Need help with a telephone privacy protector circuit(a circuit almost similar to IVRS

[pradeep_k_b]

Hi,My circuit diagram is a micrcontroller based one but my question and doubt is about analog circuit designing part,that is why I am posting this here

I am trying to design a privacy protector for my home telephone,a simple device using Arduino and this device will connect just before my phone receiver,and when someone call me the device will automatically attend the phone call and will ask them to enter a password and if the caller enter the correct password then only the call will connect to my telephone and the telephone need to ring,if the password is wrong then the device will disconnect the call.

I am doing this using Arduino and DTMF decoder.The caller will be able to enter the password using his phone keypad itself and on the receiver end the device will decode the password using a DTMF decoder and will compare the password with one that already saved in Arduino.

The problem is actually I don't know how to switch the phone lines.The device will automatically attend the call before the call reach in the receiver phone and will ask to enter the password and if the password is correct then the call will connect to receiver phone and the phone want to ring.How can I do this?as the call is already attended by the device I don't think the phone will ring again,how can I solve this issue?And also how can I detect a ringing signal to turn on the automatic hook up circuit?

Please help me.I am waiting for your feedback

 

[Audioguru]

Of course the telephone line will not ring again since it was already answered by your machine.

Do you know how to detect that the line is ringing and how to answer it?
Do you know how to feed your message into the phone line and receive DTMF from it?
Then make a ringing generator, off-hook detector (when you answer your phone) and a switching circuit.
Make it return to normal after your call.

Some of the parts you need are in a telephone answering machine.

 

[FvM]

Besides going off-hook and doing the DTMF decoding, your intended system needs basic features of a private branch exchange to transfer an accepted call to your telephone e.g. generating the ring voltage.

 

[pradeep_k_b]

@Audioguru..Thank you very much for your feedback,I already designed a DTMF decoder and its working fine,but there is no automatic hook up circuit so I tested the DTMF decoder with my mobile phone and I answered the call manually and then its working fine.I have entered a code in caller phone and I could decode it in the recieving end.But actually I am designing this circuit for land-line and I know it is different.At the time of ringing the voltage level will be higher and we will detect it and the call will connect to the DTMF recorder....But I am really stuck up the switching section

@FvM...If the password is correct then I am planning to make a connection between the input and output of the device using a relay.Then actually the device will not have any role after comparing the password and phone lines will connect directly.Do you think this method will work?

 

[crutschow]

Here's the sequence of events as I see it:

Initially your phone will be disconnected from the line by a relay NO contacts (relay2).

You will need to determine what the ringing signal level from the phone company is and detect that to answer the phone (using relay1 to connect to the phone line and disconnect your phone from the line at the same time).

When the phone is answered, it must put a minimum load on the line as determined by the phone company requirements, which signals that the call is answered.

When the code has been approved by your circuit you will need to generate the ringing signal for your phone.

Then the circuit must detect when you pick up the phone and connect the phone to the line by a relay2 while releasing relay1.

The circuit will need to detect when the call is completed and release relay2 to again disconnect the phone from the line.

 

[crutschow]

Correction to my previous post since I couldn't edit it:

Here's the corrected sequence of events as I see it:

You will need to determine what the ringing signal level from the phone company is (typically a low-frequency relatively high-voltage signal) and detect that to answer the phone by operating a relay to connect to the phone line and, at the same time, disconnecting the phone (SPDT relay).
This must put a minimum load on the line as determined by the phone company requirements, which signals that the call is answered.

When the detected code has been approved by your circuit, a ringing signal must be generated for your phone.

Then the circuit must detect when you pick up the phone, terminate the ring signal, and connect the phone to the line by releasing the relay.

 

[FvM]

I agree with the suggested sequence, except for a minor detail. Instead of unconnecting the phone on incoming call, you would better have it initially unconnected to avoid spurious ring signals. This makes the system operate more like a private branch exchange. In return, the system must detect telephone off-hook state and connect the land line. An internal loop supply isn't necessary for this operation, just a low voltage continuity test towards the telephone.

Correction to my previous post since I couldn't edit it

Time limit for editing posts was changed to 30 minutes in september. I only see subtile differences between both descriptions, by the way.

 

[pradeep_k_b]

@crutschow & FvM...Thank you very much for your suggestion,Let me list the sequence and kindly let me if I am wrong in some points

1-We are using a SPDT relay,the receiver phone will connect to NO and the device will connect to NC (SPDT1)
2-There will be another SPDT relay(SPDT2),this one will use to connect the locally generated ringing signal to the receiver phone.Normally it will connect to NO
3-The ringing signal will detect by the device and will ask the caller to enter password and when he enter the password the device will decode using a the DTMF decoder and will compare it with the password that already saved in the device
4-If the password is correct then the SPDT2 will turn on and will send the ringing signal to receiver phone
5-When the receiver attend the phone call then SPDT2 will turn off and SPDT1 will turn on,and direct connection will create between caller and receiver
6-When the receiver finished the call then the SPDT1 will return back to the previous condition (NO connection)

Am I correct?...But I think for to implement a system like this there need to check continuously the voltage level of the phone line using the micrcontroller,and with respect to the voltage level we need to take different action.The initial automatic answering and ringing signal generation will be comparatively easier but after this step we need to detect when the receiver pick up the phone,how can I do this?

 

[Audioguru]

You detect that a phone has gone off-hook when it draws current. When the phone draws current then it creates a voltage drop in a resistor in series with it. The resistor value can be very small so that it does not affect the voice levels and an opamp can detect the resulting small voltage drop.

 

[crutschow]

I agree with the suggested sequence, except for a minor detail. Instead of unconnecting the phone on incoming call, you would better have it initially unconnected to avoid spurious ring signals. This makes the system operate more like a private branch exchange. In return, the system must detect telephone off-hook state and connect the land line. An internal loop supply isn't necessary for this operation, just a low voltage continuity test towards the telephone.


Time limit for editing posts was changed to 30 minutes in september. I only see subtile differences between both descriptions, by the way.

I didn't unconnect the phone because it simplifies the circuit and doesn't require detecting when the phone hangs up after a call.

And if the phone is disconnected you also need to detect when the phone is lifted for an outgoing call and then connect it.

I thought the circuit could react fast enough so that the feedthrough ring would be short or inaudible.

My edited version is significantly different since the original did not mentioned a provision for an outgoing call.

I assume there's a reason for the 30 minute limit for editing but it's rather annoying. The other forums I visit don't have that restriction.

 

[BradtheRad]

I assume there's a reason for the 30 minute limit for editing but it's rather annoying. The other forums I visit don't have that restriction.

The reason for the time window, is because the administrators were seeing instances where spammers would let their posts sit at the forum for a while, then go back and re-edit them, putting in ads and contact info and whatnot.

To put the lid on such spammers, the administrators realized they had to restrict the time allowed, for members to go back and re-edit their posts. Currently it's 30 or 60 minutes.

 

[FvM]

It sometimes realize that my previous answer to post has been incomplete or even basically wrong. Considering the eventuality that others have already read it, it's just fair not to correct the post silently but to state explicitely what has been wrong.

I believe however that experienced users can use the edit feature responsible.

The reason for limiting the edit time window was, besides cutting the option of slipping spam, that we experienced a number of cases where forum members radically edited there previous posts or even deleted it completely which made the answers or the whole thread useless.