Old telephone ringer stompbox

[GibsonGM]

Rather than try to build a generalized 20Hz oscillator just to characterize the coil you could just try your luck with an NE555 set-up as a 20Hz oscillator.   The NE555 output is not AC, it has a DC component of Vcc/2 so you need to add a series cap on the output whatever you do.  For the non resonating cap case the output would be a large electro say 100uF.  For the resonating cap case the cap would a smaller value perhaps around 1uF to 3uF (value needs to be tuned).  Set the power rail to 10V to 15V and see how you go.

This is kind of where we were at.

If you have and know how to use a meter - What is the measured resistance of the coil, does it match what's printed on the unit?   Or does the meter read lowish resistance, then go up up up up up up to over 1Meg?  (that will indicate a series cap)

If you hit the terminals with a 9V battery rapidly on/off over and over, does it 'ring'? (pretend you are the oscillator)  You could try this w/2 batteries in series, 18V, if 9 doesn't work.    If you can make it work on 9 or 18V, at that point I would measure the current the coil is drawing (being overly 'safe' here, but don't energize the thing for long periods, just a 3-second 'on' and read).    That will tell us something.  Ask if you are unsure how to measure resistance/current...

Sometimes a mechanical 'hack it' approach works...   

[SeneX225]

When you say it rings once, I'm assuming you mean you put DC on it and it rings, yes?

Yes. Also, if connected to an AC power supply it rings continuously after physically clicking the hammer once.

Rather than try to build a generalized 20Hz oscillator just to characterize the coil you could just try your luck with an NE555 set-up as a 20Hz oscillator.   The NE555 output is not AC, it has a DC component of Vcc/2 so you need to add a series cap on the output whatever you do.  For the non resonating cap case the output would be a large electro say 100uF.  For the resonating cap case the cap would a smaller value perhaps around 1uF to 3uF (value needs to be tuned).  Set the power rail to 10V to 15V and see how you go.

If you resonate the coil inductance with the series cap then you will get more current to the coil with a low voltage.

This, I have troubles understanding, so please bear with me. I need an to build NE555 oscillator with frequency of 20Hz, current of 5mA ac to 10mA and output voltage of about 10V, is everything correct? Then I need to connect either a 100uf electrolytic capacitor or 1uF to 3uF film (?) one to output and connect the ringer to that?

If you have and know how to use a meter - What is the measured resistance of the coil, does it match what's printed on the unit?   Or does the meter read lowish resistance, then go up up up up up up to over 1Meg?  (that will indicate a series cap)

Meter reads 1700 Ohm flat without increasing.

If you hit the terminals with a 9V battery rapidly on/off over and over, does it 'ring'? (pretend you are the oscillator) 

It does not, no matter how fast or slow I'm hitting the contacts maybe I'm just a natural sawtooth.

If you can make it work on 9 or 18V, at that point I would measure the current the coil is drawing (being overly 'safe' here, but don't energize the thing for long periods, just a 3-second 'on' and read).    That will tell us something.  Ask if you are unsure how to measure resistance/current...

I did this regardless, and if my measuring is correct (which is quite possible can be not) it reads from 0.8 to 3.2 mA

[GibsonGM]

 A 555 outputs a 0 to I think 2/3 Vcc pulse, which is DC, it doesn't cross zero and go negative. No reason for a cap on the output that I can see.    So on 12V supply, you'd have nearly 8V output.   You need to find the lowest DC voltage that will drive the hammer and then run the 555 so its output will be in the right zone (1/3 higher supply to the 555 than the output you want).

At 1700 ohms, it's only drawing around 5mA...at 12V it would draw only 7.  The 555 can source 200mA!!  I believe if you run the 555 at high enough voltage, but LESS THAN 15V, you will be ok without an additional transistor provided you are getting enough output voltage from the chip.  These are the little problems we encounter in retrofitting things like this!     

If it rings on a AC adapter, you could use that without any modification, and make something (footswitch) to turn it on and off, which is probably easier.  What is the AC voltage you are using to get it to ring properly?  And I wonder what this 'physically clicking the hammer' means...anyone? Sticky, or how it's made??  If you need to touch it to get it moving, that's a big problem that needs solving before going ahead, I'd think.

[Rob Strand]

Yes. Also, if connected to an AC power supply it rings continuously after physically clicking the hammer once.

So what that tells me is 9V AC (rms) is not enough to drive that bell.  It could be marginal.   Also when you first apply AC current there is a current surge and that might be just enough to push it into working.   From the results of that test I'd guess you need at least 18V AC for it to work.

This, I have troubles understanding, so please bear with me. I need an to build NE555 oscillator with frequency of 20Hz, current of 5mA ac to 10mA and output voltage of about 10V, is everything correct? Then I need to connect either a 100uf electrolytic capacitor or 1uF to 3uF film (?) one to output and connect the ringer to that?

