Through the luminiferous aether

January 26, 2009 3:05 PM

RFInterferenceFilter: This is a continuation of this question. Summarizing: I'm experiencing a loud buzzing noise in my stereo, accompanied by interference from local FM radio stations. I've done more investigation and have eliminated my power as a (direct) cause of the noise.

The original post has a more detailed description/history of my problem. After some investigation with an oscilloscope, I ended up with this image of my AC power superimposed over the noisy preamp output (close-ups here and here). My working theory at the time was that the hiccups at the top and bottom of the sine wave in the AC output were causing the corresponding jumps in preamp output (which would conveniently explain 120Hz buzzing). In order to test this theory, I got my hands on an AC power regenerator and tried powering my preamp with that. The noise didn't go away (and didn't really change). In the course of scratching my head at this development, I also discovered that the noise doesn't go away when I physically unplug the preamp and allow it to run for a few seconds from only the power left in the power supply reservoir capacitors. Of course, the noise goes away when the energy in the capacitors is used up.

So now I'm left with this: 60/120Hz buzzing and FM radio interference when everything is physically isolated from my AC power (ie: preamp has been totally unplugged, driving the inputs of a laptop running from batteries). This leads me to believe that the cause of the noise is RFI/EMI, and that the hiccups in my power are not the cause of my issues, but are instead also caused by whatever is causing the noise in my preamp, which would explain why they are temporally coincident. I'm still inclined to believe that this is somehow power-related, as the hiccups and my AC power are in perfect synchronization, but nothing as simple as the line-noise-sails-through-power-supply that I had initially hoped for.

I'm an engineer, but not an electrical one; I've managed to debug this problem to this point, but I'm pretty much out of my element here. I don't understand enough of the principles at work to figure out this issue past this point or what I can do to try and fix it. Suggestions, please!

posted by strangecargo to technology (7 comments total)

What happened when you plugged in your headphones into the output instead of the speakers? I used to get shortwave BBC on my Creative surround sound speakers back in '99. The only way to stop that was to turn off the built-in amp. The noise went away when I moved.

If its your preamp picking up the signals, is there a way to build a faraday cage around it? (This may not be feasible, IANAEE, IAAME).
posted by defcom1 at 3:16 PM on January 26

The noise is the same regardless of what I have plugged into the output of the preamp (laptop, powered computer speakers, power amp, etc). It doesn't matter what I have plugged into the inputs (or if there's anything plugged in at all). I pretty much have the noise isolated to something happening with the preamp.
posted by strangecargo at 3:22 PM on January 26

You mention, previously, that you live near Sutro Tower - have you explored further in that direction? This page from KQED, which also broadcasts from Sutro Tower, explains some about interference types their broadcast may cause. They suggest an attenuator between your antenna and amp.

You can also read up on community opposition to tower radiation. I have no idea if this will help you or not.

Fwiw, I used to live at Stanyan & Carl and I got terrible radio interference. I chalked it up to the tower, the muni line, and the close proximity to both the fire station and the hospitals.
posted by gyusan at 4:07 PM on January 26

There's a few ways RF can get inside audio gear - power, direct pickup (inside the device), low level lines (preamp & power amp input), and high level lines (speaker output).

Power is rarely the point of ingress - the power supply almost always provides decent filtering, unless the device has an external power supply brick, in which case the RF is usually picked up on the DC / low voltage side of things, and can be treated the same as interference on inputs.

Direct pickup can be a problem with valve gear, though I'd guess not in your case, as the valves should be well shielded by the case.

Low level pickup: most likely. Long input leads (and even short leads, if you're close enough to the transmitter) or unterminated inputs are the point of ingress. What happens if you terminate the preamp inputs with a dummy load (e.g. a 1k resistor right on the connector, with no lead attached)?

High level pickup: signal is picked up through the speaker leads, and can be heard through the speakers.

