[hpsdr] An high dynamic QSD front-end

[Marco IK1ODO]

At 23.53 14/01/2007, Jim Miller wrote:


>i'm still left puzzled how any of the active integrators deal with all of
>the signals outside their gain bandwidth capability.
>
>they only present a virtual ground to frequencies at which they have
>significant excess gain.
>
>what am i missing?
>
>73
>
>jim ab3cv

Nothing, Jim.

Please have a look at my schematic at 
http://www.spin-it.com/sdr/IK1ODO_SDR1.html

The impedance seen from the RF port is always close to 50 Ohm, at any 
frequency.
The key is the diplexer after the integrator. Signal offset more than 
250kHz from F0 are grounded there. The opamp is not working at all on 
HF signals.
I measured the "blocking" dynamic range, defined as the signal level 
that causes A/D saturation or significant distortion in the mixer. 
For in-band signals it is -3dBm (for 4Vrms audio out to the EMU); at 
1MHz from the tuning frequency +10dBm are needed, then the protection 
diodes in the FST start conducting...

The opamp must not see signals outside it's bandwidth. I think I 
demonstrated it. Transients out of the QSD reach hundreds of mV with 
spectral components in the GHz range; nothing that an audio opamp 
could withstand. Blocking them in the diplexer stabilizes the mixer 
output impedance and avoids distortion in the opamp.

73 - Marco IK1ODO


 1168897334.0