Stereo Tool

When you say you cannot get them to sync on air, what exactly do you mean? Do you hear a difference in timing? Or do you mean that the combined signal doesn't receive properly?

We have seen before that there can be *very* small (but constant) differences in timing, of maybe 2-3 microseconds, which you would need to compensate for manually. You can't hear it (it's less than 0.1 sample at 48 kHz), so to see these very small differences you would need to measure it. But these differences should be small enough to not interfere with reception in the center between the two transmitters (it could move the center by a few hundred meters though). More importantly, if the overlap area between the transmitters is not in the center, you need to compensate for the speed of light (I think it's 3.336 microseconds per kilometer). It's easiest to first verify that the decoders are in sync in one location (I can't think of any reason why they wouldn't be, if the loopback is connected and used, but it's always good to verify things), then move them to the target locations and adjust for the distance.

For the transmitters: For optimal results, the modulation should be exactly the same. So, ideally, the transmitters themselves should be identical, and also locked to GPS. But even without that, just syncing the MPX signal should already improve the reception.

That's about as good as I would expect it to get without syncing the transmitters as well (and you might need to use identical ones too.)

By the way, even if the 2 transmitters are perfectly in sync, there will still be places where they cancel out each other. For the RDS (the highest part of the MPX spectrum), at 60 kHz the distance of 1/4th of a waveform is 300000000 / 60000 / 4 = 1250 meter (so from the center point the waveforms will pretty much match without TOO much difference to about 600 meters in both directions). The FM carrier frequency is much higher though, around 100 MHz, so that distance to get perfect anti-phase (a half waveform apart) is 300000000 / 100000000 / 2 = 1.5 m. So every 3 meters there will be a point where the signals cancel each other out, in between the signal will be stronger or weaker depending on where you are. But, if your FM transmitters are not synchronized these points will move around, with a speed of 3 meters per second per 1 Hz of offset in frequency. And you can easily be a few hundred Hz off without knowing (most modulation meters won't show this).

What this comes down to is that your receiver will see huge AM noise in the signal (moving possbily hundreds of times per second between no signal at all and the combined signal of both transmitters), and that makes it difficult to properly demodulate it.

How much correction did you need to use? (Or is the point where you're checking this not in the middle?)