Mini Payphone Controller Part 3

By Stan Schreier ATCA #2561

This is the last of two articles describing how to add additional features to the original Mini Payphone Controller.

http://atcaonline.com/controller.html

___________

The Original Design

For ease of construction the original circuit was very basic. It was designed to animate the payphone’s coin relay and also make the payphone ‘coin first’. The first addition to the original circuit was a ring detector to allow the phone to answer incoming calls without having to deposit a coin.

http://www.atcaonline.com/controller2.html

This article describes a technique for controlling the action of the payphone’s coin mechanism. It is a novel way of using a timer to switch the hopper’s coin vane between the refund and collect position automatically. Although it would be more authentic to detect if a call has actually been answered, any circuit capable of doing this would be beyond the scope of this series of articles and the ability of the average phone collector to build.

___________

Hopper Control

In order to make the coin relay of the payphone function from the ‘on-hook’ telephone line voltage, it was necessary to reduce the physical load on the armature of the relay. All the mechanics used to move the hopper’s coin vane were disconnected. The only function the relay had the ‘strength’ to perform was to reset the coin trigger, disconnecting the payphone from one side of the phone line when it was ‘hung-up’.

Obviously, the relay can no longer be used for coin control. However, we can ‘borrow’ the design concept used by Bell Labs when they designed the single coil coin relay in the 1960’s and Western Electric when the double coil relay was designed over 100 years ago.

http://atcaonline.com/coinrelay.html

Both of these designs used a very basic law of physics;

SIMILAR MAGNETIC FIELDS REPEL, OPPOSITE MAGNETIC FIELDS ATTRACT.

___________

Hopper Modification

Refer to Fig. 1. We will be adding two permanent magnets and an electromagnet to the payphone’s hopper. The permanent magnets are Neodymium Iron Boron disks and are attached using double-sided foam tape. The disks are 3/4 of an inch in diameter and 1/16 of an inch thick.

These magnets are ‘rare earth’ type magnets. They are called ‘rare earth’ because Neodymium is one of the ‘rare earth’ elements on the periodic table. Neodymium magnets are the strongest of the ‘rare earth’ type and are also the strongest permanent magnets that exist.

Fig.2 shows the mounted electromagnet. We will be discussing it in great detail later in this article.

Fig.1

Fig. 2

___________

Hopper Operation - Collect

Refer to Figs. 3 and 3a. Note the position of the two magnetic disks. The disk on the right is mounted to the outside wall of the hopper with the South pole facing left. The disk on the left is mounted on the right hand surface of the coin vane with the South pole facing right. The South poles of both magnets facing one another cause the magnets to repel, forcing the coin vane to pivot to the left. Remember: SIMILAR MAGNETIC FIELDS REPEL. In this position, any coins passing through the hopper will be collected and fall into the coin box. Notice there is no voltage applied to the electromagnet.

Fig. 3

Fig. 3a

___________

Hopper Operation - Refund

Refer to Figs. 4 and 4a. In the refund position, voltage has been applied to the electromagnet. The NORTH pole of the electromagnet is on the right facing the North pole of the magnetic disk mounted on the coin vane. This causes the coin vane to pivot to the right. Remember: SIMILAR MAGNETIC FIELDS REPEL.

Notice that there are two opposing similar magnetic fields; one magnetic field between the electromagnet and the magnetic disk on the coin vane and a second that is between the magnetic disk on the coin vane and the magnetic disk on the right side of the hopper. Since the electromagnet has been designed to produce a field much greater than the magnetic disks, the opposing field on the left (with the electromagnet) will override the opposing field on the right created by the two magnetic discs.

In this position any coins passing through the hopper will be directed to the coin return chute.

Fig. 4

Fig. 4a

___________

A Work In Progress

The electromagnet is the part of this project that I consider to be a work in progress. I’m not a machinist and have very rudimentary equipment. I used the machines I had available - a drill press and a Unimat lathe.

