Door bell to payphone ringer

Thread Starter

Joe LaFever

Joined May 12, 2017
11
Hello,

I have a Elcotel Series - 5 payphone in my basement/man cave. The phone has a electronic ringer. I have a 16 Vac doorbell voltage regulator close by. I would like to wire the ringer on the phone to go off when someone presses my doorbell, thus alerting me to a visitor while I'm in my man cave. Any idea's on how I should do this? I have some obstacles:

-The voltage is constant at the regulator, if hooked up ringer will just ring continuously, how do I make it sense the doorbell switch to turn on/off? I believe run off doorbell switch.
-I want to hack to the ringer directly, I understand that a standard phone is 90 volts 20 Hz to get the "proper" ring sound, using the lower voltage should work. How will it sound? I have seen circuits that use DC with a 555 chip to simulate the 20 Hz. Do I rectify the AC to a DC voltage and use a 555 chip?
-The press of a doorbell is short lived, will the ringer from phone be the same? Will it just ring for a short second? How do I elongate the ringing sound?
-I'm assuming that when the upstairs bell rings there is a drop in voltage on the regulator, how will that affect the payphone ringer?

That's all my questions for now, any help is greatly, greatly, appreciated. Any hand drawn schematics also greatly appreciated!

Thank you!
Joe
 

AnalogKid

Joined Aug 1, 2013
8,439
In terms of the electronics, there are three functions:

1. Detect that the doorbell is on.

2. Turn this into a fixed-length control signal (pulse).

3. Use that pulse to enable a 20 Hz sinewave oscillator that drives the ringer in the phone.

4. Drive a small power transformer backwards to step up the sine wave to 90 Vac. While normally the ring voltage is superimposed on the -48 Vdc phone line on-hook loop voltage, that DC is blocked by a capacitor inside the phone and never reaches the ringer. So for your circuit you can ignore the -48 Vdc bias and just generate a simple sinewave that is plus and minus about ground.

None of these steps is difficult pr very complex, just things to be tended to. What is your skill set for getting electronic components and building a circuit on perf board or in a small box? Also, where are you located.

Recommendations:

0. Either use an old wall wart as a power source or we can use the doorbell AC transformer as the source.

1. A sense transistor or optical coupler in parallel with the doorbell will generate a 50/60 Hz square wave as long as the button is held down.

2. A monostable circuit extends the start of the 60 Hz burst and produces a single long pulse. This is the ringer on time.

3. A phase shift or bubba oscillator circuit makes a reasonable sine wave and is not tempermental. It can be tweaked over a small frequency range if your ringer is picky about that 20 Hz thing.

4. A very small power transformer can step up the oscillator circuit output to drive the phone. It will not be as efficient at 20 Hz as it is at 50/50 Hz, but the system is line powered so there is plenty of juice.

How's that for a start?

ak
 

Reloadron

Joined Jan 15, 2015
5,423
A Google of telephone ringer circuits will bring up some good circuits. One of the ways to get the needed high voltage (about 70 volts) is to use a 555 chip configured as a multivibrator driving a 120 VAC / 12.6 VAC step down transformer backwards. So the 555 drives the secondary side of the transformer. Even though you are using a square rather than sine wave to drive the transformer it works, To acheive what you want use a pair of 555 timers or a 556 which is a dual 555. A button push triggers the first 555 configured as a one shot making for a ten second pulse which enables the second 555 (astable 20 to 30 Hz oscillator). The second drives the transformer through a MOSFET. The entire affair could be powered off a 12 VDC 500 mA wall wart power supply. Pretty slick idea! :)

I see AK summed it up while I was screwing around trying to type and think. :)

Ron
 

Externet

Joined Nov 29, 2005
1,493
This is tricky : !
Not to be used for communication with the payphone. Wire only to its ringer and nothing else! And is not isolated! - there is no doorbell transformer involved.

120VAC-------------ringer------------10Kohm------------doorbutton-----------120VACneutral

I did this after analyzing the properties of telephone ringers in a circuit I built to link 2 fax machines without telco and works well.
Please delete if unsafe to have it posted.

Another doorbell transformer could be used as alternative, turning the 24 VAC into 120VAC to drive the payphone ringer.

24VAC house bell in parallel to 24V winding of second transformer =====> 120V primary to 10KOhm in series to payphone ringer. -For electronic telephone ringers-

For mechanical telephone ringers, the resistor may be somewhat less resistive.
 
