Help with am demodulator

hi
this is a simple AM modulator and an easy one
I can understand the whole circuit except for those two parallel diodes
could any one explain me their role
and I guess they are not for DC consistency because the circuit works even better without them ( this is just an assumption based on my simulation results )
could you introduce me a book better than clark-hess on telecomunication circuits ?

thanks for your attention

Do you mean D2 and D3 in parallel with D1 and R2 . . . looks like a bias the base of Q1

no i meant D2 & D3 all alone
they aren't for biasing because the circuit works fine without them

sina_extreme said:

no i meant D2 & D3 all alone
they aren't for biasing because the circuit works fine without them

Click to expand...

They help limiting the EB current of Q1, in the least:
D2 and D3 are in parallel with D4, R2 and EB of Q1

Kerim

Without the two diodes the circuit functions as a "collector bend" detector, with the diodes a "base bend" detector. Whats this all about you say?. For a demodulator to function at all there needs to be some non-linearity in its circuit. The simplest demod is a single diode, when signal goes +ve it conducts, if the signal goes -ve it does not. This is a gross form of non linearity. The theory behind the original circuit is that the transistor should only be conducting pA of base current and micro amps of collector current so if the input signal goes +v it conducts more and if the signal goes negative , the transistor cuts off. When you remove the diodes the transistor goes into saturation, i.e. its conducting as much as it can, so any +ve signal will be ignored, but a -ve signal will bring it out of saturation and the collector voltage will rise, i.e. it will work as a demodulator. Rather then the the diodes, it would be better to take the base via a 100K to the centre tap of a pot, so you can adjust the pot so the collector voltage JUST fall from Vcc with no signal input.
Frank