At 21:17 4/10/99 +1000, you wrote:
>Dennis Plunkett wrote:
>
>> As for that Q1 thingo....  . The 12V zen will clamp the line voltage
>> and shunt the audio a bit,
>
>  Quite a lot, I'd think, just *asking* for distortion, quite
>undesirable for a modem.  I find it a trifle strange to put it there.
>

Yes I agree too, nominaly this is there as protection for the CCT and the
cct will nominaly hold the line voltage to under the zen limits. 



>> as for Q1, it is configured in a slow start configuration that also
>> provides a bit of dynamic current limit control the intention is not
>> to seperate the audio, but is a side advantage
>
>  I would *hope* it's intended to separate the audio, because there's a
>22 µF capacitor on the other side of the transistor.  If it was just a
>slow switch on (and why on earth would you *want* a slow switch on?),
>you wouldn't really need the 51 ohm resistor at all.
>

Take a look at TS002, it sets out the maximum current that can be drawn
from a line at any one time and also sets out the current/voltage grpahs
that *have* to be meet in order to pass. So slow switch on in a potentialy
high current part of a cct is desireable. However, for the fist 300mS you
can almost do whatever you want, but in the case of this cct the 12V zen is
too high, and should limit the line voltage to 8 to 9V, this then lets the
designer draw as much currrent as possible during this period.


>> (51 ohms is not enough to ensure that levels of +3dBr are acheived
>> with QI fully on, but should be OK for most users)
>
>  The 2k7 ohm resistor prevents saturation, so the transistor is
>definitely in constant-current mode as far as AC is concerned.  As such,
>the AC component doesn't see the 51 ohm resistor, only the 2k7.  Again,
>saturation is *most* undersirable as it generates distortion.

Huh? The 2k7 only limits maximum base current, I don't see how this
prevents saturation, as a dynamic current change on the emitter will shift
this point around, with only the cap controlling the dynamic current
response. It should then be possible to drag the emitter more than one Vbe
down and thus place the transistor into staturation, all be it for short
periods of time. Also if saturation was not a problem then the 51 ohm
resistor would not be required. If I recall this is not there to limit
current into the cct, but as a discharge path for the cap. But then again I
am more than likely incorrect as usual. 

Oh yes, add a PIC to control the ring cadence detection and off hook
control <G>



Dennis




>-- 
>  Cheers,
>        Paul B.
>
>