I was wrong. I mean like, publicly wrong. Like, in front of the whole world wrong…
Perhaps I should explain.
See, what had happened was…
Not too long ago, I posted on social media about some interference I inadvertently caused that disrupted the operation of my garage door opener. See, I have this pair of Phoenix Contact RAD900 wireless IO radios on my workbench. I love them because they operate in the 900MHz frequency range and because I am a big old nerd. (For the record, if you ask any of my children “What does a Nerd do?”, they will emphatically say “Nerds get P A I D!” no shame in our nerd game)
Anyway, after running them for about a week, I discovered that while they were powered up and connected in my lab, which happens to be adjacent to my garage, that my garage door openers wouldn’t work. I tested my theory several times and was able to confirm that the RAD900s were indeed keeping the remotes from communicating with the opener. I was essentially DOS’ing the $#@% out of my garage! Now I had to figure out why.
I knew exactly what frequencies the RAD900s used, that was a no brain-er. The mystery was, what frequency was my garage door opener using?
Garage door documentation sucks!
I pulled out the documentation that came with the opener. It’s one of those fancy MyQ ones that you can connect to your home WiFi and then control with an app. You would think since it is a wireless device with a fairly important function that it would have some decent documentation on its capabilities, right? That would be a big fat NO. It mentioned that it connected to WiFi on 2.4GHz and then talked about its rotating codes for the secure operation of the remotes. Nowhere did it talk about what frequencies it used for said remotes.
Then I hit the wild west of the Internet to practice some Google-fu. I started getting frustrated as I couldn’t find any solid information any garage door openers, let alone the model I had. After an hour or two, my brain started assuming that the opener MUST be in the 900MHz range, how else could those radios be interfering with them.
And that folks, is where I was wrong. But I am not the kind of guy that can just be wrong and be quiet. No sir. I went and made a grandiose LinkedIn post all about it. It was a great post too! I got lots of feedback and some of it was pretty funny. I even made a new friend or two. I was riding high on that post for a while.
Fast forward a couple of weeks later. I wasn’t satisfied with my results from before and I was a few beers into the evening, so I started the search over again. Several hours later I finally start finding some information, and on Quora.com of all places. It turns out that most openers operate in 300-315 MHz and 390 MHz frequencies. Now I was both confused and frustrated. This just did not compute! How was a 900MHz radio interfering with a 300-315 MHz device?
Figuring it out
The next day I was discussing my quandary with a co-worker. He happened to be a radio & RADAR guy back when he was in the Navy so I figured I would pick his brain on the subject. After going through the entire story in grueling detail (and admitting I knew that I was wrong, which still stings a little), he smiled and said, “Sounds like harmonic interference to me”. Now, I have been a wireless guy for some sixteen plus years and had never known about this phenomenon. He proceeded to tell me that while it wasn’t common, it was a known issue on Naval ships due to all of the lower frequencies that they used for communication. They had to be careful which ones they used at any given time to make sure they were not accidentally interfering with their own transmissions from other radios onboard.
That night, I tore through every CWNP book I have in search of this elusive (to me anyway) concept. I couldn’t find a reference anywhere in any of them, not on the regular track or the IoT one. Back to Google I went.
So, what are RF Harmonics? Harmonics are that part of a signal whose frequencies are integral multiples of a fundamental frequency. For example, with a 100Hz fundamental frequency, we can expect harmonics at 200Hz, 300Hz, 400Hz, you get the idea. As harmonic levels increase, so does the possibility of interference of the lower frequency. This is because the higher frequency impacts the lower frequency multiple times.
In the example above, on the left, the blue wave is the fundamental frequency (or first harmonic) while the green wave is the second harmonic. In one wavelength of the fundamental wave, the secondary harmonic impacts it at three separate points along the same plane. The result is a distortion of the fundamental wave. This distortion may not be enough to actually alter the encoding of data carried on the wavelength, but in some cases it could.
I found several harmonic calculators online and used them rather than to trust my own (sometimes) flawed math skills. Picking a first harmonic frequency of 306MHz, the harmonics breakdown looks like this:
|First Harmonic||306 MHz|
|Second Harmonic||612 MHz|
|Third Harmonic||918 MHz|
|Fourth Harmonic||1224 MHz (1.224 GHz)|
|Fifth Harmonic||1530 MHz (1.530 GHz)|
And there was my answer. My garage door opener must be using the low 300 MHz range as the RAD900’s uses the ISM 902-928 MHz bands. After running all the numbers through the calculator, my opener has to be using 301-309 as they all have third harmonics right within the ISM 900 MHz frequencies.
Now, I know what you are thinking, “Scott, bro… the RAD900 uses a frequency hopping spread spectrum method of communication for stability and interference avoidance, how can they be stomping on your poor little opener?”
Well, it’s simple really.
By default, the RAD900 radios transmit at their maximum power of 1 Watt. I had not adjusted those levels down. On top of the base transmit power, I had 3db omni-direction antennas (direct attach rubber-duck style) connected to them. I was essentially cranking out 2 Watts of transmit power on both radios. Meanwhile, my poor little battery powered garage remotes are pushing out just a few milliwatts in short bursts based off of the push of a button. Combine the extreme power overload of the higher frequency with multiple contacts on the same waveform plane of the lower frequency and you have the perfect storm scenario for signal distortion.
Through this whole process, I have learned quite a bit. The concept of harmonic interference is not something that was taught in any of my CWNP material or any of my Aruba Networks material from way back either. While I can see this as being an uncommon problem that most WiFi engineer types won’t run across, I feel that it’s worth a mention somewhere. I guess I’ll have to bring it up to the board when I finally get to apply for my CWNE.
If you have enjoyed what you have read, follow this blog as I share my experiences, blunders, how-to’s, tips and opinions in all things OT Wireless from the wonderful world of industrial WiFi!