Here's the lengthy follow up of the ringing/hangup problem using my magicjack with a Panasonic FAX/Answer machine and 1A2 system 400G line card. The problem was I would get a few cycles of ringing voltage that was detected and then the MJ would disconnect the caller. Here's a pix of the ringing waveform. Right click on it for full view.
As you can see, I would get a few cycles of the ring voltage, and then it would disconnect. I think the line card momentarily closes the line relay contact, then opens up and the call gets disconnected. This also happened with my original 400D line cards.
Anyway, the work around I came up with involves disconnecting the ring detector section of the 327A line relay (the 755 Ohm coil) and adding a 6N139 optoisolator (or equilavent) ring detector circuit to the line card. See the schematic.
The modification to the standard WE(mine is ISSue2) 400G card involves disconnecting (unsoldering & lifting) one end of R2, a 6.8k or 7.5k, (depending on the version, there's only one on the card). The end you want is the point closest to the edge of the card. This solder hole connects to C1, the 1.1 uF capacitor, which you will connect the optoisolator circuit to. This is shown as point "A" on the schematic. The 47k resistor will connect to that solder hole. The other end of C1 stays connected to the line card and eventually connects to Pin 14 (T-(CO)of the line card. The 2.2 MOhm resistor is soldered directly across the capacitor, C1, and is used to provide some small reverse bias to the optoisolator diode. The reverse bias is to help prevent false "dings" from voltage spikes. The 1N914 is for reverse voltage protection from the ~60 Volt ringing. Pin 3 of the 6N139 is connected to any convenient solder hole that goes to pin 9 (R+(CO)) of the line card to complete the ringing detector pickup circuit. The output polarity from my MJ worked out this way in order for my old TT pads to operate - yours may be different. In that case, just reverse the T&R from the MJ.
The output side of the 6N139 is a regular Darlington configuration (Pin 6 & 8 tied together) and connects across the N.O. contacts of the 327A line relay (there are no N.C. contacts). To get the polarity correct, pin 6 & 8(+) of the 6N139 goes to the solder contact closest to the blue handle of the line card. There is ~13 Volts across the output circuit. The 24 Volt zener is to protect the output from any voltage spikes from the other 24 Volt relay coils on the card, any ~24 Volt zener will work here. The line relay contact is still used for the normal line card hold functions, etc.
This circuit was constructed "dead bug style" around an 8 pin DIP IC socket and has been working for about a week now with no problems. I wanted to completely document this "fix" so others that experience the same problem don't have to reinvent the wheel. My 1A2 system continues to live! Any comments or additions - please send 'em on... (Is there a better way to post images?)
Sam - I can just solder back the one end of the 6.8k resistor and not void the warranty! - ha ha.
George