Low level pickup (input to the preamp, or between the preamp & power amp) is much more likely in situations without a consistent mains earth between devices (e.g. nothing is earthed, or only the preamp or power amp are earthed). Signal earth is 'tied' to the electrical earth via small caps; the caps ensure that RF signals are shunted to mains earth before they can be demodulated by the rest of the circuitry. If something is not earthed to mains earth, then this filtering doesn't occur.

So-called "earth loops" are much rarer than most people think - they mainly occur where the devices are plugged into different power circuits, and so have slightly differing mains earth potentials. Tying the earths together where possible - either by plugging the devices into the same mains wall socket, or tying the equipment earths together directly - is the usual solution; where this isn't possible (e.g. equipment in different locations) an isolation transformer or "braid breaker" is the next step.

FWIW, it's fairly easy to receive FM through a normal AM detector (or the rectifying action of transistors in an amplifier) through a phenomenon known as slope detection).

If all your devices are earthed, try plugging them in to the same power outlet (or power board). If that fails, try wrapping all input and output leads through filter choke rings at each end (ie both the source and preamp ends of the preamp input lead, both the preamp and poweramp ends of the preamp->power amp leads, and the power amp and speaker ends of the speaker leads.

Given that you've also go mains hum in there, I'd bet money that either (a) you've got a power supply going bad in the preamp, or (b) something in the audio chain is not properly tied to a consistent mains earth. If all that fails, then you have to really start investigating to track down how/where the hum & RF is being induced...
posted by Pinback at 4:17 PM on January 26

strangecargo...

Could be RFI/EMI. Could be bad hardware. Could be cables. Could be power-conducted noise.

Eliminate them all sequentially.

I think you've partially eliminated power quality.

Borrow someone else's stereo equipment and see if it behaves this way. If it does, external factors are more likely. If you have other radio equipment, check it out in the same location. Try a different pre-amp. Surely you have friends who have a preamp you can borrow?

Isolate the offending factor by removing what it could be from the mix, one item at a time. That's the surest way to fix problems like this.

I'd suspect you were in an unbelievably strong local field to get this kind of noise in your system, and by that, I mean a few hundred feet of a major transmitter. If that's the case, you may have few options.

One more item... you might want to find and make friends with a ham radio operator in your area. They not only love problems like this, but often they have test equipment that is useful in visualizing the RF environment. (I am a fan of spectrum analyzers, which will tell you pretty quick how strong the local signals are. ) It should take a half hour or so to find one in your circle of friends within one or two degrees of separation. (We're everywhere.)
posted by FauxScot at 6:28 PM on January 26

Does the preamp have a 3-prong (grounded) power cord? Or does it have a 2-prong, or a wall-wart kind of thing?

What I'd do, as a test at least, is to maintain the "unconnected to the world" setup, and then clip on a ground wire from the chassis of the preamp to a known ground.

If the preamp has no real ground, the signals could be "floating". I don't know if that's a real term. Meaning, the design of the circuits most likely has the negatives of the signals all tied together. But this ISN'T tied to a real ground, so it has no real zero reference. Spurious signals getting inside the chassis generate a signal in the negative, causing your interference.

If the thing is made of plastic, you might try creating a Faraday cage for it. Wrapping it in grounded aluminum foil might be enough. (It has to be connected to real ground for it to work.) If that stops the noise, you can go to some electronics hobby store and get a metal project case to mount the preamp inside of.

(I had to do this for an antenna setup- I took an old 3-prong power cord, stripped the insulation, clipped and taped the white and black wires, leaving the green ground. It appears to work, as my tv hasn't yet blown up.)
posted by gjc at 4:07 AM on January 27

Oh yeah, forgot this. Do you have any tv stations in the area on channel 5? Here in Chicago, this station picks up noise like crazy. Whatever frequency it is seems to be a harmonic of all sorts of things. Power tools, ham radio, the occasional CB radio, shredders, funky refrigerators, etc. Tune in an analog channel near that and see what you see/hear on the TV. You might hear buzzing and see white dots or squiggliness in what should be vertical lines.

This would be further evidence that you've got a noisy RF area. Good grounding and shielding should reduce this greatly.
posted by gjc at 4:12 AM on January 27