Andrews circuit design to detect when a ringing phone has been answered actually gave me a fair amount of trouble to understand. If I'm honest I would confess I still don't really understand on a technical level exactly how it works, more that it does "work". But what really didn't help was not realizing that there are two types of transistors, PNP and NPN, and for a long time was using the wrong one - no wonder it wouldn't work! I guess we learn by our mistakes and once I had it figured out I was up and running. Andrew was very helpful during this time patiently answering all my questions. As Andrew explained, and I hopefully I have this right, the way I see it is that the 47 ohm resistor generates a small current before the destination phone is answered, enough to stop the base on the transistor passing current. When the phone is answered the base current flows to earth activating the transistor which in turn activates the photo cell in the optocoupler which tells the Arduino the phone has been answered. The 10k and 10uF RC filter remove the AC ringing current so as to allow off-hook detection during a ringing cycle. Until a phone is answered the AC current flows through the bells via a DC blocking capacitor in the phone. Here is the breadboarded circuit.
To get the bells to ring I will need a transformer that will generate a decent voltage and I find a 48v one Ebay which should be enough to get a decent bell ring though the bells will probably need a bit of adjusting. Another thing that took me a long time to get my head around was understanding how that the AC voltage is superimposed over the DC by feeding the DC into the AC transformer and back out the other side, along with the AC. Andrew pointed out it will only work with a completely separate transformer, in other words I couldn't take a another feed from the same transformer I was powering the rest of the circuit with.
All that's left to do is add the circuity to the Veroboard. I have also added the "bridging" capacitor over which the AC speech currant will be transmitted between source and destination phones when they are linked and the second destination line isolation transformer. The relay that will join the source and destination lines is still not fitted as shown by the empty 8 pin socket. The IC is a dual relay so the second relay in the IC will drop out the call sense when a phone is taken off hook. I have added the remaining 1W 1k resistors and a Dupont socket so I can take a feed from the optocoupler that detects the destination phone OH to the input shift register.
The final part of the puzzle is to add the relay that will make the phone ring cadence, two 400ms rings with a 200ms pause between each ring followed by one second of silence, the standard GPO set up we all know and love. I used a
TLP3043, just to right of the destination isolation transformer, which, if I remember, was the only IC component that wasn't obsolete from Andrew's original design. I would imagine the
Panasonic AQW210EH relays I had been using for everything else would have worked just as well. This relay is fed from the 48v transformer discussed above.
Comments
Post a Comment