Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Electronic pulse dialing using micro?

Status
Not open for further replies.

neazoi

Advanced Member level 6
Joined
Jan 5, 2008
Messages
4,118
Helped
13
Reputation
26
Reaction score
15
Trophy points
1,318
Location
Greece
Activity points
36,907
Hello, how can I pulse dial a phone line using a micro?
I think a transistor must be connected with it's base to the micro, its emitter (npn) to the gnd and it's collector to one of the two telephone lines. Each time a pulse is sent by the micro, the NPN pulses the line to form the digit.
Do I need to connect a resistor somewhere between the transistor and the line to avoid hard short circuit?
 

It isn't quite a simple as that but basically your thinking is right.

First you need to connect a bridge rectifier across the two phone wires, this is for two reasons, the line polarity may not always be the same, the Telco may reverse the polarity as part of their signalling procedures and also because the ringing voltage will be higher than the DC across the line so it will overcome the DC level at it's peaks and drive the wrong polarity to your transistor.

Then you have the actual pulse dialling, use a transistor but rather than shorting the line completely, it should clamp the voltage across it to a lower level, typically 8 - 10V. The usual method is to connect the transistor (a small power type) as a constant current generator with constant base voltage and fixed resistor in it's emitter. You can use an optocoupler to turn the bias supply on and off to create the dialling pulses. Remember the transistor should be rated to at least 125V to survive peak ringng voltage and at least 100mA to survive the line current.

Before dialling the first digit you may need to generate one long current pulse to alert the phone system - unfortunately I do not have access to my Greece Telco specifications at the moment to check. I wrote the software for performance testing many of your telephones as they left the production line at the factory!

Brian.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
It isn't quite a simple as that but basically your thinking is right.

First you need to connect a bridge rectifier across the two phone wires, this is for two reasons, the line polarity may not always be the same, the Telco may reverse the polarity as part of their signalling procedures and also because the ringing voltage will be higher than the DC across the line so it will overcome the DC level at it's peaks and drive the wrong polarity to your transistor.

Then you have the actual pulse dialling, use a transistor but rather than shorting the line completely, it should clamp the voltage across it to a lower level, typically 8 - 10V. The usual method is to connect the transistor (a small power type) as a constant current generator with constant base voltage and fixed resistor in it's emitter. You can use an optocoupler to turn the bias supply on and off to create the dialling pulses. Remember the transistor should be rated to at least 125V to survive peak ringng voltage and at least 100mA to survive the line current.

Before dialling the first digit you may need to generate one long current pulse to alert the phone system - unfortunately I do not have access to my Greece Telco specifications at the moment to check. I wrote the software for performance testing many of your telephones as they left the production line at the factory!

Brian.

Thanks Brian,
You mean something like this?
What should be the value (approx) for the emitter resistor?
Also is the transistor I have used ok (not sure about the current)
 

Attachments

  • pulse dial.JPG
    pulse dial.JPG
    10 KB · Views: 47

No. That will not work.


You can use DTMF as most lines accept DTMF as well as decadic.
 

Yes, although you must be careful not to ground any part of the line circuitry. Typically a compliant phone (which exludes most Chinese ones!) will have a Zener diode rated at about 10V between the base and bottom end of R2. When a transistor base voltage is held constant the current through the emitter will be (Vb - Vbe)/R2 and as Vb is set by the Zener and Vbe is fairly constant at about 0.6V, the resistor can be chosen to make it pass whatever current you want. I would aim for about 75mA, that may not be optimum in your country but wont be far off.

I strongly advise you to provide bias current through R1 by switching it through an optocoupler so the signal from the micro is isolated. Both phone wires are above ground potential and leaking current to ground or from ground can upset the Telco equipment. Besides, if lightning strikes the lines anywhere in the neighborhood you would risk damaging the micro if it isn't isolated.

Brian.
 

This is all totally inaccurate:
"Zener diode rated at about 10V between the base and bottom end of R2. When a transistor base voltage is held constant the current through the emitter will be (Vb - Vbe)/R2 and as Vb is set by the Zener and Vbe is fairly constant at about 0.6V, the resistor can be chosen to make it pass whatever current you want. I would aim for about 75mA,"
 

I'm open to hearing your explanation Colin.

What I described is in specifications for most European and Australian telephones. They load the line with a constant current to allow extraction and insertion of audio through the appropriate hybrid network in parallel with the transistor/resistor combination.

