What's the deal with AM?

[N9IWP]

I feel compeeled to point out that one could use AM modulation at FM frequencies. Aviation communication frequencies are pretty close to FM (my local airport uses 122.9 MHz) but use AM modulation. Probably partly historical, but also (IIRC) due to issues with FM and moving/high altitude vehicles (“picket fencing” - the direct radio wave cancels out the wave that bounces off the earth)

You can have FM at lower frequencies as well, but FM in general requires more bandwidth than AM, and channel spacing becomes and issue. (not much FM below 25MHz)
Brian

[Antonius_Block]

@CurtC

The noise is not due to signal drift, and it’s not harder to lock onto an AM signal than an FM one.

I disagree with the second part of this.

As David Simmons said:

The output voltage of an FM detector depends upon how far the received frequency deviates from the center, or carrier, frequency of the wave.

Since any commercial FM transmission will be “DC-blocked” (i.e. positive frequency deviations will be balanced by negative ones in the transmitted signal, when measured over a period of hundreds of milliseconds), a low-pass filter on the output of an FM detector will provide an error signal which can be used to keep the receiver in lock. Typically this error voltage will alter the bias on a varicap diode in the Local Oscillator circuit, changing the LO frequency until the FM detector output’s low-frequency error returns to zero, thus keeping the tuner locked to the incoming FM transmission.

The above method has been used for decades, and provides a simple, relatively-cheap analog frequency-locked-loop to overcome drift in the radio’s tuned circuits. There is no equally simple equivalent analog technique for AM signals, which is why older AM radios tend to drift.

Modern digital frequency-generation techniques have essentially made drift a thing of the past for AM as well as FM, but the OP specified a non-digital tuner.

[Jon_the_Geek]

So, would a slightly higher-end tuner likely result in lower static? Would digital be more likely to hold the signal than analog? Is there anything in particular that you can do on the receiver end to get a stronger signal (larger antenna, more power, etc)?

[Askance]

@CurtC

1. The simplest is that common noise sources (engine sparks, electric motor brushes, and lots more) put out less and less of their noise at the higher frequencies. If you looked at the spectrum of that noise, they’re larger at lower frequencies and lower at higher ones. The AM band is around 1 MHz - FM is around 100 MHz. Big difference right there.

Does that also explain why the talk stations I now listen to, at frequencies such as 576, 630 and 702, are so vulnerable to interference and signal loss as I drive around the city? Where IIRC the ones at the other end of the AM dial like 1170 and 1269, that I used to listen to in my youth (when there was no FM), didn’t have so much trouble?

[David_Simmons]

@Jon the Geek

So, would a slightly higher-end tuner likely result in lower static? Would digital be more likely to hold the signal than analog? Is there anything in particular that you can do on the receiver end to get a stronger signal (larger antenna, more power, etc)?

It is difficult to improve AM reception at the frequencies used for that purpose. The problem is that AM noise is nearly 100% atmospheric noise and noise from inferfering transmitters such as auto ignition, neon signs, lightning, etc., etc. and all such signals arrive at the receiver input via the antenna. So the only way to improve the signal to noise ratio is to use a more directional antenna. Such an antenna discriminates against noise that isn’t coming from the same directions as the transmitter you want to receive. For AM broadcast frequencies such an antenna is the size of the transmitting towers that you see for AM stations. Furthermore if you want to listen to more than one station the antenna has to be steerable. About the best you can do for AM is a loop antenna and most commercial sets already use a ferrite rod loop antenna except for automibiles.

[David_Simmons]

@Antonius Block

@CurtC

The noise is not due to signal drift, and it’s not harder to lock onto an AM signal than an FM one.

I disagree with the second part of this.
.

I don’t think it is harder to lock on to an AM signal than to an FM signal in principle. The difficulty is that a frequency detector has to be added to the AM receiver while the FM receiver already includes one as part of its normal modulation detection process.

FM does have the characteristic that under some conditions there isn’t much carrier to lock on to. However I doubt this is any real problem in broadcast FM.

[engineer_comp_geek]

There’s a bit of misinformation in this thread. The static on AM is there almost exclusively because the signals are amplitude modulated, not because of the particular frequencies used. AM modulates the amplitude, which is kinda like the volume control, of the radio wave. When two different radio signals are added together, their amplitudes are simply added. This means that any noise caused by spark plugs, the sun, etc. becomes part of the signal and cannot be removed from it. The best AM receiver in the world isn’t going to be able to pick out the static from the signal, because there’s just no way for it to tell what is noise and what is signal.

Frequency modulation, in contrast, works by changing the frequency. The detector locks onto the signal and tracks it as it moves up and down in frequency. The detector is going to lock onto the strongest frequency it can find in the band of frequencies it is looking in, so just by the nature of the modulation it is going to completely reject weaker noise signals.

