Detecting when a calling party hangs up the telephone line

[Linden]

Hi,

So ive got the ring detection circuit done fine.

Problem now is when I 'answer' the phone (by detecting the ring signal
and making a microcontroller switch a relay to connect to a line
transformer) I cant get the solid state relay or even a mechanical relay
to switch off and disconnect the circuit.

I need to know how can I detect when the other party has hang up the
phone so I can disconnect the line. Otherwise the line will stay
connected indefinately and no one will be able to get through.

Help appreciated, thanks !

 

[Dbowey]

Linden posted:


<< So ive got the ring detection circuit done fine.

Problem now is when I 'answer' the phone (by detecting the ring signal
and making a microcontroller switch a relay to connect to a line
transformer) I cant get the solid state relay or even a mechanical relay
to switch off and disconnect the circuit.

I need to know how can I detect when the other party has hang up the
phone so I can disconnect the line. Otherwise the line will stay
connected indefinately and no one will be able to get through.
There is no local change of state when the far-end hangs-up.

Build a device to detect the voiceband signal amplitude, and use it to open the
loop after there is no signal detected for several seconds. I would back that
up with a timer to open the loop after about 3 minutes after the initial
answer, no matter what.

If you need some ideas, email me.

Don


Don

 

[Dbowey]

Jamie posted: last time i knew the polarity switches.

In which country or system?

Don

 

[Chris Osborn]

I need to know how can I detect when the other party has hang up the
phone so I can disconnect the line. Otherwise the line will stay
connected indefinately and no one will be able to get through.

All I've ever been able to do is wait for the dial tone to come back.

 

[Jamie]

last time i knew the polarity switches.

 

[Linden]

All I've ever been able to do is wait for the dial tone to come back.

I'm in Australia. I've heard some systems use polarity reversal im not
sure about here and I dont really want to have to make something to
detect dial tone frequencies.

With all the telephone operated devices out there surely there is an
easy method to detect hang-up.

I may end up just forcing the other side to press a # to hang up.
Otherwise time-out after not receiving a key press for a minute or two.

- Linden.

 

[Dbowey]

Linden posted:

<< > All I've ever been able to do is wait for the dial tone to come back.

I'm in Australia. I've heard some systems use polarity reversal im not
sure about here and I dont really want to have to make something to
detect dial tone frequencies.

With all the telephone operated devices out there surely there is an
easy method to detect hang-up.

I may end up just forcing the other side to press a # to hang up.
Otherwise time-out after not receiving a key press for a minute or two.
In the US there are no central offices that normally send polarity reversal on
LINES. There are offices that send a reversal on TRUNKS, for PBX and some
multi-function systems.

It isn't likely, but you can easily test if you receive polarity reversal from
the C.O. by puting a voltmeter across the line while the far end hangs-up. Let
us know what you find.

Don

 

[Bob Masta]

Hi,

So ive got the ring detection circuit done fine.

Problem now is when I 'answer' the phone (by detecting the ring signal
and making a microcontroller switch a relay to connect to a line
transformer) I cant get the solid state relay or even a mechanical relay
to switch off and disconnect the circuit.

I need to know how can I detect when the other party has hang up the
phone so I can disconnect the line. Otherwise the line will stay
connected indefinately and no one will be able to get through.

Help appreciated, thanks !

I think this may be non-trivial. I've got an answering machine
that purported to not save anything in the event that the
caller hung up without leaving a message. But it usually gets
confused by ambient line noise and records that until the
dial tone comes back. I think if you set the threshold too low
you get this symptom; set it too high, and soft-voiced callers
won't register.


Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com

 

[Linden]

I think this may be non-trivial. I've got an answering machine
that purported to not save anything in the event that the
caller hung up without leaving a message. But it usually gets
confused by ambient line noise and records that until the
dial tone comes back. I think if you set the threshold too low
you get this symptom; set it too high, and soft-voiced callers
won't register.

Hmm, no polarity reversal when the call is hung up. 47.5V on hook and
down to 1.5V when on-hook and connected to the line transformer.

Still stays at 1.5V when other caller hangs up. Need to disconnect line
from the transformer to go back to 47.5V off hook condition.

- Linden.

 

[Dbowey]

Linden posted:

(snip)
<< Hmm, no polarity reversal when the call is hung up. 47.5V on hook and
down to 1.5V when on-hook and connected to the line transformer.

Still stays at 1.5V when other caller hangs up. Need to disconnect line
from the transformer to go back to 47.5V off hook condition.
Which brings us back to needing a solution: Use a delay or use audio detection
or a combination of both.

