HFU HF Underground
Technical Topics => The RF Workbench => Topic started by: ButchKidd on May 30, 2024, 1817 UTC
-
I'd like to experiment with the gnuradio nrsc5 (HD-Radio) encoder module, and I'm wondering if it would be practical to build a rudimentary I/Q modulator on a breadboard. Or would the output of something like that be so non-linear that it's just not practical? I've got no RF design experience, so I'm just spit-balling and hoping somebody more experienced will chime in.
Does that sound naive? Or is there a version of the Chinese ready-made DDS VFO modules that will already do what I'm talking about, and I just don't realize it?
I was thinking of just taking the I and Q output from a sound card and feeding them into a pair of MC1496 balanced modulators, then combining that into some sort of output transformer, maybe with an op-amp depending on what the output levels are like. I'm sure it will need some filtering to remove all the nasties created by all the mixing and combining, and that will have to be trial and error. Initially, I'd probably use a 2-channel signal generator for the quadrature carriers, but if it works like I hope it will, I'd probably add a VFO with some type of phase shift circuit on one branch.
I've seen several articles and YouTube videos about doing demodulation in this way, so I just assume the reverse would work.
-
I've built prototype MW AM stereo modulators on breadboards in the past, and it worked. With digital modulation properly filtered in the digital domain, the only issues you will likely have is spectral regrowth, ie out of band emission. This is measured relative to carrier level, and at present -40dBc is considered acceptable.
COTS solutions may exist, but I don't know of any at a reasonable price. If you need some schematics to get you going I can pull some stuff from my archives.
+-RH
-
I've built prototype MW AM stereo modulators on breadboards in the past, and it worked. With digital modulation properly filtered in the digital domain, the only issues you will likely have is spectral regrowth, ie out of band emission. This is measured relative to carrier level, and at present -40dBc is considered acceptable.
COTS solutions may exist, but I don't know of any at a reasonable price. If you need some schematics to get you going I can pull some stuff from my archives.
+-RH
Thanks, redhat! If your schematics are easy to lay your hands on, I'd appreciate it. Otherwise, I think I understand how it should work. I guess we'll find out.
-
Apparently I don't have any schemos for the old stuff I built. I can throw something together, but the through hole versions of the 1496 are NLA. They are still available in SMD. Do you have SMD capability? It might be easier to roll a small board for this.
+-RH
-
I haven't used breadboards since the early 90s (perhaps it was earlier) but I'm going to say that proper layout for good isolation of I and Q would be the priority. Redhat's -40 dBc seems reasonable to expect. At those frequencies we can get ~-60 dBc or better in much more compact areas but with optimized layouts.