blowing up my ATMega, gyrators, and general update

[DavidRowe]

Hi, a few updates on the project:

1/ Over the past month I have put some work into integrating the
building blocks together, for example DC-DC converter, ADC and DAC sub-
systems. Here is the latest schematic:

http://www.rowetel.com/downloads/ata_sch.pdf

The latest software is available from SVN:

$ svn co https://lowcostata.svn.sourceforge.net/svnroot/lowcostata/trunk
lowcostata

All the parts are numbered now however the numbering is incorrect
(e.g. Q9 when there are only three transistors). Need to fix that some
time.

2/ I just blew up my ATMega chip! To test some of the audio I/O I had
the DC-DC converter hard-wired (no control loop) to generate about
20mA through the current loop with the phone off hook. However this
meant when I hung up the phone Vbat shot up to over -100V (as the load
was removed). When I next took the phone off hook for a moment there
was -100V across the Gyrator transistor which popped the CB junction.
This in turn connected one of the ATMega ADC inputs to -100V. Oops.
Now the ADC doesnt work, and a few of it's inputs are fused to GND,
although other bits of the chip seem OK.

To prevent this happening in future I added a 12V zener across the
Gyrator, and a current limit resistor to the ADC input.

3/ So now I am am waiting for some new ATMega163 chips to arrive.
Before I blew up the chip I was playing and recording audio OK using
the integrated software - next step is to impement the host side of
full duplex sample/replay and try some simple echo cancellation.

I also did a little analysis on the Gyrator, as I was curious as to
how the thing worked:

http://www.rowetel.com/blog/?p=51

Cheers,

David