If there is any member of the Antique Telephone Collectors Association (ATCA) that is a machinist or has a background in winding inductors, with access to coil winding equipment and material, and would like to donate his knowledge and expertise to improving this project PLEASE speak up. I can guarantee that you will receive no fame or fortune for your efforts.

___________

Making The Electromagnet

Making the electromagnet is the most complicated part of this project. I salvaged the material from the electromagnet that is used in Western Electric/Northern Electric 3 slots that have 10 cent coin tracks. Refer to Fig. 5.

Fig. 5

For years these assemblies were removed from payphones modified for home use. They were taken out of the phones to make room for 101coils that were mounted on the phones’ back castings. You should be able to locate them from the usual sources of payphone parts. Worst case if you can’t and you’re working on a Western/Northern Electric payphone you can use the one that’s in the phone. You won’t loose much by taking it out (only a little originality) because it’s doing nothing except clicking when you take the phone off hook and adding 100 ohms of resistance. You will have to add a 100 ohm resistor between terminals ‘A’ and ‘E’ of the coin track. This maintains the total resistance of the phone as it appears to the phone line. It is also necessary for the proper operation of the controller and driver boards. Don’t replace it with a jumper.

I’m not going into great detail about making the electromagnet. The pictures that follow do a good job of telling the story. Basically we will be:

1- Unwinding all the wire from the coil onto some sort of spool.

2- Cutting the length of the core in half. The half we’re using is trimmed to exactly 1/2 inch.

3- The cut end is drilled and tapped 8-32.

4- The end with the threaded stud is drilled out to clear an 8-32 screw.

5- The original terminal board and end pieces of the coil form are reused. They are held in position with a very small amount of Five Minute Epoxy.

6- The mounting bracket is made from the moveable armature of the original electromagnet.

7- A threaded 8-32 stud is used to hold the core in the chuck of the lathe to rewind the coil.

That’s it for the details, now for the pictures.

Refer to Fig. 6. The bracket is fabricated from material taken from the original electromagnet’s moveable armature. It’s made so that it positions the electromagnet in the refund opening of the hopper as far to the front and as high up in the opening as possible. Remember that refunded coins must clear the electromagnet in order to exit the hopper and fall into the refund chute. They will travel behind and below the coil. You want to give them as much room as possible to pass by.

You don’t want the bracket and the coil sticking into the top of the return chute to far to the left. Any coins deposited in the wrong slot of the coin gauge will not travel through the hopper but will fall into the return chute from above. This includes the first nickel from abandoned calls when you push the Coin Release button or hang up the handset. You want these coins to clear the left end of the electromagnet and keep on going out of the refund chute without getting caught on the bracket.

It makes no difference in which direction you wind the coil because the polarity of the finished electromagnet depends on how you connect the two wires to the driver board. Cover the metal core with scotch tape before you start winding the coil. There was an insulating tape on the core originally that probably got destroyed when the core was cut down to ½ inch. Cover the completed coil with masking tape to protect the wire.

The completed coil should have a resistance above 30 ohms. Since the resistance depends on the amount of wire you wind and the amount of wire that can get wound on the form depends on how neatly you wind it; this is a variable (the best I’ve done is 34 ohms). Don’t go crazy trying to do a perfect job winding the coil. I’ll explain why when we analyze the electromagnet driver circuit.

The assembly is mounted with a 2-56 screw that is threaded into a hole that has been drilled and tapped in the edge of the hopper. The plate under the head of the screw covers the bracket because the hole in the bracket is much larger than the head of the screw. Adjustment to the electromagnet’s position will be necessary to get maximum performance between its magnetic field and the field of the magnetic disk mounted on the coin vane. This larger hole in the bracket allows for movement of the assembly in all directions to find the optimum position before the screw is tightened.

