telephone interface

[Tommy volts]

This weekend I am going to try to build an interface circuit similar to the one presented on the web file below. <p>The question I have is:<p>Is it necessary to connect to the telephone as shown in the circuit below, or can I plug into another unused phone jack and pick up the audio signal? <p>The latter is preferred because I would rather not have my interface electronics set up in the kitchen where our telephone is.<p>Thank you for your advice.<p>
telephone interface

[Mr. K]

Should work anywhere. Remember, all of your phone jacks are wired in parallel.

[dyarker]

Correction: Will work if the jacks are wired in parallel. For example, if you have two phone numbers in the house, one or more jacks will be wired for one number, and other jack(s) wired for the other number.<p>Cheers

[Tommy volts]

Mr. K & Dale:<p>Thank you for your advice!<p>I built the interface and plugged it into the jack upstairs. It turns out that the jack upstairs is wired in parallel to the one in the kitchen( red & green lines).<p>It worked fine but with the following problems:<p>1. The interface exists in the "off-the hook" condition. That is, I have to plug it in after the phone downstairs is picked up otherwise an incoming caller will get a busy signal. Then after the phone down stairs is hung up I have to disconnect the interface or the line will remain "off the hook".<p>I think/hope I can solve this problem by using a non-polarized capacitor instead of the polarized capacitor that I used.<p>2. The audio output I get sounds like people talking from the moon, with alot of feed back and science fiction sounding humming (like a Theramin) and reverbaration.<p>I hope to solve this by trying a different amplifier (higher impedance?).<p>Thanks again for your posts.

[Mr. K]

Not sure why you experienced that problem. Maybe someone with more circuit or telephone know-how can help more. Here, in imprecise terms, is what should be going on.<p>The Telco supplies a DC voltage (they call it "battery") of around -48V to your telset. The two lines have a capacitor in series with the ringer across them, so when the AC ring signal is applied, the ringer responds. This is conceptually similar to your circuit, with the cap, xformer and resistor seriesed across the line. Going "off hook" occurs when the switchhook, which is normally held "open" by the cradled receiver, is closed and battery is applied to the phone's internal dial and voice-handling circuitry. Therefore, "off hook" implies DC continuity between the lines, with a current flow or voltage drop that the Telco can sense.<p>Your input circuit should only be passing the ring signal plus the "AC component" of the modulated DC which results from the phone conversation. If you have truly blocked DC flow between the two lines, this device should appear to be "on hook".<p>I'd recheck the project and installation and look for a DC leakage path between the phone lines.<p>HTH.

[frhrwa]

why not just do the circuit like miguel says in the article above.. insert the transformer in series with either side of the line in (t or r), then take the secondary side off to the recorder.. the impedance doesn't come into play across the line so it will never trip the ringing. when your off hook on any phone, those two wires, t & r are alive and will put your voicestream onto the recorder..

[dyarker]

You absolutely must have the capacitor in series with the transformer winding, and it must NOT be electrolytic. <p>Off-hook is caused by drawing 20mA (nominal) through the phone line. You guessed right about using a non-polarized cap.
Mr K - that's a pretty good description (my job is the care and feeding of a Nortel 500 line PBX and 300+ lines of Norstar key system).<p>-------------
Changing the capacitor may help the sound too. Most audio transformers aren't designed to carry DC current. When they do, the core saturates; then only the peaks of one polarity of voice or music get thru. The other signal polarity gets through because it's subtracting from the DC. The result looks a mess on an oscilloscope, and sounds BBBAD.<p>Some telephone transformers are designed for 75-90mA of DC current. They larger then equivalent zero DC type because the core is larger to avoid saturation.<p>Try replacing the 4.7 KOhm 1/2W, with 3.9K and 5.6K to see if that helps.<p>Some small amplifiers use RCA jacks for speaker output. This circuit needs line out levels. To use a speaker output, use a 600 Ohm to 8 Ohm transformer backwards. The center tap on a 1K Ohm is okay. The other side can be 8 or 16 Ohm. Impedance matching is important, but not critical for this job.<p>-------------
The transformer winding can be in series with one wire of the line, BUT:<p>a. It must be in the line before any parallel wiring to the jacks (in series with ALL extensions).<p>b. The transformer must be able to handle DC current (see above).<p>c. To avoid loss of conversation volume on the phones (yours and distant end), the winding in series needs to be less than 600 Ohms. (else from the phone company equipment's point of view your line is now 1200 Ohms, your phone gets/sends less than half the sound power it should.)
A fix is to use a transformer with a 600 Ohm primary and two 300 Ohm secondaries, use the secondaries in parallel for 150 Ohms. The conversation volume will be nearly the same, increase the volume of the music to compensate for the transformer step-down.<p>-------------------------
One last thing. Move the zeners to the other transformer winding. That way the current they draw when clamping ringing doesn't have to go through the transformer.
***********
Oops, you're doing pick-up. Everything is still correct except "speaker jack" comment. And that is correct for music-on-hold projects.<p>C U L