Just follow the schematic.

With the 100uF cap you will not get maximum voltage output.   It still might work.

However with the smaller cap you can get more voltage across the coil because the cap resonates with coil.  It acts like a voltage booster.   If you look the waveforms I posted earlier we are getting +/- 27V from a 15V supply; a finer point is the 20Hz fundamental is less than 27V pk.   The thing with this method is you need to play with the cap value to match your coil.    There's more chance of this method working.    However it still might not be enough voltage.   We simply don't know what minimum voltage the coil requires to ring.

The circuit should work with a crappy electro with the + electrode connected to the NE555.   Using a poly is also fine.  Because you have to play with the value you might need to start with 1uF then add a another 1uF in parallel etc.    You might need finer steps like 470nF.  Measure the DC current on the supply and choose the cap which draws the most current.   (You might even need to go below 1uF or above 10uF.)

No reason for a cap on the output that I can see

Ideally the coil requires AC.  Without the cap the NE555 output is AC plus a DC offset of VCC/2; think of the output NE555 not swinging negative, it stops at 0V.   The DC component causes an unnecessary DC current to flow through the bell and may actually affect how the bell functions.

For the resonant cap case the cap gives a voltage boost at the bell terminals (see waveforms).   I suspect this is going to be necessary to get enough output.


These waveforms show,
- To be within +/-10% of the maximum bell current the resonant cap needs to be
   within Coptimum -19% to Coptimum +25%
   So if you choose standard cap steps of 2.7u, 3.3u, 3.9u etc. you should get close to optimum.
- For the 100uF case you can see the output voltage is square but the bell voltage and current are lower
  than if you choose the cap to resonate with the coil.   Hence this version is less likely to work.



A few more snippets:

* Minimum voltage required for bell * (observed)
https://forums.raspberrypi.com/viewtopic.php?t=328297
"Voltage isn't critical, they were designed to work on the end of long cable runs, 24Vac gets them ringing but is probably below minimum spec"

At 13:40 this guy shows the phone bell working with 25Vrms 60Hz
https://www.youtube.com/watch?v=E2ytI7DknaY

* Minimum current * (spec)
https://maker.pro/forums/threads/aus-telephone-voltages-and-frequencies.226865/
"A call indicating device shall operate satisfactorily on a minimum of 4mA of
17Hz ring signal, 5mA of 25Hz ring signal and 6mA of 50Hz ring signal."

I suspect there's zillions of bell designs out there so who knows what voltage the specific bell you have actually operates at.

My feeling is the NE555 design with the resonant cap, and operating from 15V, might work.   However, the output is a little lower than the 24VAC in the post I referenced; and of course his experience is based on a different bell.   Finding the minimum voltage a specific bell requires can only be found by experiment!  Specs won't even help because operating at minimum is doing the wrong thing and getting away with it.  [My best estimate for the minimum spec is about 40Vrms using info from multiple sources, so the 24V AC minimum for the bell is well under the minimum spec.]

As a side note, phone bells are specifically prohibited from operate below 10Vrms.    If the minimum operating spec is 40Vrms then it's no too far fetched to assume the minimum bell voltage will be between 10Vrms and 40Vrms, which is 20Vrms to 25Vrms.  That agrees with the 24Vrms  in the post.  It also means the NE555 circuit with a 15V supply, which produces 16.4Vrms is in the zone where it works/doesn't work - so very much pot luck of it working.

[SeneX225]

If it rings on a AC adapter, you could use that without any modification, and make something (footswitch) to turn it on and off, which is probably easier. 

I thought on that, however it would require s footswitch with a double-arm physical lever to work, which I have no idea if it exists or how to make one if it doesn't.

What is the AC voltage you are using to get it to ring properly? 

It works on 9V 1A AC.

And I wonder what this 'physically clicking the hammer' means...anyone? Sticky, or how it's made??  If you need to touch it to get it moving, that's a big problem that needs solving before going ahead, I'd think.

I made a post about it a little while ago. By "touching" I mean pressing plates connected to hammer to make (or break) contact.

While testing this an interesting development occurred: I've plugged in an AC 9V 1A power supply mentioned earlier and accidentally touched the plate connected to the hammer, and it rang... and did not stop! Not until I touched another side of said plate, which stopped the hammer.

[GibsonGM]

Ideally the coil requires AC.  Without the cap the NE555 output is AC plus a DC offset of VCC/2; think of the output NE555 not swinging negative, it stops at 0V.   The DC component causes an unnecessary DC current to flow through the bell and may actually affect how the bell functions.