Fig. 6

___________

Positioning The Magnetic Disks

Refer to Fig.7. The magnetic disks are held in place with double-sided foam tape. The positions of both disks are shown in the pictures. This view is looking up into the coin hopper from the opening in the bottom. This is the collect port that would be sitting flat on the tray in the payphone when the hopper is mounted. Scroll up to Figs. 3 through Figs. 4a for the proper pole direction of the magnetic disks.

Fig. 7

Fig. 8 is a more detailed view of the disk on the right hand wall of the hopper.

Fig. 8

A compass is used to identify the north and south poles of the magnetic disks. I’d write N or S on the disk faces before you start the assembly.

Fig. 9

___________

Positioning The Electromagnet

Positioning the electromagnet is a little tricky. A ‘catch 22’ situation exists. You would think that the right side of the electromagnet should be as close as possible to the magnetic disk mounted on the coin vane. That would make the repelling field created by the electromagnet and the disk as strong as possible and cause the coin vane to pivot to the right (refund) with lower voltage applied to the electromagnet. In reality the opposite occurs. Since the core of the electromagnet is soft iron and the disk is a magnet, there is an ATTRACTION of the magnet to the iron core. We are looking for the opposite to occur when voltage is applied to the electromagnet. That means the voltage applied to the electromagnet must be HIGHER to generate a stronger OPPOSING magnetic field the closer the electromagnet is to the magnetic disk. The electromagnet’s field must first overcome this natural attraction and then go to work over-riding the other opposing the magnetic field created between the two magnetic disks.

With experimentation and a very precise low voltage variable power supply you can find the best mounting position for the electromagnet. The optimum situation is:

1 - The lowest voltage (about 4 volts) that reliably causes the coin vane to pivot to the right

(refund).

2 - The bracket of the electromagnet sticking into the refund chute opening the least amount

possible.

3 - The electromagnet mounted as high up and to the right in the refund opening of the hopper.

The target voltage for reliable operation of the coin vane is 4 volts or less. There is a voltage drop across two transistors Q1 and Q2 that are on the driver board. If you are starting out with a 6 volt battery pack there is very little extra voltage available.

You can determine the proper polarity of the two wires from the electromagnet that cause the hopper to pivot by switching them on the terminals of the power supply. Start with an output of 8 volts from the power supply. This will guarantee enough voltage to get the coin vane to move even if the electromagnet is in the worst possible position. Remember to label the polarity of the two wires. You will have to know this in order to connect them to the driver board properly.

These adjustments are done with the hopper assembly sitting on your bench in front of you, not mounted in the payphone. Refer to Fig. 10. This is the power supply I use to adjust the position of the electromagnet. Notice that it takes a minimum of 4.6 volts to flip the coin vane to the right (refund) with this particular assembly. Refer to Fig. 11. At 4.6 volts the current drawn by the electromagnet is 110 ma. This is a reasonable starting point. With further adjustment to the position of the electromagnet the voltage necessary to operate the coin vane reliably will be decreased.

Fig.10

Fig. 11

The Electromagnet Driver

Fig. 12 is the circuit diagram of the Electromagnet Driver. Fig. 13 shows the parts placement on the pc board. Fig. 14 is a close-up of a completed driver board. Note that C2 has been deleted.

Fig. 12

Fig. 13

Fig. 14

The purpose of the electromagnet driver is to apply voltage to the electromagnet putting the coin vane in the refund position for a pre-set period of time, while the driver circuit senses an input voltage.

The circuit is designed so that the timing is reset to zero when the dial is ‘wound up’ to the finger stop, or a Touch Tone button is pushed. It makes no difference the amount of time needed to dial the entire phone number. The number of digits is unimportant and the speed at which the person dials is not a factor. After the last digit is dialed the timer starts (again) and energizes the electromagnet, putting the coin vane in the refund position for 40 seconds. After 40 seconds the voltage is removed. The coin vane flips to the collect position and for the remainder of the call the circuit idles with a current drain of less than 4 ma, until the phone is hung up.

The circuitry of the driver is fairly straightforward. IC1 is a darlington opto-isolator. Its input ‘samples’ the voltage at a point in the payphone that has 3-5 volts on it when the phone is off-hook, and no voltage when the dial is ‘wound up’ to the finger stop, or a Touch Tone button is pushed.

When the input of IC1 ‘sees’ a voltage above 3 volts its output goes to ground, biasing transistor Q1 ‘on’. This puts B+ from the battery pack on IC2 (a CMOS 555 timer) starting the timing cycle. The timer is set for 40 seconds by the value of parts C3 and R4. While the timer is running, its output is ‘high’, applying a positive voltage to the base of transistor Q2, biasing it ‘on’. Q2-biased ‘on’ grounds the electromagnet energizing it. After 40 seconds the timer turns off. Its output goes to ground. Q2 is turned off and the voltage to the electromagnet is removed and it is de-energized.

The circuit is battery powered so it was designed for minimum current consumption. When the hopper is in the refund position and the electromagnet is energized, the circuit draws about 140 ma. Remember, this is only for 40 seconds. That’s the reason I said don’t go crazy trying to make a perfectly wound coil. Even if the resistance of the coil is a little low, and the current consumption a little high, it’s only for 40 seconds. Forty seconds after the last digit is dialed or the last Touch Tone button is pushed, the circuit idles at under 4 ma. When the phone is hung up, the circuit draws no current.

The battery pack can be any type from 6.0-7.2 volts; there’s plenty of room in back of the coin return chute in the vault compartment for the batteries. I’ve used everything from 4 AA Alkaline batteries in a plastic holder to a 7.2V 1.4 Amp Li-Ion 2 cell rechargeable battery pack. I have no idea how long a fully charged Nimh, NiCd or Li-Ion pack with a capacity of over 1000 mah will last in this application. Probably longer than your next new car will last, before the battery pack needs to be recharged.

If you’re looking for a good source for rechargeable batteries you might give these folks a try. They specialize in every type of battery pack and the prices are excellent.

http://www.onlybatteries.com

___________

Electromagnet Driver Parts List

Part Description Value Part Number

C1, C2* Electrolytic 1 mfd

(deleted) Capacitor

C3 Electrolytic 220 mfd

Capacitor

C4 Capacitor .1 mfd

D1 Diode 1N4001

IC1 Opto-Isolator 4N33

IC2 CMOS Timer LMC555CN

Q1 PNP Transistor ZTX550

Q2 NPN Transistor ZTX450

R1 Resistor, 5.6K, ¼ Watt, 5%

R2, R5 Resistor, 1K, ¼ Watt, 5%

R3 Resistor, 10K, ¼ Watt, 5%

R4 Resistor, 150K, ¼ Watt, 5%

PC Board

2 Neodymium Iron Boron Magnetic Disks

A kit of parts including an etched and drilled pc board is available to members of the ‘Antique Telephone Collectors’ ATCA. This (as with the original controller parts kit) is a service for club members only.

I can’t help you with the electromagnet. You’re on your own finding that.

__________

Testing

Testing the hopper and the driver board should be done to make sure they work before they are installed in the payphone.

Connect the battery pack to the terminals marked Bat + and –. Be sure to observe proper polarity. Connect the wires from the electromagnet to the terminals marked Mag + and -. Once again observe the proper polarity. Set the output voltage of the variable power supply to 5 volts then turn the power supply off. Connect ‘L’ to the positive output terminal of the power supply, and ‘Y’ to the negative output terminal. Turn the power supply on. The coin vane should flip to the right (refund) and remain in that position for 40 seconds. After 40 seconds it should automatically flip back to the left (collect). Turn the power supply off and then on again. The operation of the coin vane should be repeated. Assuming all this ‘good stuff’ has happened, go ahead and mount everything in the payphone.

Connect the ‘L’ and ‘Y’ leads from the driver pc board using the list below. Connect the magnet to the driver board observing proper polarity. Make a cable that matches whatever battery pack you decide to use. I recommend 4 AA Alkaline cells. It’s a simple, inexpensive choice and you won’t have to worry about a battery charger. Radio Shack stocks a cable with battery snaps (Catalog # 270-326) and a mating 4 cell plastic battery holder (Catalog #270-383). By the time the batteries have to be replaced you’ll probably forget there are batteries in the payphone.

Scroll down to the last part of this article and mount the board (boards) using the pictures as a guide. Then, close up the phone, plug it in the line and make a call. Each time you ‘wind up’ the dial to the finger stop or push a Touch Tone button the coin vane in the hopper should flip between refund and collect. After the last digit, the coin vane should be in the refund position and stay there for 40 seconds. This is more than enough time for 5 rings and one ‘hello’. After 40 seconds you will hear the coin vane flip to the collect position. If you hang up before 40 seconds, your initial deposit will be returned. If you hang up after 40 seconds your deposit will be collected.

The action of the Electromagnet Driver circuit is dependant on the control voltage that appears on ‘L’ and ‘Y’ of the board. This voltage is a function of the characteristics of the switching equipment your payphone is connected to. If you are reverted to a dial tone when the party you called hangs up, or the voltage on the line drops for a split second, this will reset the driver circuit to the refund position. In a situation like this if you want your payphone to simulate a completed call properly, YOU must hang up before the person you called does. I’ve noticed on some of my phone lines that this characteristic depends on where I’m calling. Don’t ask me why. I have no idea and could care less.

VOIP? It might work or it might not. There are so many suppliers and different ATAs that I have no way of knowing what will work and what won’t. From my experience the ATA wants to see either absolute ‘on-hook’ or absolute ‘off-hook’ voltage, one or the other, with no in-between voltage steps. The Mini Controller and the Electromagnet Driver circuits draw current from the phone line that cause different voltage steps to occur after the phone is hung-up but before the line goes to complete ‘on hook’. So for VOIP I’m just going to say FORGET ABOUT IT! A POTS on them.

___________

Connecting To The Payphone

Below is a table of connection points for the input of the driver board terminals ‘L’ and “Y’ for various model phones. I will be adding to this list as I install these boards in additional models.

For Western Electric 190 and 200 series prepay 3 slots.

Northern Electric prepay 3 slots are the same.

Connect terminal ‘L’ of the pc board to ‘R’ in the payphone

Connect terminal ‘Y’ of the pc board to ‘BKX’ in the payphone.

For Western Electric 1234G Touch Tone 3 slots.

Connect terminal ‘L’ of the pc board to ‘BB’ in the payphone.

Connect terminal ‘Y’ of the pc board to ‘TR’ in the payphone.

For Automatic Electric LPB series 3 slots.

Connect terminal ‘L’ of the pc board to terminal ‘14’ of the induction coil.

Connect terminal ‘Y’ of the pc board to terminal ‘1’ of the induction coil.

__________

Installation

This is a collection of pictures that can be used as a guide to installing the three boards from this series of articles in various model payphones.

The Northern Electric NE 233QF and Western Electric 1234G also have ITT 1427 networks installed.

http://atcaonline.com/WE200payphones.html

Automatic Electric LPB

Northern Electric NE233QF

Western Electric 1234G

___________

The End

I can’t think of any other function I can add to further complicate what started out to be a very simple construction project.

If you find any errors in this or the previous two articles please let me know. If there is any point that is unclear please let me know. If you built any of the boards and can’t get them working please let me know. It you can IMPROVE upon anything I’ve done PLEASE definitely let me know. If you have any complaints please just go away!

___________

Hey, This Stuff Is Interesting

If you were brought here by a Search Engine and you have an interest in, or would like further information about collecting, modifying, repairing or restoring antique phones, I suggest you go to:

http://atcaonline.com/index.html