i am the ghost host

[solarbird]

I just spent the whole day – daytime day, I quit at dinner – rewiring my studio to better isolate the AC power lines from everything else, because it’s pretty clear at this point that our AC power is just scary with RFI, and the new high-gain preamp is AC all the way to its motherboard, unlike everything else, which is DC power once you get past the adaptors.

Result? All this did NOT A FUCKING THING for the high-gain preamp. In fact, it’s worse tonight. Enough that were I trying to use it, it’d be difficult and I’d have to be stupidly careful with cable placement, or something, and hope I get lucky.

At that point I thought I’d achieved nothing at all, and had in fact fucked it up somehow – which didn’t make any sense but that’s never stopped me before – until then I poked at my primary input board for semi-unrelated reasons, and HOLY HOPPING CHRIST ON A POGO STICK the difference.

SO MUCH QUIETER so much quieter

Even at max gain, zero input shows up as zero on my main input board. It’s not actually zero – I can hear it – but it’s too far down to light the metres, which puts it at somewhere around -75db.

So at least I didn’t fuck things up. That is a thing which is good to know. I actually achieved my design goal for the rewiring, to a surprising – nay, astonishing – degree of success!

And it did absolutely fuck all for the problem.

goddammit.

I’m still sure AC Power is a problem though, because if I plug the preamp straight into the wall, the RF pick up is REALLY LOUD; through a computer power strip (which filters, some) it’s less; through the battery-backup and power strip, less still. I suppose it’s possible there’s another reason for that, but I don’t know what it would be.

eta: I took apart the preamp, convinced myself of a few problems, maybe improved it, maybe the RFI just calmed down a bit on its own. Who knows? Here’s a recording of a suboptimal cable position after the resoldering. That’s boosted all to hell, of course; at normal recording levels, it’s inaudible, but I hate that it can be found. It’s also still cable-position-dependent. I tested literally every cable I have; this was the most RF-resistant one. If I have it on the right side of the mic stand, it sounds like this. On the left: no audible signal.

eta2: I opened up the case again and twisted the internal side of the AC power lead (9v AC) as tightly as I thought I could get away with. The test afterwards contained far lower RFI than previous tests, but that’s not necessarily meaningful – it’s a different time of day. (All my previous daytime tests have been mid-afternoon.)

That said, for the first time, there was no sign at all of KUOW 94.9, which may be a first. And I didn’t hear BBC World Service in there either. I was picking up KIRO, dimly – but I don’t know whether I was picking up the AM or FM station. Most of the time they have separate programming, but they simulcast during Seahawks handegg, from pregame through post, and that’s what was on. And I had to dig at it to get that much, which is actually good – if I can’t stop it entirely, I can at least make it something you really have to work to find.

So, yeah. Ongoing. I think that twisting the internal AC power lead bumped it down another level. But I’m not sure. It gives me some hope that the RFI chokes I’ve ordered might help.

Fucking ghosts.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

Comments

[angelwolfgeek]

IIRC there's devices specifically for filtering noise for/from powerline networking/X10 devices. Might be worth messing with. Can't recommend anything specifically but it might be a useful thing to look into.

[solarbird]

I've got some RFI filters coming, I'm gonna slap them on EVERYTHING and hope that helps.

But I'm not confident. I took apart the preamp and reconfirmed all the solder points, and managed to convince myself a few weren't optimal and redid them. I also found an actual problem - a stray bit of copper - on the gain control, which I fixed. And I moved a connection point I thought was too close to another one in the original design. This may've helped, or the RF levels may've just dropped while I was working on it. No way to know. I do know that opening the end of the preamp while it's in operation sends the RF signal level WAY up, which implies my unit ground is good. I guess.

Here's a recording of a suboptimal cable position after the resoldering. That's boosted all to hell, of course; at normal recording levels, it's inaudible, but I hate that it can be found. It's also still cable-position-dependent. I tested literally every cable I have; this was the most RF-resistant one. If I have it on the right side of the mic stand, it sounds like this. On the left: no audible signal.

[stickmaker]

Recalling from my ham radio days, I'm wondering if the source is from outside. Are there any transformers near that part of your house? Old ones, especially if they're slowly going bad, can create annoying RFI. If the problem is on only one leg of the three-phase, there's your 60 Hz.

[solarbird]

No, listen to the recording - this is FM radio. Like, you can hear it. Or shortwave. I suspect FM tho'; it was definitely FM earlier in the day.

[stickmaker]

My mistake. I thought you wrote somewhere it was 60 Hz, but looking back I see that was in my head. (In my defense, that is the most common type of RFI.)

I don't see how it could be FM, unless you've got an FM decoder somewhere in the circuit. AM, yes, including shortwave.

I heard at least two stations, one of them possibly the BBC. They're commercial, not ham, too. Or maybe shortwave propaganda stations.

I've had my share of mysterious, spurious signals. One of them in my landline phone was tracked back to the TV cable! I was suspicious when I discovered the coax coming into my house carried enough voltage to give me a distinct tingle. The coax itself wouldn't have leaked, but the things it was connected to could and likely did.

[solarbird]

I don't understand the FM thing either, but it is definitively there. It is, in fact, the most reliably there. I've heard KUOW FM 94.9 specifically - I caught a call sign, so I know it's not some other public station floating around out there.

I have also picked up what I'm still reasonably convinced has to be BBC shortwave. There is a local 24-hour rebroadcast of BBC World Service, but it's digital, and having magical digital decoding is a step too far, even for me. (And KUOW does some FM rebroadcast, but not at that time of day - this was afternoon and early evening, for hours. They don't do that in analogue, just HD3.)

I took the box apart and twisted the internal side of the AC power lead as tightly as I thought I could get away with. The test afterwards contained far lower RFI than previous tests, but that's not necessarily meaningful - it's a different time of day. (All my previous daytime tests have been mid-afternoon.) But that said, for the first time, there was no sign anymore of KUOW 94.9. I was picking up KIRO, dimly - but I don't know whether I was picking up the AM or FM station. (Most of the time they have separate programming, but thye simulcast during Seahawks handegg, from pregame through post.)

So, yeah. Ongoing. I think that twisting the internal AC power lead bumped it down another level. But I'm not sure.

[stickmaker]

Okay, here's a thought, also from my ham radio days. Receivers emit RF. Badly designed receivers can emit a lot of it. The same is true for audio amplifiers. Someone's radio receiver is picking up FM radio signals, amping them up, and somewhere in the process rebroadcasting the formerly FM signal as AM.

One culprit could be a PA system at a gas station or some other business, which plays local radio stations when they're not making announcements. Lots of wiring there to serve as a transmitting antenna. You could even be getting more than one of these at a time.

This would only be noticed by someone working with sensitive equipment, which you are.

Is the interference constant, only during some part of the day, or what?