[Ronja] worst-case with Q103 in the receiver?

[Karel Kulhavy]

On Fri, Dec 08, 2006 at 10:11:09PM +0100, Thomas Egenhofer wrote:
> hi again,
> i'm back with another few and annoying question but please bear with it.

You don't have to humiliate yourself in advance to make sure you won't be
kicked or shouted at. This is neither OpenBSD mailing list, nor Gentoo
bugtracker, nor a date with a girl.

> (i'm just asking because i had the very same problem with my own
> receiver circuit of a slower, low-cost fso)
> it's about Q103 in the receiver.
> from what i figured out it's why you have to limit the receiver for
> short-distance links.
> well isn't it possible that the BE-voltage of Q103 can swing into a
> negative value ( if the signal is strong enough.. means about
> 2xVbe-pp(>~1,4V)and more).

The circuit is designed to cope with this case. When the BE voltage is
negative, the transistor just turns off.

> in my case this caused the current to flow through the BE-diode part of
> the transistor right onto the base-capacitor and "flooding" it.

I don't understand how it could. If BE voltage is negative, no current flows
through the BE diode.

> than the be differential voltage on the base of Q102 and Q103 would be
> unfortunate and causing the circuit to not work like it's supposed to
> do.
> 
> well in case this IS the case and you want to correct it (thus making
> low-distance tuning not neccessary anymore... you could either use the 2

The low distance tuning is not because of the end transistors. It's because
the NE592 doesn't seem to like too strong signals. Or maybe the Q101. I am
not sure. I am trying to improve this with my new all-transistor design.

Though it's difficult. Imagine you want to genetically engineer a musician to
be able to hear a flute over 1.4km, but at the same time to be resistant to
damage when someone places a gun next to his ear and pulls the trigger.

CL<
> transistors in a cascode wiring (little bit bad to tune the needed
> voltages- i did this in my circuit(remember mine is not the ronja rec.)
> or add 2 diodes. one in front of each transistor base. (the one of Q102
> might be better build between the 2 6.8k and the 1k resistor.
> 
> jep this is pure theory and i won't stop you from blaming me for.
> but can anyone confirm this behavior of the circuit?
> 
> oh and another small addition. from my experiments it's very good to
> highpass the low-frequency's out of the signal after every (reasonable
> high) amplification. so adding a second capacitor after C167 and ground
> the wire between them over an ~500 Ohm resistor might improve signal
> quality,too.
> 
> greetings
> thomas e
> 
> ps:this is the only part of the receiver which worries me. the rest is a
> prefect design as far as i can tell.
> ps:ps:if you use 2 diodes for reverse-voltage protection you can avoid
> to short-circuit the power-supply :)
> 
> 
> -- 
> Twibright Ronja mailing list http://ronja.twibright.com
> Ronja at lists.pointless.net
> http://pointless.net/mailman/listinfo/ronja