Last edited:

AnalogKid

Joined Aug 1, 2013
8,439
Re-thinking things ...

If the ringer circuit is powered from the doorbell transformer *after* the switch, there there is no need to detect a button press. Power comes on, oscillator starts, ringer rings. The power section bulk filter capacitor is sized to run the ringer for a minimum amount of time. This eliminates the switch detection and the monostable timing circuit. Prelim schematic later.

ak
 

AnalogKid

Joined Aug 1, 2013
8,439
This is tricky : !
Not to be used for communication with the payphone. Wire only to its ringer and nothing else! And is not isolated! - there is no doorbell transformer involved.

120VAC-------------ringer------------10Kohm------------doorbutton-----------120VACneutral
Live non-isolated AC powerline through a doorbell pushbutton switch? I'll pass. Also, what about the real doorbell for the rest of the house? Did I miss something?

ak
 
Take a look here: http://www.camblab.com/datashts/pbmrspec.pdf

Any design of a ring generator could be detrimental IF the phone is picked up and the line is still ringing.
So, that's something to worry about.

from pdf said:
Output is NOT PROTECTED against shorts;
circuit must have series resistance of at least 300 Ω, normally part of ring trip cir-cuit. If the latter’s current sensing resistor is less than 300Ω, additional resistance must be added. Ring trip circuit should
terminate ringing within 200 ms of off-hook transition.
Ringing only the bell, by rewiring, is something that can be looked at too. Let's for the moment assume that,

Ok, so you need AC at 20 Hz. So, sine or square.

Your CADENCE would have to either be as long as button is depressed or you could create your own. e.g. 1.5 s on for every press or it continually rings when pressed.

True, most doorbels are 16 VAC. You can modify one of these to detect 16 VAC.

So with a 5 v supply and say a a ring generator, you can pretty much be there. The only other suggestion is some sort of cadence, like it rings for 1.5 seconds when pushed.

Liner timerblox, 555 timer, whatever.

Knowing the wiring scheme will help. You have to figure out where things terminate, You need access to the 16 VAC delivered to the bell. So, if the power heads to the doorbell and the switch heads to the bell, you have a problem unless someone used less conductors than they needed.

Look at this http://www.linear.com/product/LTC6993-1 part for the delay and on that same page is a reference to a de-bouncer using another part. Long delays with the 555 are problematic.
 

djsfantasi

Joined Apr 11, 2010
6,407
Just a suggestion for a quick test. If you applied the 16VAC to the ringer , does it ring?

I think the 20Hz signal is the nominal values, but the ringer responds to a broad range of frequencies. We used to use an AC transformer to ring onstage phones in the college drama club. It was a technical school, too. Ga Tech.
 

Externet

Joined Nov 29, 2005
1,493
Correct. The live mains on the ringer button is a 'no, thanks' warned in my post. I was explaining how simple can it be. Please moderate/delete the pertaining paragraph.

The second transformer keeps functionality of the existing doorbell and supplies a simple signal for the telephone ringer only. The telephony cadence is ignored, as it is not a telephony signaling. It is a doorbell pushbutton signaling.
 
But you can run a transformer backwards, so 16 VAC gives you 90 VAC (1:5.6) isolated. e.g. not 16:120
Somthiing like 120 VAC to 21 VAC. You can add and subtract too. So, when the 20 V winding sees 16 VAC, it generates about 90 VAC ISOLATED.

With party lines you had selective ringing based on frequency. I don't know if 60 Hz would work.
 

AnalogKid

Joined Aug 1, 2013
8,439
Here is a first pass at a circuit. T1 is the doorbell transformer. C1 is to be adjusted to give the correct minimum ring time. U1 is a power sine wave oscillator. The feedback network can be phase shift, bubba, Twin-T, or any other 180 degree network at 20 Hz. T2 is a small power transformer running backwards.

Note: this is a preliminary concept schematic, intended to anchor a discussion of a minimalist circuit that gets the job done.

ak
Doorbell-Phone-1-c.gif
 

Attachments

Thread Starter

Joe LaFever

Joined May 12, 2017
11
Here is a first pass at a circuit. T1 is the doorbell transformer. C1 is to be adjusted to give the correct minimum ring time. U1 is a power sine wave oscillator. The feedback network can be phase shift, bubba, Twin-T, or any other 180 degree network at 20 Hz. T2 is a small power transformer running backwards.