You do get static occasionally on FM signals, but this is due to the fact that a noise or possibly another broadcast signal has managed to get strong enough that the detector is sometimes tracking one signal and sometimes tracking the other. This gets to be especially annoying when you have two stations of similar strength, and the detector bounces back and forth between which one it happens to lock on to.

FM, since it tracks the frequency, also doesn’t care too much if the signal gets stronger or weaker. Stronger or weaker is the amplitude of the signal, so if the signal gets weaker on an AM signal the audio seems to “fade” in and out.

Aside from the noise and fade problems inherant in amplitude modulation, AM signals also sound crappy because the frequency range they use is limited by the FCC, which means they have an effective bandwidth of about 10 kHz, IIRC. For reference, the human hearing range is somewhere around 20 Hz to 20 kHz, so this bandwidth limit really cuts into the high frequency range that people can hear. AM radio signals are also set up to be one audio channel only, which means you don’t get stereo sound (stereo is more pleasing to the ears). You could theoretically broadcast AM in stereo, and you could increase the frequency allocations (though you would need more total bandwidth set aside to do it), but you still aren’t going to get around the noise/static problem.

FM radio is set up by the FCC as stereo with a bandwidth of about 15 kHz, which sounds a lot better than AM but is still a step down from CD quality.

Back in the 70’s I remember there being both AM and FM stations, but by the end of the 70’s most of the stations had moved to FM. Not at all coincidentally, good quality radio receivers were cheap and common by then too. For a while it looked like AM was quickly becoming a dead band. The FCC considered some proposals to make AM stereo stations and increase the bandwidth to try and revive what was otherwise a bunch of wasted frequencies. However, in the late 80’s and early 90’s talk radio started really taking off. Voice signals don’t need quite so much bandwidth to sound acceptable (compared to music), so many stations started broadcasting talk radio on AM (which meant they didn’t have to fight for space on the crowded FM band in large metro markets). Since the AM band is now used quite a bit, the FCC is no longer seriously considering making any improvements to it.

[engineer_comp_geek]

@Birdmonster

On the same topic, I have a bass amp that sometimes picks up radio frequencies when it’s on and the chord is in the amp but not the bass. Is that AM or FM?

AM.

All you need for an AM receiver is a tuned circuit to filter off the frequencies you don’t want, something that conducts better in one direction than the other. It’s very difficult to not accidentally make an AM receiver out of most electrical circuits. This is especially true when amplifiers are involved, since they can take a really weak signal that accidentally got coupled into the audio path and amplify it into a really strong annoying signal.

[Jon_the_Geek]

Ok, a better antenna won’t help, but will a digitally-tuned receiver (or some other more-high-tech-than-my-crappy-old-radio solution) have a better chance of holding the “real” signal, at least preventing the drift? I doubt I’ll actually upgrade, but I’m curious if it’s possible to alleviate some of the trouble I’m having listening to talk radio; if it’s all at the transmitter (or in the intervening miles), or if it’s partly the transmission and partly my crappy receiver.

[engineer_comp_geek]

AM signals don’t really “drift” as in change in frequency. They do fade in and out a bit due to things like atmospheric conditions, reflections and interference patterns off of objects as you move, etc. which there is nothing you can do about.

If you are getting frequency drifting, then one of two things is happening.

First, it could be that you weren’t tuned directly to the center of the frequency you were listening to, and as the signal gets weaker you need to adjust the dial so you get more towards the center of its broadcast frequency. It may be that if you are close to the broadcasting station that your receiver saturates so it doesn’t sound right if you are tuned exactly to the right frequency, then as you move away the signal gets weaker and you have to adjust it.

The second thing that could be happening is that your radio’s tuner is drifting, not the station you are listening to. Every old fashioned analog type radio I’ve ever taken apart used a simple plate type variable capacitor for the tuning. This device is just two sets of metal plates. As you move the plates closer and farther apart you change the resonant frequency of the tuned circuit, which is what selects which station you are tuned to. The problem with metal plates is that their size and spacing is going to change as the radio heats up. Since their size and spacing effects the radio receiver’s frequency, this may force you to adjust your tuner.

A digital tuner will help with both of these problems. First, it’s going to automatically tune to the center of the broadcast frequency. You don’t have a choice in this. Second, it’s going to use a different type of receiver circuit which is a bit better at rejecting unwanted signals and generates less internal noise itself, which is going to make for a better signal in general.

Don’t expect it to work miracles though. It’s still AM, and there’s just no getting away from static and fading on AM.

[Jon_the_Geek]

@engineer_comp_geek

The second thing that could be happening is that your radio’s tuner is drifting, not the station you are listening to. Every old fashioned analog type radio I’ve ever taken apart used a simple plate type variable capacitor for the tuning. This device is just two sets of metal plates. As you move the plates closer and farther apart you change the resonant frequency of the tuned circuit, which is what selects which station you are tuned to. The problem with metal plates is that their size and spacing is going to change as the radio heats up. Since their size and spacing effects the radio receiver’s frequency, this may force you to adjust your tuner.