Brian.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Could I just use a transistor controlled relay to do the pulse dialing more easily, with less components?
I have noticed that (at least in Greece) if you hook off and hook on the line very quickly, you can "create the pulses" needed for dialing numbers, whereas at the same time the line is not terminated, because a longer duration pulse is needed for that (when you lift down your headset).

But I do not know what is the case in other European countries? I believe that Germany has the same telephone system, since most of our telco stuff comes from Germany.

The purpose is to have these limited pins microcontrollers to dial numbers, without the need for an external dtmf tone generator chip like the 8870.
 

To review the exact analog telephone specifications valid for Greece, you can download ETSI EN 300 001 from etsi.org. It contains the general european and country specific regulations.


For a "minimal" line termination circuit, you can refer to schematics of classical analog phones. They implement a simple resistive termination in off-hook state. During pulse dialing, the resistive termination formed by the headset hybrid circuit is shorted so that the input impedance is pulsed between open and short circuit. A RC snubber across the dialing contacts (e.g. 1µF/100-200 ohm) is limiting the transient voltage. See W48 schematic:

1159108700_1406459747.jpg


I believe that the current source termination suggested by betwixt will work too, but it doesn't exactly comply to ETSI regulations. FCC specifications may be different.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Ok so based on this information I came up with the next circuit.
Does it seem right, or am I missing something?

dial2.JPG
 

Minimally, yes that will work but you can omit the dialling relay and use the 'hookswitch' relay instead.

For safetys sake add parallel 'head to tail' diodes across the secondary of the audio transformer or you may get >75V of signal when someone calls you. Alternatively, connect the capacitors across the 600 Ohm resistor so they are isolated until you go off hook. Doing both might be a good idea!

Brian.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Minimally, yes that will work but you can omit the dialling relay and use the 'hookswitch' relay instead.

For safetys sake add parallel 'head to tail' diodes across the secondary of the audio transformer or you may get >75V of signal when someone calls you. Alternatively, connect the capacitors across the 600 Ohm resistor so they are isolated until you go off hook. Doing both might be a good idea!

Brian.


Ok let me see if I understood that correctly.
1. Connect a pair of anti-parallel diodes (say 1n4148?) at the secondary transformer terminals (the right hand side of the transformer coil)
2. Where should I connect the capacitor? Parallel to the 600R?
 

Correct. What you should try to avoid is the chance of in incoming call being routed to the transformer before the hookswitch is operated. A Telco line typically has around 50V DC across it, dropping to about 10V when the phone draws current (off hook, hookswitch closed). The Telco monitors the voltage current through the line so it knows if you are on hook or not. If someone calls your number the Telco checks your hookswitch state, if off-hook it alerts the caller your line is busy but if on-hook it superimposes around 75V AC on the 50V DC supply that is already there. So when ringing voltage is present, the line peaks at about -25V and +125V (50 +/- 75) and you don't want that to be routed to the audio transformer!

A telephone usually has an additional capacitor across the line (to block the DC) before the bridge then back to back Zeners to drop around 4V in both polarities and anything remaining goes to the ringng voltage detector so you can monitor for incoming calls. In other words it looks for the 75V AC and ignores the DC and any voice signal on the lines.

Brian.
 

A telephone usually has an additional capacitor across the line (to block the DC) before the bridge then back to back Zeners to drop around 4V in both polarities and anything remaining goes to the ringng voltage detector so you can monitor for incoming calls. In other words it looks for the 75V AC and ignores the DC and any voice signal on the lines.
Brian.

Thanks Brian,
I am a bit confused about this capacitor. Should it be in series with the line (like shown in the schematic above) or in parallel to it (connected one line wire to the other through the capacitor).
 

The RC snubber should be connected parallel to the dialing contact. It shouldn't be connected parallel to the line in normal operation because it disturbs the line impedance matching.
 

The RC snubber should be connected parallel to the dialing contact. It shouldn't be connected parallel to the line in normal operation because it disturbs the line impedance matching.

So the RC should be placed in parallel to the on/off hook switch of post #10 and in series with the 600R? (asumming only the hook switch is used for dialing numbers, as described earlier.
 

Correct - and the audio coupling capacitor and transformer should be in series and across the 600R.

Without my data sheets I can't check at the moment but something at the back of my mind makes me think the line impedance in Greece is 400 Ohms. It's really an impedance not a resistance but for your purposes it should be OK to use a simple resistor.

Brian.
 
  • Like
Reactions: neazoi

    neazoi

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top