Getting Big AM Sound — Correctly

To the detriment of AM, almost gone is the old crusty chief engineer who cut his teeth on an RCA BTA 1 and knew the difference between a tank circuit and a swinging choke.
Author:
Publish date:

Some AM stations have a stellar sound: clean, bright, natural and loud. Some sound so bad you wonder if anyone in management ever listens at all.

To the detriment of AM, almost gone is the old crusty chief engineer who cut his teeth on an RCA BTA 1 and knew the difference between a tank circuit and a swinging choke. Today's younger chief engineers probably knew computers before transmitters and transmitters were more FM than AM.

In FM, engineers strive to keep the 100 percent modulation lights flashing as much as possible without overmodulation. Those who insist on the same for AM need to read on, and get educated.

Those of us who have a passion for AM cringe at the sound some stations pass off as broadcast quality. Sure AM has problems, poor signal-to-noise in many areas and of course the cesspool called IBOC — don't get me started on that again — but the medium can sound decent even under present constraints.

Too many engineers get the latest and loudest box, put it inline and crank it up till all the peak lights come on and that's it. Wrong! In order to set up modulation correctly you need to understand the characteristics of your type of transmitter.

(click thumbnail)This shows the so-called cliff effect that occurs with some PDM transmitters. You can easily see the cliff that is formed as the negative modulation fails to properly track as it nears cutoff. The sharp 'cliff' causes audible distortion. The graphic is not to scale. The orange line normally appears at about 96 percent on some PDM transmitters.
One hundred percent negative and 125 percent positive are legal limits, not necessarily obtainable goals! There are still some plate-modulated transmitters around, many designed and built in the late '60s or early '70s. These behemoths modulate well as long as both the modulator tubes and the RF finals are fresh.

Aging tubes

As these tubes age they lose ability to handle high levels of modulation and often, even when fresh, their power supplies can not handle the demands of positive peaks over 100 percent at all. Back then, the legal limit for positive modulation was 100 percent.

Despite those drawbacks, some AM neophytes insist on trying to hit 125 percent modulation because it's the legal limit, with terrible results. The audio is bad and system components are taxed to early failure.

Back it down! The chokes, tubes, modulation transformer and power supply will thank you and the sound will be a lot better. If you have one of these old timers and use modern processing I suggest you set peaks at 86-90 percent negative and positive, which is still legal.

For those with Pulse Width Modulation transmitters, more common today, the story is different. It's the negative modulation that can be a problem.

If you're using a high-performance processor like Omnia or Optimod you may find the modulation sounds a little edgy like the clipper is too high. Before thinking your $5,000 processor is bad, check your modulation settings.

Some PWM transmitters are prone to cliff effect. They can't handle extremely high negative modulation without running away to nasty-sounding cutoff.

To see this effect, use a 1 kHz tone and modulate at about 93 percent negative. Then slowly raise that modulation level and watch the monitor. If you see it get to maybe 95 or 96 and then jump to 100 percent then you have cliff effect.

This occurs as the result of asymmetric antenna loading to the final PA. The PDM 70 kHz filter cannot properly handle the reflected energy. The effects are more noticeable with high frequency modulation. Usually simple phase rotation can improve that.

My remedy is simple. Set the maximum negative modulation for just under 95 percent. I personally set it at 93 percent. The sound will be much cleaner and the loudness won't really suffer.

And you will find you can push the processing harder if you want. You should never see a flash from the 100 percent negative indicator.

PWM transmitters do an excellent job of passing positive peaks but personally I like the sound of symmetrical modulation so I set both positive and negative peak at under 95 percent. It might surprise you to know the most successful AM stations in the past and present were never known for beating modulation peaks to death.

Top 40 king CKLW in Windsor, Ontario, back in the '60s used a lot of RMS compression but set modulation peaks at 85 to 89 percent. WGN, Chicago's current big AM biller, uses only moderate processing for a non-fatiguing warm and pleasing sound.

Then there is the receiver factor. Most cheaply designed AM receivers show audible distortion when hit with extremely high peak modulation levels.

In the end we must remember it's what the listener hears that counts. Take a cue and go for quality vs. maximum peaks at any cost. It pays off in the long run.

Image placeholder title

Related