edhall@rand.org (Ed Hall) (06/06/91)
Other people have taken a shot at the problem of a line-driven in-use indicator. Here is my attempt: 2.2K +--------------+-----+---v^v^v-----+ | | | \-/ | > > 2.4M _V_ LED | + 2.4M < < | -------- > > | ------+ O---|200PIV| | |_____|/ | Phone |Bridge| | | |\ | O---| Rect.| | | / | V | / -------- |__|/ \_|/ | - | |\ |\ | > | V | V | 100K < \ \ | > | | | | | | +--------------+-----+-------------+ All transistors are NPN and should have a BVce of 150V or so, and an Hfe of about 100. The LED should run on 2 or 3 mA. The 100K resistor may need to be adjusted if the on-hook voltage is less than 48V, or if the phone doesn't hang up properly. It should be set to allow the LED to come on when the line is down to 12V or so. Of course, conecting anything that isn't type-approved to your phone line is a violation of FCC rules, so this circuit is for educational purposes only. :-) The theory of operation is simple. The bridge rectifier guards against polarity reversals. When the voltage divided between the first 2.4M and 100K resistors goes above 0.6V or so, the first transistor will begin to conduct; this happens at a line voltage of about 12V. This conduction shunts the current passing through the second 2.4M resistor so that the bias current through the other two transistors (connected in a gain-multiplying Darlington configuration) is no longer able to make them conduct, extinguishing the LED. When the phone line is in use, the voltage reduces to 9V or less. This prevents the first transistor from conducting, and thus allows the other two to conduct, lighting the LED. The 2.2K resistor is chosen so that the LED doesn't draw enough current to keep the line below 12V (assuming the line will source a minimum of 5mA or so); thus when the line is no longer in use, the circuit can shut the LED off. Increasing the current through the LED by reducing the value of this resistor might increase the current enough to keep the circuit from allowing the line to go back "on-hook". I'm afraid I can't give part-numbers for the transistors. The BVce is likely to be the limiting factor in choosing them; you can choose a smaller breakdown voltage, but you'll risk frying things the first time the phone rings. (The junkbox transistors I tested the circuit with were hand-picked and didn't have intelligible part numbers.) Putting a 150V MOV at the input might not be a bad idea, come to think of it. -Ed Hall edhall@rand.org