In article ,
wrote:
Hold on a minute here. You can't just throw 2 signals through the IF strip and then the limiter. the
signal would be a mess and neither would be intelligible.

The signal does tend to be a mess, but you can end up with semi-
intelligible audio, especially if you're moving, and the relative
strengths of the two signals are varying a lot.

Am I wrong here ? Two carriers ? Just how would that work ?

Well, here's how I see/understand it.

Both RF signals come through the front end and the mixer. The signal
entering the first part of the IF strip is a down-mixed (10.7 MHz)
combination of the two, with the relative IF strengths of the two
being proportional to the corresponding RF signal strengths.

As this "combination" signal goes through the first part of the IF
strip, it's amplified linearly... nothing distinguishes the two
signals.

The interesting thing happens when the amplified IF signal is strong
enough to limit. This is where the stronger signal wins (usually).

If one RF/IF signal is a lot stronger than the other (say, 10 dB or
more), it's the one which limits. The detector/discriminator "sees"
only this signal - the other one's contributions are too small to be
detected - the detector is fully "captured". You hear the audio from
this signal.

If the difference is smaller (say, 2 dB), the stronger signal
limits and "captures" the detector and you hear its audio. However,
the presence of the weaker signal in the detector input can
perturb the detector enough to have some audible effect... an increase
in distortion, or a faint "growling" sound. In the ham FM-repeater
community we call this "doubling". As the difference between the
signals grows less, and you fall towards the detector's "capture
ratio threshold", the doubling growl/distortion grows worse.

When the two signals are of nearly the same strength, neither signal
is detected properly... you either hear silence, or nothing but
doubling growl/buzzing.

Now, consider the case the OP was talking about, where he's driving
(not stationary). Under these conditions, the RF paths from the
transmitters to his radio are constantly changing. It's quite common
for there to be a large amount of signal-strength variation, from
moment to moment... variations of 10 or 20 dB or more are not
uncommon, due to multipath cancellations and reinforcements,
reflections and diffractions, etc.

When listening to a single transmitter of sufficient strength, you
don't notice this at all - with enough IF gain, the detector still
gets a fully-limited signal and there's no noise.

When listening to a single transmitter with a weaker signal, you hear
"picket fencing" - bursts of noise and distortion each time the signal
strength drops too low, and the detector no longer gets a clean,
limited signal.

Now, throw in a second transmitter, also weaker, in a different
direction. The signals from the two transmitters will be
picket-fencing in a completely un-correlated manner, because they're
coming from different directions.

At one instant, one may be strong and the other weak, and you hear the
first signal. A moment later, you've moved 50 feet, and the opposite
is true - A is weak, B is strong, and you hear B. Somewhere in
between those two locations, the signal strengths were nearly equal,
neither captures the limiter, you hear neither cleanly (just silence,
or a burst of noise and buzzing, or etc.).

In effect, the audio ends up "chopping" back and forth between the two
transmitters' signals, at a rate of anywhere up to multiple times per
second, with noise and distortion thrown in during the chopping.

That is, I think, what the OP was hearing. I've heard essentially the
same effect when crossing the boundary between the service areas of two
different (utterly unrelated) FM stations on the same channel. In this
case the audio programs were completely different - one was not a
delayed version of the other - but the RF issues were the same.