For the resonant cap case the cap gives a voltage boost at the bell terminals (see waveforms).   I suspect this is going to be necessary to get enough output.

Thanks, Rob, that clarified the issue for me.  You have far more experience w/this kind of circuit than I do!  I didn't consider any problems that DC current could cause.    Sene, listen to Rob, as he knows much more than I do  

I'm trying to 'hack' this, not to make it function as designed.   Now I'm thinking about those 30+ volt computer power bricks most of us have lying around, and a 555 with a power transistor on its output...you would need to run the 555 at a safe (lower) voltage, of course - those bricks often have multiple outputs, with a tap for lower voltage tho.

[duck_arse]

that little black thumbwheel at the top between of the bells - that's meant to dull or kill the ringing. have you twiddled that to see if you get easier/harder hammering action?

[Vandoog]

Fun project. I did a similar thing with an old Alarm bell/contact mic and called it the "Hellringer" Was super simple for me, just a momentary stomp switch to a 9v jack.

https://youtube.com/shorts/hnVSOWX-DUQ?feature=share

[SeneX225]

After a bit of looking around I found this schematic:



Will it fit the need here (after fitting it with a capacitor on the output as Rob suggested, of course)?

Also, I found this as a backup plan (even though my math is very rusty):

that little black thumbwheel at the top between of the bells - that's meant to dull or kill the ringing. have you twiddled that to see if you get easier/harder hammering action?

It makes the action harder, because it gradually reduces swinging length of the hammer.

Fun project. I did a similar thing with an old Alarm bell/contact mic and called it the "Hellringer" Was super simple for me, just a momentary stomp switch to a 9v jack.

https://youtube.com/shorts/hnVSOWX-DUQ?feature=share

This is exactly the result I want to achieve! I even thought calling it the Deadringer but after about a minute I realized how cheesy it sounds!

[Rob Strand]

I'm trying to 'hack' this, not to make it function as designed.   Now I'm thinking about those 30+ volt computer power bricks most of us have lying around, and a 555 with a power transistor on its output...you would need to run the 555 at a safe (lower) voltage, of course - those bricks often have multiple outputs, with a tap for lower voltage tho.

If you have a single transistor booster open collector output you end-up with DC on the output and it creates problems with adding the series cap.

One issue is finding parts which will handle 30V.   The NE555 poops out at 16V or so.   Opamps and transistors handle 30V.

The resonant cap is a trick to get more output voltage within the NE555 limits.

A different spin on the 30V power brick would to use an opamp.  You can add a two transistor output booster (like a headphone amp) to prevent the opamp overheating.  Wire that opamp as a sine oscillator (a square output Schmitt-trigger oscillator might do).  Due to the single supply it will need an output cap.   The thing is we still only get +/- 14V output.   With a resonant output cap you will get quite a bit more.   (It might be possible to set-up a sine oscillator using the output cap and the bell coil inductance as the resonant circuit for the oscillator but that's going to take a bit of effort.)

I don't know what power supply rail the OP intended on using.  In order to come-up with a design we need to know what voltage the bell operates at and what supply rails we are dealing with.   It's all too open ended.

Here's an example of a crude open collector output stage but with a collector pull-up resistor to provide a return current path for the bell.
(It's not a working circuit. It just demonstrates what you can get out of a 30V DC rail; with some loss of swing due to the 470R.)

[GibsonGM]

That's really cool, Rob!  File that as 555 Trick No. 422.  Makes perfect sense. 

Yes, there isn't enough digging in (yet) to understand what makes the most sense here. A major disadvantage of the 'good old days' tech (tho I really like it a lot) is the need for higher voltages/currents in their operation, which are kind of hard to approximate.   

My father in law runs a BIG model train, using 16VAC as the foundation for everything.  He doesn't understand why everything I show him to run lights, sounds, track switching (solenoid snap switches and so on)...requires WORK to design and is klutzy.   He will not go to 5V or 9V DC (stubborn).  He wants time delays for things like track switching - you can imagine my frustration when trying to set up a 555 starting from 16VAC at EACH point rather than some central DC supply!  LOL   

[Rob Strand]

My father in law runs a BIG model train, using 16VAC as the foundation for everything.  He doesn't understand why everything I show him to run lights, sounds, track switching (solenoid snap switches and so on)...requires WORK to design and is klutzy.   He will not go to 5V or 9V DC (stubborn).  He wants time delays for things like track switching - you can imagine my frustration when trying to set up a 555 starting from 16VAC at EACH point rather than some central DC supply!  LOL 

That stuff is awesome but an insane amount of work.   But yes, a 16VAC standard is a headache for modern electronics - adds a lot of baggage to any devices.  I've still got some 12VDC + 15/16VAC supplies for trains/slotcars.