Your voltage readings indicate a problem in your design. The DC resistance of
your equipment is too low. The equipment puting the line off hook should place
less than about 230 Ohms DC across the line.

Don

 

[Rich Grise]

Linden posted:

(snip)
<< Hmm, no polarity reversal when the call is hung up. 47.5V on hook and
down to 1.5V when on-hook and connected to the line transformer.

Still stays at 1.5V when other caller hangs up. Need to disconnect line
from the transformer to go back to 47.5V off hook condition.

Which brings us back to needing a solution: Use a delay or use audio detection
or a combination of both.

Your voltage readings indicate a problem in your design. The DC resistance of
your equipment is too low. The equipment puting the line off hook should place
less than about 230 Ohms DC across the line.

I'm sure you mean "less load" - 230 ohms is the _minimum_ resistance,
_maximum_ load. I've heard that you should use a 600 ohm resistor for
"hold", so I'd say that would be a minimum for the resistance value.

But detecting when the far end hangs up, you'd have to ask the telco.

Good Luck!
Rich

 

[Dbowey]

Rich posted:

Linden posted:

(snip)
<< Hmm, no polarity reversal when the call is hung up. 47.5V on hook and
down to 1.5V when on-hook and connected to the line transformer.

Still stays at 1.5V when other caller hangs up. Need to disconnect line
from the transformer to go back to 47.5V off hook condition.

Which brings us back to needing a solution: Use a delay or use audio detection
or a combination of both.

Your voltage readings indicate a problem in your design. The DC resistance of
your equipment is too low. The equipment puting the line off hook should place
less than about 230 Ohms DC across the line.

I'm sure you mean "less load" - 230 ohms is the _minimum_ resistance,
_maximum_ load. I've heard that you should use a 600 ohm resistor for
"hold", so I'd say that would be a minimum for the resistance value.

But detecting when the far end hangs up, you'd have to ask the telco.

It's an age thing--- I meant to say "...should place no less...." Thanks.
Also my memory failed me about the 230 Ohms; It is 330 Ohms - see below. I
really hate to add to the confusion perfusion.

I think the 600 Ohm idea comes from info passing through too many hands, and
may be a result of the incorrect insight that telephone lines are all 600 Ohms
(impedance). This is one of the faults of the internet; there is a great deal
of information available, but so much of it is wrong or at least twisted.

On a long enough loop from a C.O., 600 Ohms DC resistance *might not* place a
loop in the off-hook condition. If a 330 Ohm closure is used to answer the
call, and then 600 Ohms resistance is placed on the line before removing the
330 Ohms, then the 600 Ohms will probably hold the loop in off-hook. Given a
properly zoned phone line, a 330 Ohm resistance will always provide at least
18.5 mA. of loop current at least 6.6 Volts across the resistance.

By the way, 330 Ohms is the Standard resistance for a rotary dial phone and 430
Ohms is the Standard resistance for a "touchtone" phone.

Don

 

[Bill Bowden]

I'm sure you mean "less load" - 230 ohms is the _minimum_ resistance,
_maximum_ load. I've heard that you should use a 600 ohm resistor for
"hold", so I'd say that would be a minimum for the resistance value.

But detecting when the far end hangs up, you'd have to ask the telco.

So, how does an answering machine know when to stop recording
after the caller hangs up?

-Bill

 

[Rich Grise]

I think the 600 Ohm idea comes from info passing through too many hands, and
may be a result of the incorrect insight that telephone lines are all 600 Ohms
(impedance). This is one of the faults of the internet; there is a great deal
of information available, but so much of it is wrong or at least twisted.

In all fairness, I heard it in 1988. ;-) And, of course, just took the
guy's word for it, because he was older and wiser, and answered with such
certainty that it _had_ to be right.

On a long enough loop from a C.O., 600 Ohms DC resistance *might not* place a
loop in the off-hook condition. If a 330 Ohm closure is used to answer the
call, and then 600 Ohms resistance is placed on the line before removing the
330 Ohms, then the 600 Ohms will probably hold the loop in off-hook. Given a
properly zoned phone line, a 330 Ohm resistance will always provide at least
18.5 mA. of loop current at least 6.6 Volts across the resistance.

By the way, 330 Ohms is the Standard resistance for a rotary dial phone and 430
Ohms is the Standard resistance for a "touchtone" phone.

Is that where that "Western Electric 430" comes from?


Thanks,
Rich

 

[Dbowey]

Rich posted:
(big snip)

Is that where that "Western Electric 430" comes from?

That could be...... 430pm was the WECo Day shift installation crew quitting
time, and held dearly by so many that they might have memorialized it. It's
the only 430 I can recall.

Ok, what's a 430?

Don