Note: this is a preliminary concept schematic, intended to anchor a discussion of a minimalist circuit that gets the job done.

ak
View attachment 126611

This is great stuff thank you. I'm going to take another look to all responses Monday and reply from work, run it by a couple people.
 

AnalogKid

Joined Aug 1, 2013
8,439
I debated whether or not to put a ground symbol on the schematic, because if you connect that point to earth ground the entire doorbell system no long is completely floating.

If someone thinks an LM386 can't make enough output current for your ringer, there are many other single-chip audio power amps to choose from.

ak
 

rherber1

Joined Jan 6, 2008
26
Hello,

I have a Elcotel Series - 5 payphone in my basement/man cave. The phone has a electronic ringer. I have a 16 Vac doorbell voltage regulator close by. I would like to wire the ringer on the phone to go off when someone presses my doorbell, thus alerting me to a visitor while I'm in my man cave. Any idea's on how I should do this? I have some obstacles:

-The voltage is constant at the regulator, if hooked up ringer will just ring continuously, how do I make it sense the doorbell switch to turn on/off? I believe run off doorbell switch.
-I want to hack to the ringer directly, I understand that a standard phone is 90 volts 20 Hz to get the "proper" ring sound, using the lower voltage should work. How will it sound? I have seen circuits that use DC with a 555 chip to simulate the 20 Hz. Do I rectify the AC to a DC voltage and use a 555 chip?
-The press of a doorbell is short lived, will the ringer from phone be the same? Will it just ring for a short second? How do I elongate the ringing sound?
-I'm assuming that when the upstairs bell rings there is a drop in voltage on the regulator, how will that affect the payphone ringer?

That's all my questions for now, any help is greatly, greatly, appreciated. Any hand drawn schematics also greatly appreciated!

Thank you!
Joe
I am not familiar with the Elcotel Series 5 payphone but in any case I suspect that the ringer transducer is just a piezo sounder.

If this is the case then I would use one of the common electronic ringer IC's such as the KA2411 or CS8205.

This page might give you some ideas http://www.eleccircuit.com/loud-ringer-for-phone-using-ka2411/

The incoming 90V ringing signal is simply reduced in amplitude by the series 1uF cap and 5k6 resistor before being rectified and filtered to provide a DC trigger signal to the IC. After that the IC takes care of driving the ringer transducer.
 

Thread Starter

Joe LaFever

Joined May 12, 2017
11
In terms of the electronics, there are three functions:

1. Detect that the doorbell is on.

2. Turn this into a fixed-length control signal (pulse).

3. Use that pulse to enable a 20 Hz sinewave oscillator that drives the ringer in the phone.

4. Drive a small power transformer backwards to step up the sine wave to 90 Vac. While normally the ring voltage is superimposed on the -48 Vdc phone line on-hook loop voltage, that DC is blocked by a capacitor inside the phone and never reaches the ringer. So for your circuit you can ignore the -48 Vdc bias and just generate a simple sinewave that is plus and minus about ground.

None of these steps is difficult pr very complex, just things to be tended to. What is your skill set for getting electronic components and building a circuit on perf board or in a small box? Also, where are you located.

Recommendations:

0. Either use an old wall wart as a power source or we can use the doorbell AC transformer as the source.

1. A sense transistor or optical coupler in parallel with the doorbell will generate a 50/60 Hz square wave as long as the button is held down.

2. A monostable circuit extends the start of the 60 Hz burst and produces a single long pulse. This is the ringer on time.

3. A phase shift or bubba oscillator circuit makes a reasonable sine wave and is not tempermental. It can be tweaked over a small frequency range if your ringer is picky about that 20 Hz thing.

4. A very small power transformer can step up the oscillator circuit output to drive the phone. It will not be as efficient at 20 Hz as it is at 50/50 Hz, but the system is line powered so there is plenty of juice.

How's that for a start?

ak
I have a Bachelors in EE from University at Buffalo. I live in Buffalo, NY.
 

Thread Starter

Joe LaFever

Joined May 12, 2017
11
Just a suggestion for a quick test. If you applied the 16VAC to the ringer , does it ring?

I think the 20Hz signal is the nominal values, but the ringer responds to a broad range of frequencies. We used to use an AC transformer to ring onstage phones in the college drama club. It was a technical school, too. Ga Tech.
Fair question and I'm going to try that just to see what happens.
 
Top