On 2/12/19 10:05 PM, Pierre Martel wrote:
http://qrp-labs.com/progrock.html
looks like this could maybe usefull on multicasting???
Simulcasting is what this would be called. There are a number of issues with simulcast, disciplining the transmitter is just one of them. This looks like it uses the 1pps output, not the 10 mhz output from the GPS, which will not have the resolution to lock the reference properly. The SI labs chip used in the PCGM used the 10 mhz output, but these had their own issues too. The digital clock generators are harmonic rich, and really need filtering to be used. The traditional way to use the RTCM/voter is to remove the crystal reference on the transmitter PLL replacing it with the generated GPS clock. In most radios this will remove the dual point modulation provided by a diode across the crystal. Now modulation balance will be way off, and it will be much harder to get audio to line up. Most stations which have provisions for external reference already will use a 10 MHz or 5 MHz clock and lock the reference to that. The Quantar and Tait will handle this just fine. I'm been looking at the CDM radios (waris) for this too, as they use dither modulation of the reference frequency. I've not had time to look further into it. The other issue is the audio needs to be delayed the same through each transmitter. Some DSP based transmitters (SLR repeaters, etc) have a non-deterministic delay in the audio. 73's -- Bryan Fields 727-409-1194 - Voice http://bryanfields.net