Debugging an RFI problem on Field Day…
I spent most of the week in New York, so Field Day snuck up on me without me making any serious plans. But since Jeri has been homebrewing her own SDR I’ve been thinking more about that, and thought that I would dust off my SDR-IQ receiver and at least make some recordings of the crowded bands for field day.
So, I hooked it up, downloaded the latest version of Spectravue, and fired it up. And immediately found that there was some periodic high intensity noise, about 30khz wide and at intervals of around 160khz, all up and down the range covered by the receiver. Tuning down to the bottom, I found that there was indeed a sharper louder noise peak centered around 81Khz, and that we we were seeing mostly odd (but some even) harmonics all the way up to 30Mhz.
It seemed likely that it was a local noise source, so I immediately went around the house and checked all my fluorescent fixtures to make sure they are turned off. I knew from experience that the one in my garage can emit some pretty ugly hash, but switching them all off didn’t seem to help. I remembered that I had an air purifier upstairs turned on, so I switched that off. No help. I turned off TVs (which had never presented a problem before, so I couldn’t see why they would cause an issue now) but they also didn’t help.
I strained my brain: I had never seen this kind of noise before, it was probably local. What has changed in the electrical environment of my house? I couldn’t figure it out. I mused that maybe I had some bad connection in my antenna (which is probably not adequately weather proofed with coax seal) and that it was rectifying some local noise source. I contemplated tearing it down, and putting up another antenna.
I needed a break. Carmen and I took a walk down to W6CUS’s Field Day operations near the Berkeley Marina, and then went to Home Depot and bought ourselves a new replacement screen door for our patio. When we tried to get it in the Honda, it wouldn’t fit, so I sent her home to get the Expedition and waited, mulling it over.
When I got home, I was thinking of a different problem: the design of a VLF loop antenna. I disconnected my laptop and moved to the couch, where I dug up a copy of some loop antenna design software and played with some numbers for a while.
After an hour or so, I got bored, and decided to give the SDR-IQ a try. I went back to my antenna hookup, plugged in the SDR-IQ to my laptop, and fired up Spectravue. And… miracle of miracle… the RFI was gone!
Perhaps, think I, it was something like a motor in an air conditioning unit that had switched off. But huzzah! I could at least tune around and listen. So, I put the laptop back on the table. And.. the RFI is back.
Take it off. It’s gone. Put it back. It’s on.
Alright Mark, think. What the heck is going on?
Sure enough, even proximity to the power supply was enough to bring back the hash. Then I remembered, this laptop power supply was a replacement (an OEM replacement) for the original that I had used previously. That could explain why I had never seen it before. My wife’s laptop has the same kind of power supply, so I swapped that in. Silent as a church mouse. No problem at all.
Problem solved! And with that, I could go ahead and listen (and see, using Spectravue) the traffic on Field Day. By then it was 10:00PM, a couple of hours beyond sunset, but 20M was still open:
I recorded a half an hour of the full band for later experimentation, and then moved down to 40m, where you could see both the CW and voice portions of the band.
Anyway, it points out to me that the hardest part of debugging is to figure out what is relevant and what is not, what is changed and what has not. I had a preconceived notion that since I had used this computer before, nothing related to it could possibly have anything to do with it. But of course something had changed. In retrospect, noise from switching power supplies should always be on my mind when debugging these things.
Hope everyone had a good Field Day!