The Spirit of Radio
May 11, 2011 8:15 AM Subscribe
The science behind FM transmitter behaviour... how does it really work and what's going on?
So, simple enough.. you pick a station and the transmitter pushes your music from your iPod on that wavelength, and the power and proximity to your receiving antenna overrrides any other signals.
So, what I can't seem to find is why the behavior seems so random.
Regular setting recommendations is to find a FM frequency setting that doesn't have a station broadcasting. So, I pick two stations with nothing but static, A and B.
Station A broadcasts my music without a touch of static or hiss. Yet station B is all static-y. Yet there are no stations near either A or B on the radio dial.
Now, stations C and D have strong, perfectly good radio stations. Trying to play through station C's signal is a mess - fade in, out, static. But station D I completely overrride and my music plays flawlessly.
And why do people recommend using the minimum or maximum frequencies?
So, what's going on?
So, simple enough.. you pick a station and the transmitter pushes your music from your iPod on that wavelength, and the power and proximity to your receiving antenna overrrides any other signals.
So, what I can't seem to find is why the behavior seems so random.
Regular setting recommendations is to find a FM frequency setting that doesn't have a station broadcasting. So, I pick two stations with nothing but static, A and B.
Station A broadcasts my music without a touch of static or hiss. Yet station B is all static-y. Yet there are no stations near either A or B on the radio dial.
Now, stations C and D have strong, perfectly good radio stations. Trying to play through station C's signal is a mess - fade in, out, static. But station D I completely overrride and my music plays flawlessly.
And why do people recommend using the minimum or maximum frequencies?
So, what's going on?
B might not have worked because there was strong interference there that wasn't from a radio station. So while your radio wouldn't have been able to make anything out of this interference on account of it being random uncorrelated noise (and thus seem like there was nothing there), it still might be strong enough to lower the SNR of your transmitter to the point where it sounded bad. And the station at D could have been low enough power that there was still significant SNR for the radio to pick up the station, but not enough that it could overcome your local transmitter.
posted by Rhomboid at 8:46 AM on May 11, 2011
posted by Rhomboid at 8:46 AM on May 11, 2011
FM radio has something called capture effect, which means the strongest FM signal (at the receiver) takes precedence. Signals of roughly equal strength will exhibit the symptoms you describe with respect to station C (weird mixing, fading back and forth, etc.).
You should also keep in mind that just because there is no nearby broadcast station on a certain frequency (e.g., station B) doesn't mean there isn't some sort of signal which is interfering. When signals of two different frequencies mix, they can produce interference on a third (or fourth) seemingly-unrelated frequency. There could be signals in your area completely outside of the FM radio band which are mixing to produce interference on frequency B. Such interference might be demodulated by your FM receiver as noise/static, but it could in fact be stronger than the very weak signal put out by your iPod transmitter.
As for station D, just because it sounds "strong" doesn't mean the signal is necessarily strong, which is why your little iPod transmitter can easily override it. Keep in mind radio signal strength falls off proportionally to the square of the distance. Every time you double your distance from the transmitter, the signal is only 1/4 as strong as it was before.
posted by Nothlit at 8:48 AM on May 11, 2011 [1 favorite]
You should also keep in mind that just because there is no nearby broadcast station on a certain frequency (e.g., station B) doesn't mean there isn't some sort of signal which is interfering. When signals of two different frequencies mix, they can produce interference on a third (or fourth) seemingly-unrelated frequency. There could be signals in your area completely outside of the FM radio band which are mixing to produce interference on frequency B. Such interference might be demodulated by your FM receiver as noise/static, but it could in fact be stronger than the very weak signal put out by your iPod transmitter.
As for station D, just because it sounds "strong" doesn't mean the signal is necessarily strong, which is why your little iPod transmitter can easily override it. Keep in mind radio signal strength falls off proportionally to the square of the distance. Every time you double your distance from the transmitter, the signal is only 1/4 as strong as it was before.
posted by Nothlit at 8:48 AM on May 11, 2011 [1 favorite]
Another possibility is the antenna on the transmitter itself.
The antenna on these things is often just a trace on the PCB or a short bit of wire. The wire or PCB trace has a resonant frequency where it works best; one of your ABCD choices might be hitting that and working really well, while another may be not as good. Combined with other factors, you get this hard-to-pin-down behavior.
posted by fake at 8:57 AM on May 11, 2011
The antenna on these things is often just a trace on the PCB or a short bit of wire. The wire or PCB trace has a resonant frequency where it works best; one of your ABCD choices might be hitting that and working really well, while another may be not as good. Combined with other factors, you get this hard-to-pin-down behavior.
posted by fake at 8:57 AM on May 11, 2011
FWIW, I have modified several to boost the output signal. FCC limits the power level to 3 milliWatts (into the antenna), I think. The chips can output a lot more, on the order of 100 mW. You have to know where to find the attenuator parts and how to remove them from the circuit, and it doesn't sound like that's something you know. (I also have a lot of measurement gear and can see these signals, imitate them, and measure their strength.)
I will tell you that even though these modified systems have higher power and work undeniably better, it's one of those 'tragedy of the commons' things that would be a mess if everyone did it, so I will keep my superpowers secret except to the local mefi crew of EEs.... (you know who you are!).
If your unit plugs into the cigarette lighter, the cord that goes between the ipod and your lighter socket IS the transmitting antenna usually. Drape it differently and you'll get stronger/weaker signals to the radio. It might also explain why at one instant you get good signal and a minute later you get a properly polarized signal from a co-channel transmitter and the capture effect mentioned above kicks in. Also note that some car antennas have integral amplifiers and that FM is line-of-sight, more or less. Exact situation as before, except in an area (like a coastal metro area), you can come into/out of sight of stations a long way off pretty capriciously as you drive.
