Well, I was bored...so this is what you get //content.invisioncic.com/y282845/emoticons/smile.gif.1ebc41e1811405b213edfc4622c41e27.gif
I got curious about the oft-mentioned problem of Class D and other switching-based amplifier topologies (where the active devices are turned on and off rapidly) causing interference to nearby radio receivers. Since I've got the setup to do some tests of this, I ran a few quasi-experimental trials, just to see what happens for my install. YMMV, as always.
Little background info: the two amplifiers used are a PG Tantrum 1200.1 and Tantrum 500.2, both securely mounted and grounded to the metal wall of the vehicle. The T1200.1 is a Class T amplifier, and the 500.2 is a Class A/B. The radio used was an Icom 706MKIIG communications receiver/transmitter, with all preamplifiers and filters turned off. The squelch was left wide open. A remotely-tuned 102" stainless steel whip mounted on the rear quarterpanel is the antenna for 50MHz and below, and a 19" vertical whip mounted on the roof is used for 144MHz and up. Both antennas are mounted through the exterior panels for the best RF grounding.
With the amplifiers off, I started by recording the background noise for each of the frequency ranges in question:
HF (1.8 MHz, 160 meters)
low-band VHF (50 MHz, six meters)
high-band VHF (144 MHz, two meters)
UHF (446 MHz, 70 centimeters)
Each frequency was measured in several modes - Single Sideband (SSB), FM, and AM. For each measurement, I was recording the signal strength on the receiver - expressed in a non-standard value called an "S-unit" (no correlation to G-unit /badpun //content.invisioncic.com/y282845/emoticons/wink.gif.608e3ea05f1a9f98611af0861652f8fb.gif) Each S-unit on the receiver's meter indicates 6 more decibels of signal strength - so a S5 signal would be 6dB stronger than an S4 signal.
Results:
For the 160 meter band, it didn't matter if the amps were turned on or off. The background noise level remained constant through all modes. Of course, on 1.8 MHz, a nine foot whip is maybe 2% efficient. Not a good antenna by any means; way too much radiation resistance. That's why good AM reception mobile is tough to come by...a 300-foot long Beverage antenna is nice to have for transatlantic AM reception, but you can't drag it behind you on the freeway //content.invisioncic.com/y282845/emoticons/biggrin.gif.d71a5d36fcbab170f2364c9f2e3946cb.gif
The low-VHF range is where things got interesting. While the amps were turned off, I recorded a S0 background noise level, but when the amps were switched on, I immediately got broadband hash-like interference all throughout the band. In SSB, the noise was a solid S8; much the same for AM. In FM, however, the interference was S9+20dB - enough to blot out any other signals that might happen to be on frequency. I tuned around to see if the noise got worse, and I found several peaks where the noise was full-quieting on the receiver - basically, completely saturating the radio's front end. Switching in the receiver's DSP helped some, but the S/N ratio sucked.
Both high-band VHF and UHF were more immune to interference.
Now, there were tons of variables at work - the antenna used for 50MHz and below is mounted on the same side as the amplifiers, and even though there's a metal skin seperating the two, it's a nine foot antenna compared to the 19" stubby whip that's on the roof. The increased capture area probably contributes to the higher interference potential. I noticed the noise level dropped significantly above 60MHz (where the roof mounted antenna took over) so I wouldn't be surprised if simply changing the position of the antenna is all that's needed.
Take what you want from this...I realize that 90% of people out there aren't going to have as much or as sensitive radio equipment in their vehicles as this, but the same principles apply if you're receiving radiated interference through your head unit or wireless CD changer. Good RF grounding by using mounts that drill through a metal panel is imperative, and making sure the amps are well-grounded to avoid spurious radiation from the interconnects or speaker wires.
