Phone ring detector circuit needed

Status
Not open for further replies.
J

jack0987

Member
Got a new land line phone but it does not have a blinker led that tells me there has been a call while I was away.
After I notice the blinking, I then just reset it.

Anyone know of a circuit that I can build that will do this?
 
Telephone ring voltage is AC and quite a big voltage. If you are in the UK, the ring voltage, between pins 3 and 5 of a phone socket, will be dc and won't have significant ac voltage, until the phone rings.

The ac voltage when the phone rings will be at least 30 V ac. You could just rectify that and feed a small 48 V dc relay.

It's probably against the regulations to connect anything home-made to the phone line. You should certainly not connect anything that could put power back into the phone line. If you don't put any voltage back into the phone line, and your phone does what you want, I very much doubt if anyone will notice that anything connected isn't approved.
 
OK. I have made a little progress.

The left side of the optocoupler of this circuit partly reworked by me with components I have and appears to be working for me.
The right side was replaced with a led to test. It comes on when ringing which is what I want to start.



On the left side, what can I do to improve this circuit?
The original circuit called for a 0.82uf cap but I only had a 0.47uf ones. How do I increase the cap value? Place two in parallel?
 
Two caps in parallel will add their capacitances, so it will give you around 0.94 µF.

If it's working, there is no need to change things.

The only possible problem that you could have is that there is not really a minimum voltage or time for the ring to make the relay operate. If you were to put 10 kOhms in parallel with the input to the opto coupler, then you would know that it would not work below around 5 V.

The circuit is arranged so that the relay will normally be on, and it will turn off when the phone rings. Is that what you wanted?

I don't think that the 1kOhm resistor is performing any useful function. It could be a link instead.
 
Phone lines are very noisy. You cannot rely on the first voltage transition above some ring-voltage threshold to indicate an actual incoming call.

At the secondary of the optocoupler, I would have an R-C integrator to increment a voltage over many cycles of ring voltage, followed by a set-reset latch, followed by an oscillator to make the LED flash, followed by an inverter for the correct logic polarity, followed by a transistor to drive the LED.

Sounds like a lot, but all of this can be done with one CD4093 quad Schmitt trigger NAND gate plus one 2N7000 small MOSFET, plus maybe four R's (one is for the LED) and 2 C's.

ak
 
Feeding the telco pair rectified to a latching relay will do it too.
Ringing is ~86VAC 20Hz
 
Last edited:
Just a hint that might persuade your design... When I was a kid I designed a phone ring detector that would detect and pickup the phone line before any of the other phones in the house had a chance to ring. HOW? There is a minimal 5V to 12V on the phone line to power the handset. just before the phone rings with the high voltage, the 5V to 12V drops to zero volts. When this happened, I would pick the line up, and flash a light to let me know the phone had picked up.

Why all of the trouble? I was a typical teenager in the mid 80's and I wanted to talk late at night on the phone with my "friends" without disturbing my parents.
 
Beau Schwabe said:
Why all of the trouble? I was a typical teenager in the mid 80's and I wanted to talk late at night on the phone with my "friends" without disturbing my parents.
Click to expand...

LOL. Great Reply. Been there. Would you like to share that circuit here for others?
 
OK. My progress so far.



For the set reset, I am thinking of a flip flop as soon as I figure out how to connect it???
Later I will make LED1 blink.
 
This is the general circuit idea.... keep in mind I was probably 15 when I put this together. The GND is from the supplied power GND which is an isolated 12V "wall wart" supply. Basic operation: R2, R3, and R5 form a voltage divider. R6 and T4's B-E are in parallel with R5. A small voltage (5V or so) is enough to turn 'ON' T4 which keeps T1 'OFF' during normal operation, If the voltage falls to 0V on the Phone line (which the old land lines used to do just before the phones would ring) then T4 turns 'OFF' allowing T1 to turn 'ON'. T2, T3, R7, R8, and C1 form an SCR latch, C1 (100nF) is for stability. When T1 turns 'ON' so does T2 and T3. If T1 turns 'OFF', T2 and T3 remain 'ON'. When the SCR is 'ON' the relay is energized and the LED turns 'ON' indicating the Relay is energized. The Relay in turn provides a lower resistance path equivalent to 'picking up' the phone. With the phone off the hook, the other phones in the house never had a chance to ring, because the High Voltage ring voltage never arrives. Thus making this circuit able to answer the phone before any other phones in the house. To reset the SCR simply press the SW1 switch which is a normally closed switch.
 
