I’ve found that when someone tries to convince you that something is anything except what it appears to be at face value, they’re gaslighting you. Or as I sometimes call it, “Occam’s Knife Game” in which the player stabs around at every possible explanation except the simplest one, and in so doing reveals that they know what that explanation really is — it’s the one they’re avoiding.
In our latest episode, PG&E actually put forth some minimal effort for a change and sent some techs out to go noise hunting. Note that we’ve been down this road before, and they affirmatively located a noise source within the Poplar Street substation. This time their motivation in signal hunting appears to me not to have been for the purposes of actually locating the source of their noise, but rather to find something — anything — else they can blame the interference on.
They don’t seem to understand that when I’m hunting noise, I’m using a directional antenna, specifically a 3-element 220 MHz Yagi, which has a front-to-back ratio of about 20dB with a small amount of gain in the reverse direction. It’s fairly easy for me to tell the general direction a noise is coming from, provided I take measurements from more than one location. This is important: because there is indeed a small amount of gain in the reverse direction, and some small lobes of gain off to the side, it’s always necessary to take more than one directional reading, and not in the same line.
When I first tracked the interference from the PG&E Poplar substation, I did exactly this: I took the initial measurements from my station location. Then I took additional measurements from an area accessible to the public some distance away. By moving myself to another location, if the original reading was in fact coming from behind me, then from the second location, my reading would no longer point to the same general area as the first, and I’d need to start over. But that’s not what happened on any occasion that I’ve reconfirmed the source of the interference. No matter if I stand at my station location, or stand at the Poplar pump station, the noise always appears to be coming from the Poplar substation. These two locations are almost perpendicular to each other.
From their hunting efforts, PG&E believed they’d found the “real” culprit, a cable TV service amplifier giving off spurious emissions on 30 MHz, and I was asked to see if I could confirm whether this was the same noise I was hearing at my station. There are a few problems here already: I’m not monitoring or trying to use 30 MHz — that’s not even in an Amateur Radio band — and the type of noise I’m receiving is not the kind you’d normally expect to hear from a switching power supply or from spurious emissions from a cable TV amplifier. I am receiving, and specifically monitoring for, 120 pulse-per-second arc noise, the kind of noise you’d get if the two peaks of a 60 Hz current were arcing to a ground wire or grounded object. (My monitoring software watches 3.540 MHz and specifically cross-correlates the time-domain signal looking for a 120pps noise. It will not easily identify other types of signals that don’t match this pattern.)
So I went out, again tonight, this time looking for this cable amplifier. I was not able to locate it where PG&E asserted it could be found. I did detect a “birdie” signal in the general vicinity of its supposed location at about 2300 Hz, but again, this is not the noise that’s causing interference at my station and not the noise I’m monitoring.
PG&E also attempted to point their fingers at a number of photovoltaic rooftop installations. This is, of course, silly: the noise I’m receiving persists well into the night. No sun means no noise from photovoltaic arrays. We can immediately discount all of those.
But while I was out hunting for this mysterious cable box, I did find a utility pole making a hellacious racket. I hadn’t noted this before, but it’s immediately next door to the pole that I reported back in 2024 and which PG&E actually fixed a few months later.
Then I went to stand in the middle of the street with the cable box’s alleged location to my left and the substation in front of me, 90 degrees to each other. If the true source of the trouble was the cable box, this would be a perfect location to tell whether the noise was coming from in front of me, or to my left, a vantage point not available from my station location, where the alleged cable box would have been more or less behind me with the substation in front of me.
And guess what: no noise to the left, plenty of noise to my front. Shocking.
For the sake of completeness, I then headed out to the Bay Trail and north to the Poplar Creek pump station. This is as close as someone can get to the PG&E Poplar Substation while remaining on public land. I’ve been here before, many times, for the same reason: to reconfirm that I’m not crazy, and that the substation is really where the signal is coming from. You can probably guess the rest: yep, the signal is coming from the substation. Exactly where I’ve been finding it for the last two years and exactly where I’m likely to find it again tomorrow, and next week, and at this rate, in another two years.
And much like my effort at the first location, here, too I found another utility pole producing a massive amount of RFI, this time one adjacent to Poplar Creek. There are actually two poles on opposite sides of the creek here, and because of thick vegetation and this area being known to harbor rattlesnakes, I opted not to get sufficiently close to differentiate the poles at this time. I had detected some intermittent noise from this pole before, but nothing like what I heard tonight. I suspect that like much of PG&E’s aging infrastructure, this and the other pole I found tonight have just been degrading over time, and until they actually cause an emergency situation, PG&E is happy to simply ignore any maintenance.
I’ve captured my direction-finding and some audio recordings of the noise I observed on an annotated Google Map. Here’s a preview:
Gee, where could that noise be coming from?
To be completely scientific about it, I need to take another set of directional readings outside when there isn’t any notable interference, to prove a correlation (which I have also done before, but I’ll do it again for completeness). The RFI has been extensive and nonstop the entire day today, however, and even as I write this after 9 PM PDT, the noise is still S9+6. It will be necessary to get up quite early one morning, or to get lucky in the middle of the day for another set of observations.
In their effort to try to convince me that the problem is everything except PG&E, what they’ve actually done is gotten me to confirm that the source is still exactly where I’ve been saying it is for the last two years and I’ve also found two additional sources that weren’t there before.
As best as I can tell, something inside the Poplar Substation is still producing arc noise. Even PG&E has confirmed in the past that they identified a problem inside the substation, though they’re trying to deny it now. They want this to be everything except their responsibility, which is the kind of cultural defect that’s brought about multiple PG&E-caused wildfires, a recent substation fire in San Francisco, and, I suspect, some kind of accident waiting to happen at the Poplar substation here in San Mateo.
PG&E has now been dodging this problem for a few months shy of two years. If there ever is a disaster at that substation, I sincerely hope that all the records I’ve kept over all this time end up being called something like “People’s Exhibit A.”