are certifying compliance. I say “almost” because if say it is a childrens’ toy then by implication you are certifying compliance with toy regs. But for electronics there is any number of regs which could potentially be applied, depending on the market sector.
I did more tests today, this time with the spectrum analyser properly set up.
This is the electric field radiated. This is background, 150MHz span, 30cm long antenna 20cm away, feeding into a 50 ohm input
This is with the charger charging the phone. You can see plenty of muck in the VHF band
This is the magnetic field. This is background, with a 10cm diameter loop, 20cm away, feeding a 50 ohm input
This is with the charger charging the phone, and again plenty of muck in the VHF band
Of course one cannot tell how this translates into real interference with avionics. The vertical scale is 10db/division so taking into account the 150MHz span shown, there is quite a lot of muck around -50db in the 100-150MHz section where VHF comms, VOR, LOC, are.
I had a look at the GS frequencies, with a 400MHz span, electric field. This is the background
and this is with the above 30cm antenna
So there is very little at GS frequencies.
The charger runs at 144kHz. This is with the magnetic loop around it. First the background; nothing on there
and this is the charger running. Lots of radiation, and obviously no attempt was made at any sort of sinewave generation
20cm behind the charger we have this, showing how fast the field drops off with distance
With the phone being charged directly with a USB-C cable we see nothing, which really surprised me. It shows just how good a job Samsung did. They had to, of course, otherwise the phone would have wiped out its own reception, but I am still surprised at such a lack of emissions at these low frequencies where they could afford to generate some muck
My conclusion would be that while this simple investigation doesn’t tell us about avionics interference, the fact that the charger is running such a crappy waveform means that interference is pretty likely – as shown above in the VHF spectrum.
I wonder how a car radio would handle this? VHF might be OK, with a rooftop antenna, but DAB radios (does anyone use them?) work only marginally at the best of times
I hear my Qi charger pulsating in the radio on my bedstand if they are too close together.
Of course neither have any isolation of any kind, but FM radio is the 20MHz band below the VORs that sit just below our VHF COMs, so I’d expect any gap would pick it up.
That is interesting. I didn’t expect the good results for the Samsung charger but also didn’t expect the wireless charger to be that bad. You are probably right with your assumption about square waves being used to drive the charger coil. You can do that kind of stuff almost entirely in software on a rather slow micro controller. I did that for RFID a decade ago and as far as I remember the waveform in idle mode didn’t look that bad (at least on an oscilloscope, never looked at the spectrum). With another transponder in reach things got ugly, however.
With VHF communication being AM it doesn’t take much noise to disturb the radio reception. Well, not the reception itself but enough to make the squelch useless which will force you to go crazy or to switch the radio off. I once connected a radio to my (European made, half decent) switch mode power supply but could totally forget airband reception with the antenna less than one meter away from the radio.