Where does the +12 V come from? A separate power supply not isolated from the phone line?

ak
 
First pass at the approach mentioned above. Non-inductive, galvanic isolation from the phone line, serious false-ring prevention, low power operation, LED flash oscillator included.

During ring voltage positive half-cycles, C1 charges through R4. When the voltage is far enough below Vcc, the U1A-U1B flipflop sets. This enables the U1C oscillator to flash the LED. U1D is necessary to keep the LED off when the circuit is reset.

With no details about your country or phone system, this is not a production-ready schematic. R1 and R2 are adjusted for the minimum input LED current needed to energize the output transistor briefly at the peak of each ring voltage cycle. The secondary current is a max of 130 uA at 12 V Vcc, so R1 and R2 probably can be higher than shown, like 47K to 100K each. Be sure to have the optocoupler's CTR (current transfer ratio) in your calculations.

Adjust R4 so that the flipflop changes state late in the first ring or somewhere in the second ring.

R3 sets the time delay after ringing stops before the ff is reset.

U1C is a standard Schmitt trigger oscillator. Adjust R6 for the flash frequency.

Adjust R7 for the desired LED current. Vcc can be anything between 5 V and 15 V.

If you want the circuit always to power up in the reset state, add a small capacitor across SW1.

The circuit should run for a long time on batteries. All circuit impedances are relatively high. When the circuit is sitting idle waiting for a ring, the only currents are the leakage currents through the active devices.

ak

 
Last edited:
Part of the third sentence - "...an isolated 12V "wall wart" supply."

"A separate power supply not isolated from the phone line?" - Not isolated from the phone line, but isolated from the mains
 
jack0987 said:
OK. My progress so far.

View attachment 134023

For the set reset, I am thinking of a flip flop as soon as I figure out how to connect it???
Later I will make LED1 blink.
Click to expand...

I have noticed that in my test circuit the LED1 comes on when the phone receiver is also lifted.
What change might I make so that it only comes on when it detects a ring? Please view attachment.
I also noticed that if I increase the capacitance the caller hears something like noise interference rather than a ringing tone on their end.
Please comment.
 

Neat. It's above my head right now but I am sure others will be pleased. Thanks.
 
Because your optocoupler has a Darlington transistor output stage, you should be able to increase the 22 K input resistor (reference designators - ?) significantly, like over 100 K and probably in the 470 K to 1 M range.

Also, you can decrease the value of the input capacitor to eliminate the reported noise. For example, your current input corner frequency is 15.4 Hz. Decreasing the cap to 0.1 uF and increasing the resistor to 100 K shifts the corner freq to 15.9 Hz, a trivial change. This will decrease significantly the load on the phone line during a ring, and increase the loading impedance across the line during talk by 5 x.

ak
 

Thanks so much. I did a quick test of changing the resistor value and the range for me seems to be about 60K to 100K.
I will know more after some additional testing.
 

Thanks. Just taking a look at his now. I do now have a sort of working circuit using a 7474 as a latch but at a glance this seems to be a better design.
 
The circuit will work with just about any "standard" optocoupler. The 4N35 is just one I grabbed from the design library. The 4N33 should be no problem. Again, it has so much gain that input resistors R1 and R2 can be increased to 100 K.

The 2N7000 is nice because it does not require an input resistor. It can be replaced with any NPN small signal transistor such as 2N2222, 3904, 4401, etc. For these, add a base resistor that yields a base current of 2 mA - ish. This will work well for any LED current up to 30 mA. The standard rule of thumb indicates a max collector current of 20 mA (10x the base current), but that is for transistors designed in the 1950's.

ak
 
Status
Not open for further replies.

Similar threads

  • Question
Filtering for a signal level detector
Replies
13
Views
3K
throbscottle
C
  • Locked
  • Question
Water Leak Detector equipment
Replies
13
Views
3K
rjenkinsgb
G
Is there such a thing as a 2 way switch for a 1/8" phone jack?
Replies
18
Views
719
ZipZapOuch
  • Question
Batteries overheating in Cordless House Phone
2
Replies
21
Views
5K
vidalv
S
  • Question
Circuit Board help
Replies
7
Views
3K
aliarifat794
A
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…