Facebook
Google
GitHub
Linkedin
DSB-SC demodulator
- Thread starter Kerim
- Start date
kerim .if your dsbsc demod unit using a pll is based on alternate sequential inversions of demodulated oudio switched at the crossover points of the DSB signal . that idea was patented by leonard Khan in USA a brilliant comms engineer in the 1950's I think. he of course did it all with valves ! ... , I have the patent copy somewhere . .I just have to dig out the wirelss world circuit in jan 1987 to compare cheers Mike
Mike,
Me too while working on my thesis (1979), I had the idea (it was before the last one) to recover the original audio amplitude by inverting the demodulated one at the appropriate crossover points of the DSB-SC.
I even discussed this possible method with my supervisor, an American prof. (and amateur) in communications. So I tried to build my own version of it. But I found out that it wasn't the easiest and the 100% reliable solution I was looking for. So although I made it to work, I considered it as another failure since the idea wasn't just designing a DSB-SC detector.
My last attempt was using a conventional PLL though its formulas say clearly it shouldn't work. I built it anyway and I did it on my last day in the lab and at the university (since I had to return home very soon later). To my big surprise, the circuit locked to the suppressed carrier!!!
I thought first that I was imagining things. But when I changed the carrier frequency of which the DSB-SC signal is generated, the PLL lost its lock state outside a small bandwidth around it. At that moment, I had no idea how this could happen. So I watched on the oscilloscope every voltage trace on my circuit. I found out that the lock was possible due to an imperfection in one gate which wasn't included in the formulas.
Obviously, after I returned home I re-analyzed the PLL circuit considering the effect of this new ‘imperfect’ factor as well. After the study, I was able to increase the lock range from the 0.2% when at lab to more than ± 5% while keeping the circuit simple and without having a degradation in the recovered audio (musical) signal. But ideally, the highest possible lock range is within ± 25% of the suppressed carrier.
As you see, FCC still needs time to be real good in communications.
Cheers,
Kerim
Me too while working on my thesis (1979), I had the idea (it was before the last one) to recover the original audio amplitude by inverting the demodulated one at the appropriate crossover points of the DSB-SC.
I even discussed this possible method with my supervisor, an American prof. (and amateur) in communications. So I tried to build my own version of it. But I found out that it wasn't the easiest and the 100% reliable solution I was looking for. So although I made it to work, I considered it as another failure since the idea wasn't just designing a DSB-SC detector.
My last attempt was using a conventional PLL though its formulas say clearly it shouldn't work. I built it anyway and I did it on my last day in the lab and at the university (since I had to return home very soon later). To my big surprise, the circuit locked to the suppressed carrier!!!
I thought first that I was imagining things. But when I changed the carrier frequency of which the DSB-SC signal is generated, the PLL lost its lock state outside a small bandwidth around it. At that moment, I had no idea how this could happen. So I watched on the oscilloscope every voltage trace on my circuit. I found out that the lock was possible due to an imperfection in one gate which wasn't included in the formulas.
Obviously, after I returned home I re-analyzed the PLL circuit considering the effect of this new ‘imperfect’ factor as well. After the study, I was able to increase the lock range from the 0.2% when at lab to more than ± 5% while keeping the circuit simple and without having a degradation in the recovered audio (musical) signal. But ideally, the highest possible lock range is within ± 25% of the suppressed carrier.
As you see, FCC still needs time to be real good in communications.
Cheers,
Kerim
Last edited:
