News:

"The phone is a remarkably complex, simple device,
and very rarely ever needs repairs, once you fix them." - Dan/Panther

Main Menu

Resistors in switchboard circuits

Started by RB, March 22, 2017, 03:01:28 PM

Previous topic - Next topic

Victor Laszlo

#15
Since you don't want to install them the way they would have been installed historically, the simplest way to do this is as follows:

Designate two magneto telephones as "subscribers" 1 & 2.

Install them at the ends of two pairs of wires.

Designate third telephone as "central."

Make or buy a wooden box with three telephone station jacks.

Wire 2 of the jacks across the line wires 1 & 2.

Install 2 line ringers at the box, across the lines of the two subscribers.

Install 2 ~60 Volt, A-C relays, each in series with a 2.0 mfd capacitor, in parallel with the above ringers.

Install 2, 6 Vdc relays that operate on the make contacts of the above relays.

Provide a 6 Vdc supply at the central box. (4 "D" cells)

Install 2,  LED's at the box, adjacent to the jacks that they serve.

Wire the D-C relays to lock up on their own contacts.

Provide a N/C (break) push button in the lock-up supply path to release those relays.

Wire the lamps on a second set of contacts of the relays through the local supply.

Connect the third (central) magneto phone to the third jack.

Use a short double-ended cord to plug into the station jacks.

OPERATION

Either subscriber (1 or 2) rings in to central.

Ringer activates.

Ring up relay operates.

Lamp relay operates and holds.

Lamp comes on.

Central takes down receiver and plugs from jack 3 into calling jack 1 or jack 2 and asks for instructions.

Calling party asks for called party.

Central plugs into called jack and retires lamp by operating push button.

Central rings called party and announces call.

Central plugs cord from jack 1 to jack 2 and hangs up.

Parties converse and ring off.

Central pulls down connection.

---------------

This scheme approximates a situation where there are two party lines connected via a central magneto office. In this scenario, there is only one party on each party line.

Materials:

Box w/ three jacks.

2 relays.

2 ringers.  WE type E1B or E1C are perfect. They have capacitors inside.

One double-ended two-conductor cord.  Surplus or make your own.

The LED's and latching relays are added to substitute for mechanical drops. If you can find drops, you can eliminate the LED's and the 6 Volt supply.

Small Western Electric wooden telegraph service boxes with three jacks are often found at phone shows or on Ebay.  If you want to know why there are lots of 3-jack boxes around, I can teach you how long-distance telegraph lines work in another thread.

No keys, coils, switches, or any other stuff required.



unbeldi

#16
A corded switchboard as described would indeed be a better start to gain some experience, and if nicely assembled would work much like some old real board.

Shown below are the Western Electric cord boards from a 1907 catalog.
The first needs a separate operator telephone, and the second has it built in.

unbeldi

A simple cord board like Victor described can easily be expanded simply by adding more line circuits beyond the two initially installed. The design also permits as many simultaneous talking paths as desired, by using multiple jumper cords.  Most cord boards and especially all cordless boards (with switches) had a fixed number of calls that could be conducted simultaneously, simply because each talking path requires some hardware and then the design involves a cost-versus-usage argument.

However, Victor's design has one drawback which I feel needs to be addressed.  It does not provide any means for the operator (central) to listen in on an established connection to determine whether the parties are still in communication if they forgot to ring off at the end of the call. The operator would have to interrupt the connection to plug herself into one of the lines and ask the subscriber. Forgetting to ring off was a common problem with magneto switchboards, many people even forgot to hang up the receiver to take the phone off-line.  To combat the latter problem on party-lines, each set was equipped with a "sure-ring" condenser, which provided a higher impedance to the set at ringing frequency to permit other ringers to still operate.

To permit the operator to listen in on a connection, you may simply provide two jacks for each line, wired in parallel. The second jack is then always available to patch the operator into the circuit.   It can also be used to connect more than just two lines into a call, creating a conference call.

poplar1

Quote from: unbeldi on April 08, 2017, 10:13:44 AM

...However, Victor's design has one drawback which I feel needs to be addressed.  It does not provide any means for the operator (central) to listen in on an established connection to determine whether the parties are still in communication if they forgot to ring off at the end of the call. The operator would have to interrupt the connection to plug herself into one of the lines and ask the subscriber.....

To permit the operator to listen in on a connection, you may simply provide two jacks for each line, wired in parallel. The second jack is then always available to patch the operator into the circuit.   It can also be used to connect more than just two lines into a call, creating a conference call.


Or, the extra jacks could be associated with each cord pair, rather than each line. Notice that in the top photo, there are 4 additional jacks, one per cord pair. The single cord (with some kind of designation above it -- TK? --) is plugged into the jack that is located above the 2 cords that are being use to connect lines 5 and 8.  That single cord is probably associated with the operator's desk stand. Thus, only 4 additional jacks (one per cord pair) are needed rather than 10 (one per line). This is consistent with the design of later cord boards where the talk and ring keys are associated with each cord pair.
Mets-en, c'est pas de l'onguent!

"C'est pas une restauration, c'est une rénovation."--François Martin.

unbeldi

#19
Quote from: poplar1 on April 08, 2017, 10:50:12 AM
Or, the extra jacks could be associated with each cord pair, rather than each line. Notice that in the top photo, there are 4 additional jacks, one per cord pair. The single cord (with some kind of designation above it -- TK? --) is plugged into the jack that is located above the 2 cords that are being use to connect lines 5 and 8.  That single cord is probably associated with the operator's desk stand. Thus, only 4 additional jacks (one per cord pair) are needed rather than 10 (one per line). This is consistent with the design of later cord boards where the talk and ring keys are associated with each cord pair.

Certainly. That's what historical boards do.  However, it makes the mechanical design more complex and introduces the limitation of a fixed number of talking paths.
It's the same trade-off that I mentioned earlier.  1/4" phono jacks are cheap and easy to add.

Another option is the addition of a simple two-pole push button for each line to connect TIP and SLEEVE of the line to the operator circuit.

unbeldi

Instead of using cords to make the connections between lines, some switchboards also used plugs that shorted two closely positioned jacks, arranged as part of an array of lines and talk circuits.

Pictured here is such a German switchboard by Siemens & Halske, a cordless switchboard.  I haven't seen any of this kind made by American manufacturers, whose 'cordless switchboards' used keys instead of plugs.

unbeldi

#21
I drew a circuit diagram of Victor's switchboard design.

I added two items:
- the previously discussed monitoring jack
- a resistor R1 to limit the current through the LED

R1 is needed to prevent the LED from burning out, and needs to be dimensioned according to the supply voltage, the forward voltage drop of the LED that is used (it depends greatly on the LED color), and its maximum current specification.  So, if it is a 20 mA diode and has a forward voltage of 1.8 V, and the supply voltage is 5V, then the resistor should be (5 – 1.8 ) / 0.02 in ohms, but I would use something larger as long as the LED still glows.  220 Ω should be good enough for any 20 mA LED.

You can also use a lamp instead of the LED, but in that case I would wire the lamp directly into the relay coil circuit. Doing so with an LED requires a relay that operates on only 20 mA current. Those probably get a little more expensive, and antique relays from old switchboards require more current than can be pushed through modern LEDs.

If I wanted to build this, I would also try to avoid using two relays, and instead use a semiconductor, or optocoupler, to detect ringing on the line.

I indicated a supply voltage of 5V in the diagram.  This is not important, but it is easy and cheap to find an old cell phone charger, that almost everyone these days has in a drawer of old stuff these days. It would make a good and cheap power supply.

The diagram shows only one station line circuit.  Of course you want to duplicate that for as many lines as your switchboard requires.  When building for more than two lines, you might also move the reset push button switch into each LED relay circuit, so that individual lines can be reset, in case more than one call alert comes in at the same time.

For only two lines it is not a big problem to use one ringer per line, but for more lines, the space problem multiplies rapidly, as can be seen in one of those historic switchboards that I showed earlier.  I would devise a method to use only one ringer that is triggered by a relay from any of the lines.  Of course, the ringers could be omitted altogether, if no audible alarm is needed and you are sitting in front of the board all day to play operator for your kids.  ;D  It might also be possible to replace that AC relay (RY1) with a DC relay and use it as a buzzer, as it will flutter about 16 times per second driven by the ringing voltage from the magneto generator of the calling station.  If the relay itself does not make enough noise, a second set of switch contacts could drive a DC buzzer or ringer.

unbeldi


Victor Laszlo

Great schematic! It's as though you read my mind! (or my post)

I would have included some way to monitor the line, if the application had been more...commercial... in nature, but since the system proposed has two subscribers it didn't seem to be immediately necessary.

There are other refinements to a real working magneto system, to be discussed if and when the OP returns to the thread.

RB

Hi Guys
Yup, still here, in spirit, anyway.
been busy doin grand babies, and life in general...
I am currently working on the repeat coil circuit.
it has to be switchable from LB, to CB...so takin me a bit to get the switches wired.
big fingers, small work space.
Thanks for all the wisdom!!!
will study the new diagrams, and post back, thanks again.

unbeldi

Quote from: RB on April 18, 2017, 03:57:30 PM
Hi Guys
Yup, still here, in spirit, anyway.
been busy doin grand babies, and life in general...
I am currently working on the repeat coil circuit.
it has to be switchable from LB, to CB...so takin me a bit to get the switches wired.
big fingers, small work space.
Thanks for all the wisdom!!!
will study the new diagrams, and post back, thanks again.

Switchable from LB to CB ?
Hmm.  Never heard of that.
Common battery PBX switchboards often had the option of installing a tie trunk unit to permit connection to an alternate magneto exchange or local magneto board.  Those tie trunk are dry and basically the same as a local battery line.
Is that what you mean ?

unbeldi

Here is an interesting and quite stylish looking cordless magneto board of ca. 1908 by the Century Telephone Construction Company.  This is almost the same time as the first cordless boards by the Western Electric Company, the 505 for CB and the 10-Line Magneto Switchboard for LB, both of which where clunky rectangular wood boxes.

Found in Telephone magazine.

unbeldi

#27
Another board of the time was this 10-line magneto board by Stromberg-Carlson.

RB

Good morning.
not sure what a tie trunk unit is...still learnin.
I do believe, however, that we are saying basically the same thing.
I have collected schematics for both LB and CB repeat coil configurations.
From those, I have combined "via 2 rotary switches", a solution which allows me to convert on the fly, from LB to CB.
I had to use 2 switches because I could not get my hands on the correct 1 switch solution. So, there we are.
Still in the wiring up stage, There are a lot of wires associated with this part, and my eyes are not gettin any better.
Should have it completed by end of weekend.
I intend to use this board as a teaching tool. As well as a fully functional switchboard. That is why I go to these extents.
The board will be able to support any basic 2 wire crank phone connections, "straight through the board".
And, if needed, the ability to insert the repeat coil, to isolate the two lines from each other.
Overkill you say??? sure it is, but who is judging? I can do this any way I choose. AND, it is fun as heck!!!
Well, not all fun, I hate that I need to work under a BIG-EYE mag lite to see well enuf to solder.
I WILL, however, emerge victorious!
Next week, if this weekend goes well, I will get back to the signaling circuit. ??? is that the right term?
Thanks to all of you for your interest/input. It is helping speed this up a bunch!

unbeldi

#29
A tie trunk, is just a telephone line connecting two switch boards.  For magneto service, it has no battery on it, and is essentially the same as a local battery telephone line.

A repeating coil is typically used for transmission between two dissimilar type of lines to avoid inductive noise problems.  An example is when one line is a ground-ringing line and the second rings across tip and ring only without the need for ground.
When building a switchboard, you will probably only have one type of line, so a repeating coil would not be needed.  When transmitting from a local battery line to a common battery line, you don't necessarily need a repeating coil, you can often simply use a capacitive bridge circuit.   Some switchboards with repeating coil circuits even had a feature key that eliminated the coil, called a zero-loss key, because the use of repeating coils does introduce transmission losses.

A signaling circuit is the part of a telecommunication equipment that receives or sends the information for call control.  So, that includes ringing, on- and off-hook detection (supervision), and address transmission (dialing).

I am a bit concerned about your project, to be honest.  I know how much fun it is, indeed, but it would be more fun to build something simpler that in the end works. Eventually, you might be able to find a cordless board for little money, they do exist and collectors do retire from their hobby as well—everyone MUST eventually.