Make it stop!

August 8, 2006 11:42 AM

Why is my Mac G5 picking up radio signals?

My dual-2ghz G5 (running OS X 10.4.7, FWIW) seems to be sending an unintended audio signal to my amplifier through the built-in soundcard.

I run a standard Monster stereo-miniplug-to-RCA line from the back of my G5 to a Sansui 331 receiver. I started noticing little audio artifacts in between tracks when the receiver is turned up, so I cut the iTunes and cranked the Sansui's volume to max and could plainly hear a Christian radio broadcast.

At first I thought it was bleed-over from my Sansui's radio, but 1) the dial is tuned to NPR, and 2) when I mute the G5 via the keyboard mute button, the radio sound disappears. Strangely, when I lower the G5's volume all the way instead of muting it, (with iTunes, etc not playing) the radio audio doesn't change. So, only toggling the G5's mute on and off makes the radio appear and diappear.

Also, moving the Monster minipug line from port on the back of the computer to the headphone jack on the front makes the radio audio disappear.

So, in summary, I have isolated the source of the radio sound to only the stereo-mini out port on the back of the G5, and I can control it's appearance with the mute button, but on the sound volume buttons.

Has anyone here experienced or heard of this phenomenon? I'm looking for fixes (I have a work-around) but I'm also looking for insight.

Here are the details on the two built-in soundcards. I'm not sure which controls what exactly:

Built In Sound Card:

Devices:
Burr Brown PCM3052:
Inputs and Outputs:
Line Level Input:
Controls: Mute, Master
Playthrough: No
PluginID: Onyx
S/PDIF Digital Output:
Controls: Mute
PluginID: Onyx
Headphones:
Controls: Mute, Left, Right
PluginID: Onyx
Internal Speakers:
Controls: Mute, Master
PluginID: Onyx
Line Level Output:
Controls: Mute, Left, Right
PluginID: Onyx
Formats:
PCM 16:
Bit Depth: 16
Bit Width: 16
Channels: 2
Mixable: Yes
Sample Rates: 32 KHz, 44.1 KHz, 48 KHz, 64 KHz, 88.2 KHz, 96 KHz
PCM 24:
Bit Depth: 24
Bit Width: 32
Channels: 2
Mixable: Yes
Sample Rates: 32 KHz, 44.1 KHz, 48 KHz, 64 KHz, 88.2 KHz, 96 KHz
AC3 16:
Bit Depth: 16
Bit Width: 16
Channels: 2
Mixable: No
Sample Rates: 32 KHz, 44.1 KHz, 48 KHz, 64 KHz, 88.2 KHz, 96 KHz


Built In Sound Card:


Devices:
Crystal Semiconductor CS84xx:
Inputs and Outputs:
S/PDIF Digital Input:
Controls: Mute
Playthrough: No
PluginID: Topaz
Formats:
PCM 16:
Bit Depth: 16
Bit Width: 16
Channels: 2
Mixable: Yes
Sample Rates: 32 KHz, 44.1 KHz, 48 KHz, 64 KHz, 88.2 KHz, 96 KHz
PCM 24:
Bit Depth: 24
Bit Width: 32
Channels: 2
Mixable: Yes
Sample Rates: 32 KHz, 44.1 KHz, 48 KHz, 64 KHz, 88.2 KHz, 96 KHz
AC3 16:
Bit Depth: 16
Bit Width: 16
Channels: 2
Mixable: No
Sample Rates: 32 KHz, 44.1 KHz, 48 KHz, 64 KHz, 88.2 KHz, 96 KHz

posted by squirrel to computers & internet (11 comments total)

If there's a broadcast transmitter nearby it will produce induction in weakly-shielded cabling. We used to live a few miles from the KRLD-AM clear-channel transmitter in Dallas and sometimes I'd pick it up weakly in microphone and speaker runs. The question now seems to be whether better cables or components will fix it.
posted by hodyoaten at 11:51 AM on August 8, 2006

Cabling was my first thought too, but if the problem goes away when using the headphone port it may be something else.

Maybe it has something to do with the built-in amplifier for the stereo out port? Is there an option for line-level out?

Out of curiousity, what kind of surface does your Mac sit on?
posted by polyhedron at 12:42 PM on August 8, 2006

I don't think that there's a broadcasting source very close to my home, but I could be wrong. The diligent can find my general location in my profile. The G5 sits on a short plastic foot stool on top of a wood floor. Thanks for your thoughts so far.
posted by squirrel at 12:46 PM on August 8, 2006

This is the reason i got an external sound box rather than a card. The firewire out is clean even if the analog portion of the computers audio path is hosed. Short of building a faraday cage for your Mac, I doubt there's a way to truly isolate it.
posted by doctor_negative at 1:20 PM on August 8, 2006 [1 favorite]

It's probably a shortwave signal, since most of the english shortwave stations I've managed to tune in have been Christian.
posted by knave at 1:32 PM on August 8, 2006

I'm thinking it is induction from a nearby transmitter as hodayten says. I used to have a porrly shielded set of speakers that would intermittently pick up the christian AM station down the road from me. Every now and then I'd get a burst of "Jesus loves you!" from the speakers (which is pretty scary, as we haven't spoken in some time). I'm betting that some part inside the case is acting as an antenna.

You can get a list of local radio stations here. My stalking skills reveal that there are several close to you, including a high school radio about a mile away (callsign KECG)...
posted by baggers at 1:34 PM on August 8, 2006

Turning the volume all the way down leaves some signal travelling through the cables that can be modulated, while turning mute on kills the signal completely so that the cables can no longer act as an antenna. My 5.1 system (many long audio cables) picks up police radio traffic, among other things.
posted by Derive the Hamiltonian of... at 6:56 PM on August 8, 2006

So do the speakers actually work when plugged into the headphone jack, or does the sound card detect that the impedance is wrong and shut down the output? I'm putting all my money on the cables acting as an antenna and saying that the different properties of the sound card on mute or low volume and the headphone port are determining whether the baseline signal through the cables is strong enough to result in audible modulation and amplification.
posted by Derive the Hamiltonian of... at 7:02 PM on August 8, 2006

Virtually any non-linearity in the circuit has the potential to demodulate AM signals. The mathematical explanation is actually pretty straight forward - if you have an open mind about math, that is. If the words taylor series and double angle formula have already made you lose concentration, it may not be worth going any further..

You also need an antenna, to pick up the signal in the first place, and a filter to emphasize one AM station over the others. These things can happen accidentally in all kinds of electronics.

Try a different audio cable, but the antenna is just as likely to be something inside the laptop, I think. See if running off battery or off the AC adapter makes a difference - power supply noise is often the culprit. I might be overlooking some other simple ideas..

It's possible that a non-critical component has failed (like a particularly important bypass capacitor falling off, or who knows what else), but unless you are relatively skilled with circuits and soldering iron, there isn't much you can do about it..
posted by Chuckles at 9:40 PM on August 8, 2006

For some reason I thought it was a laptop, sorry..

There are a few other things worth trying, actually.. In the end, it is basically an EMI problem, so try looking through the suggestions in the sound of silence.

And.. Try disconnecting all cables except the one leading to the amplifier. This might even include pulling the keyboard & mouse cables as soon as you have the experiment set up. It definitely includes pulling the video cable, and turning off the monitor.
posted by Chuckles at 9:55 PM on August 8, 2006

Thanks for all the tips and conjecture, y'all. I didn't know that there was a difference between the high-level signal put out by the "sound out" mini jack on the back and high-level signal put out by the "headphone" mini jack on the front. They sound the same through my amp.
posted by squirrel at 8:06 AM on August 9, 2006