[Rob Strand]

As far as old technology solutions.   Another angle to drive the bell with the correct voltage is to use a mains transformer in reverse.   If you have a 115V rms 60Hz transformer, then at 20Hz you will get an output voltage of about 115 * (20/60) = 38V rms.

In reality the transformer turns ratio is rigged to compensate for regulation.  This means operating the transformer in reverse will mean the output is say 20% less than the basic calculations, so 32Vrms.   While it doesn't make the required minimum 40Vrms it's likely to be enough to at least drive the bell.

You could have a 9V DC supply running an LM386 amplifier configured as a 20Hz sine wave oscillator.   If we allows say 3V pk output, that's 2.1V rms output.  So we need a transformer ratio of 38/2.1 = 18.   If the transformer was 115V rms then the secondary voltage required from the transformer would be 115/18 = 6.4V rms.   A 4.5Vrms transformer secondary will give a bit more margin for error.  The secondary shouldn't be over driven to get more output voltage as the core can saturate.   If the transformer is 50Hz/60Hz then it will give you a but more slack.   Another way to get more slack is to operate the bell at 25Hz.  High enough frequency to cut some slack for the transformer but not enough to make the bell sound bad.

Probably need a 2.5VA to 4VA transformer to avoid resistive losses.

The bottom line is this idea barely works.

[GibsonGM]

My father in law runs a BIG model train, using 16VAC as the foundation for everything.  He doesn't understand why everything I show him to run lights, sounds, track switching (solenoid snap switches and so on)...requires WORK to design and is klutzy.   He will not go to 5V or 9V DC (stubborn).  He wants time delays for things like track switching - you can imagine my frustration when trying to set up a 555 starting from 16VAC at EACH point rather than some central DC supply!  LOL 

That stuff is awesome but an insane amount of work.   But yes, a 16VAC standard is a headache for modern electronics - adds a lot of baggage to any devices.  I've still got some 12VDC + 15/16VAC supplies for trains/slotcars.

I have even GIVEN him a 12V/1A DC power supply, but he won't run his 'peripherals' from it; doesn't understand why DC is so much easier to work with (cuz I'm the one doing the work ha ha).   We rectify the 16VAC for every item he wants...and at 84, I don't think anyone can ever change his mind, LOL.   He probably has an old-time way to run the OP's bell, too, ha ha.     Way OT, sorry.

[SeneX225]

I'm back with yet another blunder.

I've converted that earlier schematic into a layout in DYILC and put it all together on actual breadboard, but sadly it doesn't run even without connecting the bell with capacitor. I cannot see what exactly is wrong, but yet again, I know very little of the works as it is.

Here's the schematic:



Here's the layout:

[Phend]

Maybe this will help.

https://electronics.stackexchange.com/questions/502133/make-old-telephone-ring

https://vetco.net/products/nte1648-ic-telephone-ringer

[amptramp]

There is another way to do this that leaves the poor old 555 in peace.

If you have 60 Hz power, set up a counter to count three positive alternations of voltage then allow the ringer to get current via an SCR.  Then it waits another three pulses and so on.  If you are on 50 Hz power, count the alternations at the output of a bridge rectifier and hit the ringer through the SCR after five counts.  The latter is more generalized because you can use a divide by five / divide by six to get a universal unit that can run on 50 Hz or 60 Hz.  You may have to use a transformer or resistor to get the correct voltage to the coil.  There is no worry about having the coil see the correct frequency because it is set by the power line.

[PRR]

>  https://vetco.net/products/nte1648-ic-telephone-ringer

That's to "ring" a piezo beeper, not an honest electro-mechanical dinger.

This seriously is a case for a pre-packaged solution.
https://www.surplussales.com/Communications/Comms-1.html
Ringing Generator  20 Hz (TEL) TA182/U  $9

[Rob Strand]

There's no guarantee any of the NE555 designs will work.   The intent of those circuits is to see whether you can get away with it.  Also the only hope is to run it at 15V DC.  If they don't work because of a circuit error then you need to debug it.   Check some basic things like: is there 20Hz coming out of the NE555.

The key piece of info missing is the minimum voltage required to operate the bell.   That needs to be worked out but without the right equipment you have to build something.

That's to "ring" a piezo beeper, not an honest electro-mechanical dinger.

Those devices were intended to *replace* the mechanical bells.  It's the ring "tone" you hear from a modern phone, which comes from a speaker.

[Phend]

ok, back to the olden days to deliver what worked. The....
Western Electric field Telephone crank magneto / generator.

I remember the old days at the ski hill, where the lift operator at the bottom would use this to contact the operator at the top.
Aah, the good ole days of commutator communication.
(Either crank with yer foot or incorporate a drill motor)