That sounds like a very likely candidate for the problem. Can you tell me which direction the signal should seem to drift in as the plates heat up (or will it depend on the exact setup of the plates)? The signal “moves” toward 1590 (meaning I have to tune to the left to find the signal) at night, which would be when my radio is hottest. When I listen as I get ready for work in the morning, I have to tune back to the right to get the station. If I can find a way to do it that won’t involve more work than curiosity can impel me to put forth, I’ll experiment to see if I can confirm (or, well, fail to rule out) heat as the culprit… mostly because I’m curious now

The “drift” is what pushed me over the edge to the point of posting this thread, so thanks a ton for giving me a logical explanation for that particular problem

[RaftPeople]

Assuming my electronics book was accurate, another point is that FM is a ground wave (whereas AM bounces off the atmosphere) which means you get reception in tunnels, etc.

[Tabby_Cat]

Er, what? You most definately do not get FM in tunnels. Personal experience.

[Exapno_Mapcase]

This could be another U.S.-centric item, but one big advantage here of FM is that FM does not cut out the way AM does. We don’t have any mile-long tunnels around here, but underpasses and the like are no problem.

What frequencies are your FM stations?

[Tabby_Cat]

Oh, well, I don’t have any problems with underpasses either, but once you get about 200m or so into a tunnel, no more reception unless there are repeaters installed.

The frequencies range from about 88.3Mhz to about 100Mhz.

[David_Simmons]

@RaftPeople

Assuming my electronics book was accurate, another point is that FM is a ground wave (whereas AM bounces off the atmosphere) which means you get reception in tunnels, etc.

Not exactly. At AM frequencies during the daytime radio waves propagate only by ground wave. This limits good reception for a 50 kW, clear channel station to maybe 100 miles with adequate reception to maybe 150-175 mi. At night the sky wave enters the picture and at distances beyond about 100 miles or so that is the primary mechanism.

At FM frequencies the sky wave never enters the picture because it isn’t reflected by the ionosphere.

I’m not sure of the mechanism that attenuates radio waves in tunnels. It might be that of a waveguide smaller than cut-off wavelength. In order to propagate down a rectangular or square tube with conductive walls the tube must be at least 1/2 wavelength wide. For example, at the highest AM broadcast frequency of 1.5 mHz the wavelength is 200 meters. Such a wave will not propagate very far down a tube smaller than 100 meters (328 ft.) wide. The lowest FM broadcast frequency is 88 mHz with a wavelength of 3.4 meters. A guide to transmit such a wave only has to be about 1.7 m (~ 5.5 ft) wide.

The walls of a tunnel are not perfectly conducting and might allow transmission of the wave better than a true wave guide. However, the AM wavelengths are so far beyond cut-off that they are eliminated almost immediately.

[CurtC]

@Jon the Geek

That sounds like a very likely candidate for the problem. Can you tell me which direction the signal should seem to drift in as the plates heat up (or will it depend on the exact setup of the plates)?

He was talking about very old radios with parallel plates. More modern ones use some different technique, like a varactor or something.

[Jon_the_Geek]

What qualifies as “very old”? I’m guessing mine isn’t old enough, but the description fit the problem so well that it seemed likely…

So, what’s a varactor? How old would a radio have to be to be likely to not have one?

[Spectre_of_Pithecanthropus]

@FlippyFly

1 – The signal can travel farther… you can get an AM signal for sometimes as much as 120 miles, where FM seem to only go about 30-60 (depending on the power of the station of course). AM signals tend to bounce around in the atmosphere more, therefore they travel further.
.

I’ve noticed that very late at night or very early in the morning the AM signals seem to bounce remarkably far, and in L.A. I’ve picked up stations in Utah and Colorado. This was always while driving outside the city, since I don’t listen to AM elsewhere. Within L.A. itself, a few minutes after the 1994 earthquake, when the electricity was out, basically citywide, and there was virtually no radio stations broadcasting, I was able to get the San Francisco KNX affiliate on my car radio; it was the only way I could find out any news, beyond the obvious, of course.

[engineer_comp_geek]

@Jon the Geek

What qualifies as “very old”? I’m guessing mine isn’t old enough, but the description fit the problem so well that it seemed likely…

So, what’s a varactor? How old would a radio have to be to be likely to not have one?

Here are a couple of pics of the variable capacitor I was talking about.

oselectronics.com

Note that there is one set of plates that stays stationary while another set is rotated closer or farther away.

A varactor is a type of diode where the capacitance of the device varies according to the reverse bias voltage across the diode. For you non-techy folks out there, it’s a little semi-conductor thingamabob that you can vary it’s capacitance electrically.

Which ever type you have, the capacitor is part of a tuned circuit along with an inductor (which is just a coil of wire). The resonnant frequency of the circuit, which determines what station you are tuned to, varies according to the capacitance and inductance of the components (google “RLC bandpass filter” if you want more gory details).

I’m betting that you have a plate style variable capacitor in your radio, but I could be wrong.