Michael Black wrote:
) writes:

Just because a few people get it wrong on the net, doesn't make it
right. I've seen DSBAM used for AM, DSBSC used for DSB. Check out this
college text:
http://www.utdallas.edu/~mtacca/cour...-2006/Exp3.pdf
In the equation, you would have to add a DC bias in x(t) to produce AM.
Hence they are producing DSB, pure and simple.

The only reason you'd need a DC bias on the transmitting mixer is if
the mixer was of a balanced type. IN that case, the DC bias unbalances
the mixer, allowing for the carrier to get through to the output.

But, long before balanced mixers came along, there were lots of mixers
that let the carrier through.

Hence to get DSBc you'd either need an inherently unbalanced mixer
at the transmitter end, or if it is a balanced mixer of some sort,
unbalance it, with a DC bias or messing with the balance of
the circuit. (It is often easier to apply a DC voltage, but it's
not the only way).

Of course, you don't even need a balanced mixer to send DSBsc. You
could notch out the carrier at the output of the mixer. Just put a very
sharp filter there, that only gets rid of the carrier but not the sidebands.
Not even close to a common practice, but I've seen it played with. I've
also seen SSB transmitters that started with a balanced mixer, but also
used a single crystal notch filter at the output of that mixer, to
increase carrier suppression; in the days of tubes the balance of the
mixer sometimes varied with temperature (and wear) and the filter just
guaranteed good carrier suppresion.

Let's not forget that early experiments with SSB used a very low
carrier frequency, and used the low bandwidth of the antenna to
filter out the carrier and the extra sideband. Obviously a filter
(the antenna) was used to get rid of the unwanted sideband, but it
was also used to get rid of the carrier. The transmitter was simply
a DSBc transmitter, as seen on the "AM broadcast band" today.

Michael

It never ever ends. You do not need to talk about mixer designs to
describe modulation. It is handy since most people that studies
communications can look at a signal flow graph and visualize the
modulation. However, you can explain double side band simply from trig
identities. In communications, you do the theory, then build the
hardware. Those who do it backwards eventually get screwed.

cos(a)*cos(b)=0.5*(cos(a-b) +cos(a+b))

Make cos(a) here is the message, while cos(b) is the carrier. In this
case you get two tones, i.e. the sum and difference frequencies that
are located above and below the carrier. Why, it's like, dare I say,
double sideband! Hence DSB is all you need. There is no carrier to
suppress.