The most obvious things are the hardest to spot: RF interference and the bluetooth keyboard

So I'm a bit embarrassed how long this took me to clue in. I have a spiffy bluetooth keyboard. Its great, its got a touchpoint on it, great key travel. Its basically the same keyboard I had on much much loved x300 years ago.

But, funny thing, in the evenings it would behave badly. The touchpoint would wander, the keys would stick or not register.

OK, you should know bluetooth and WiFi don't get along. They both operate on 2.4GHz and there is lots of papers written on this, patents covering ways to avoid, etc. But the upshot is, they don't play nice.

So I moved the bluetooth dongle to as close as I could get it (in the monitors USB hub). And,well, it got better I thought. But still some evenings it would be better, some worse.

So last night I'm looking at my desk lamp. And my misbehaving keyboard. And I'm thinking back to my shortwave radio days (Halicrafters SX-130!). And I'm thinking about how much better the propagation was at night, and can that be related? But, shortwave is, well, shortwave. Its measured in KHz. And bluetooth is 2.4GHz. That propagation is due to things that are **much** to large to affect my bluetooth. So that can't be it.

But for sure this happens at night. And I'm looking at my desk and my fancy desk lamp. I'm pretty happy with the desk lamp. My eyes are getting old and dim and sometimes I need a lot of illumination. And it has 2 x 100W 'equivalent' LED bulbs in it. And, like the rest of my house, is voice-activated and WiFi controlled. And its right by the keyboard...

Argh. The light. Its radiating like a banshee back that lamp cord!

So I go and find a big old DVI cable and rip off the ferrite. And then I take apart the light control module (this is one of the ones I used a Sonoff on). I note that the lamp cord is kinda dangerous in the nail-head splice junction. So I put some ferrules on. I then thread the cable through the ferrite twice (the reduction is the square of the number of windings).

I reassemble, and, well, its better. Is it perfect? Hard to say. But, now it makes sense why it was worse at night (the light is on! and the light is WiFi!).

Here we can see the sonoff module, all naked, and the new ferrules in the inlet side.

And now lets look at the final product.

Its, well, ferrite'd. And, that should help prevent some of the RF from backfeeding down that 'antenna' that is a 2m unshielded power cable, running underneath my monitor and its precious cargo of bluetooth dongle bits.

Maybe I should have tried a bigger ferrite and 2 turns? 3 turns? O well, lets give it a try.

Morale of the story... its not that complex, don't go looking for 'ionospheric bounces of nighttime RF propagation'. Its the $!# WiFi lamp that is next to the keyboard!

2 comments on “The most obvious things are the hardest to spot: RF interference and the bluetooth keyboard
  1. db Lee says:

    My favourite shortwave radio ionospheric propagation story is the little girl in St Petersburg, Florida, who picked up some of Amelia Earhart’s distress signals and logged them in her diary. No one believed her at the time, of course, but she was later vindicated with the whole Nikumaroro thing.

  2. db db says:

    The shortwave was my ‘Internet’ in those days. I could tune into Voice of America, BBC World, Radio Canada International, Radio Moscow, various German ‘numbers stations’, etc. This was the ‘world news’ for me in rural early 80’s ontario.

Leave a Reply

Your email address will not be published. Required fields are marked *

*