It works at all because the microwave RF is very narrowband, though the frequency drifts all the time. Bluetooth has FHSS so it will eventually go though, WiFi can also often survive but it's a bit harder. Although at some short distance you'll saturate the analog frontend of the receiver, then the microwave frequency or becomes irrelevant.

>but I don't have a microwave :)

Living the dream!

If it's an old microwave replacing the door seal would be cheaper and easier.

> You can thank Hedy Lamar for that.

eh... maybe. Don't forget the microwave isn't CW so there's plenty of transmission slots available on the off cycle.

Wifi has an "interference robustness" option which is designed with microwave ovens in mind. The power output of a microwave is a sine wave, because it's powered from your AC line power, and doesn't emit energy during the zero crossings. "Interference robustness" times packet sends (and lengths) to be when the microwave is producing the least amount of power. Thus, you spend half of every 60th of a second irradiating your food, and the other half downloading videos. (Needless to say, having this on and your microwave going in the background reduces throughput. But, dropped packets are worse than small packets that always get through, and this aims to eliminate completely stomped-on packets.)

Wifi also has a listen-before-talk model. If it "hears" a microwave running, it thinks it's another station transmitting, and backs off. This feature of the protocol is why long-range networks worked so poorly in the past. If station A can't hear station B, but the AP can hear both of them, then A and B are going to step on each other and the AP won't be able to communicate with either station. This is why the "enterprise" way of deploying networks was to have a ton of access points running at low power; that works well with the listen-before-talk model since the AP likely can't talk to or hear stations that are too far away for the stations in its range to hear.

I don't know if interference robustness still exists in modern standards, as I haven't seen it in a control panel for decades, but it was definitely in 802.11b. I have never tested Bluetooth (or read the standard), but basically... the industry knows this is a problem, and handled it a long time ago. Bluetooth might ignore the problem because it plans on frequency-hopping (away from the microwave) anyway, but like all software, that can easily be bugged.

I have an inverter microwave. Before I went crazy with APs around my house, that microwave would win the war; no zero-crossings to sneak a packet by it.

Now my bluetooth only drops when I'm fairly close to the microwave as it runs.

Are you saying that adding WiFi APs improved your Bluetooth connections when the microwave is running? Or is there such a thing as Bluetooth APs? Pardon my confusion.

Wow. I have no idea how I confused that.

Yeah, the WiFi is improved. The Bluetooth is the same as it ever was.

I pardon your confusion.

Cool, I just thought I was missing out on something. :-)

There is also a fun regulatory hack in bluetooth. As it is frequency hopping, FCC considers it spread spectrum, but only if it hops across a large enough set of channels. This matters because spread spectrum radios can run significantly more power.

Bluetooths frequency hopping system avoids interference to it and from it by dropping channels with interfernce or other users from the hop set.

So with enough interference, even monentary interference and the hop set reduces and regulatory limits require dropping power from 100-200mW to 25mW. This usually cuts the connection.

> Try cleaning the mating surfaces around the door thoroughly.

(I'm obviously not an electrical engineer, given this is 10x stuff) If we're talking a 2.45 GHz microwave signal, that's a 12.2 cm wavelength.

But I thought for shielding you only needed to have gaps of <wavelength to null emissions.

Is there some fractional-wavelength propagation, or is my understanding of EM shielding off-base? How are microwaves noisy? E.g. https://physics.stackexchange.com/questions/269672/does-a-fa...

> Is there some fractional-wavelength propagation?

"Propagation" is probably not the right word, but fractional wavelengths can "leak" some amount of field.

https://en.wikipedia.org/wiki/Evanescent_field

Notably, that field is powerful enough that microwave doors need to be about an inch thick. The rest of the inch is just empty space and a sheet of plastic to stop you putting your fingers into the evanescent field.

Unfortunately there were a few microwaves sold recently in Europe which forgot that design element, and if you pointed at food while cooking it, you would cook your fingers too.

You're right. The rule of thumb for EMC engineers is to bond joints at no further than wavelength/10 where the wavelength corresponds to the highest frequency that you wish to maintain good shielding effectiveness. Some MIL projects use wavelength/100.

But the gap around the door might only be millimeters wide, but it is 30cm or more long.

It is "both* dimensions that must be a long way below the wavelength to keep microwaves in.

Instead the door seal uses a technique called a quarter wave choke. It relies on reflecting back any microwaves wanting to escape, and by making incoming and reflected waves perfectly cancel, no power is transmitted.

As OP said, even a small amount of soup dripped onto the seal and your microwaves will all start escaping.

I wonder if building a faraday cage around it would do anything.