I recently modified both my car radios to accept 'aux' inputs. Much better alternative as you can quit dealing with flaky transmitters for good.
posted by FauxScot at 10:33 AM on May 11, 2011
I will tell you that even though these modified systems have higher power and work undeniably better, it's one of those 'tragedy of the commons' things that would be a mess if everyone did it, so I will keep my superpowers secret except to the local mefi crew of EEs.... (you know who you are!).
If your unit plugs into the cigarette lighter, the cord that goes between the ipod and your lighter socket IS the transmitting antenna usually. Drape it differently and you'll get stronger/weaker signals to the radio. It might also explain why at one instant you get good signal and a minute later you get a properly polarized signal from a co-channel transmitter and the capture effect mentioned above kicks in. Also note that some car antennas have integral amplifiers and that FM is line-of-sight, more or less. Exact situation as before, except in an area (like a coastal metro area), you can come into/out of sight of stations a long way off pretty capriciously as you drive.
I recently modified both my car radios to accept 'aux' inputs. Much better alternative as you can quit dealing with flaky transmitters for good.
posted by FauxScot at 10:33 AM on May 11, 2011
Also, the cord as antenna often likes to use your body as part of the antenna. I frequently can get a much stronger signal by simply maintaining contact with the cord as I drive. I simply loop part of the cord around my wrist or under my leg.
posted by COD at 11:49 AM on May 11, 2011 [1 favorite]
posted by COD at 11:49 AM on May 11, 2011 [1 favorite]
COD: "The frequencies at the ends of the dial tend to be less frequently used by powerful stations, so your odds are better of getting a clean signal. If you transmitter can go into international mode, you can access 87.5 and 87.7, which are not used in the US at all. I have great luck keeping a clean signal on 87.7. You'll have to use Google to figure out if your transmitter can do it, as it is usually an undocumented feature."
My old 1994 cavalier had 87.7 and so did my transmitter. It was (mostly) great. A little interference now and then, but mostly not a problem. I just got a 2003 car, and was utterly disappointed to see no 87.7, and all the other channels suck. My friend pointed out that 87.7 in the US isn't considered FM.
So it's true that I might be able to find a signal that is "strong" and override it? I may have to try that, because I miss using my mp3 player and only using CDs.
I've seen a kit that taps in a line between your antenna and radio, and switches the antenna off with a switch and adds an aux input. Someone told me that it still would suck, because it's feeding through the antenna, even though it "turns off" the outer reception it can still have problems. Is that true?
Finally - do most FM transmitters have issues at stoplights? I noticed that certain stoplights seemed to really interfere and I could hear buzzing and switching when the lights changed.
PS - sorry for threadjacking.
posted by symbioid at 2:07 PM on May 12, 2011
My old 1994 cavalier had 87.7 and so did my transmitter. It was (mostly) great. A little interference now and then, but mostly not a problem. I just got a 2003 car, and was utterly disappointed to see no 87.7, and all the other channels suck. My friend pointed out that 87.7 in the US isn't considered FM.
So it's true that I might be able to find a signal that is "strong" and override it? I may have to try that, because I miss using my mp3 player and only using CDs.
I've seen a kit that taps in a line between your antenna and radio, and switches the antenna off with a switch and adds an aux input. Someone told me that it still would suck, because it's feeding through the antenna, even though it "turns off" the outer reception it can still have problems. Is that true?
Finally - do most FM transmitters have issues at stoplights? I noticed that certain stoplights seemed to really interfere and I could hear buzzing and switching when the lights changed.
PS - sorry for threadjacking.
posted by symbioid at 2:07 PM on May 12, 2011
Someone told me that it still would suck, because it's feeding through the antenna, even though it "turns off" the outer reception it can still have problems. Is that true?
Yes, it's true. The FM modulation scheme limits frequency response to about 15 kHz and channel separation of about 40 dB. You also lose quality because the signal goes through a pre-emphasis filter, a stereo encoder, an IF modulator, an RF modulator, an RF demodulator, an IF demodulator, a stereo decoder, and finally a de-emphasis filter. Each of the first four is undone by the last four, resulting in the same signal coming out as went in. But that's 8 stages of totally extraneous analog manipulation of the signal which adds noise and distortion, on top of the limited bandwidth and finite channel separation. You skip all of this by just hooking directly into a line level input. Now, of course, this is a car and there's lots of noise to begin with and not everyone has golden ears so maybe you'll be fine with this; a lot of people are content with FM quality sound. But from a purely engineering standpoint, feeding audio over FM is hugely suboptimal over just sending it directly.
posted by Rhomboid at 2:36 PM on May 12, 2011
Yes, it's true. The FM modulation scheme limits frequency response to about 15 kHz and channel separation of about 40 dB. You also lose quality because the signal goes through a pre-emphasis filter, a stereo encoder, an IF modulator, an RF modulator, an RF demodulator, an IF demodulator, a stereo decoder, and finally a de-emphasis filter. Each of the first four is undone by the last four, resulting in the same signal coming out as went in. But that's 8 stages of totally extraneous analog manipulation of the signal which adds noise and distortion, on top of the limited bandwidth and finite channel separation. You skip all of this by just hooking directly into a line level input. Now, of course, this is a car and there's lots of noise to begin with and not everyone has golden ears so maybe you'll be fine with this; a lot of people are content with FM quality sound. But from a purely engineering standpoint, feeding audio over FM is hugely suboptimal over just sending it directly.
posted by Rhomboid at 2:36 PM on May 12, 2011
This thread is closed to new comments.
posted by COD at 8:28 AM on May 11, 2011