Multi-Line Telephone Intercom 

Obviously if you want a second talk circuit, you'd need numerous more relays. It becomes an entirely different project. This project is basically "selective ringing."

Good luck with your project! If your phones are *really* ancient, with electro-mechanical ringers, this power supply may not provide adequate ringing voltage.

I am! Thanks, it's nice to be here.

This is not a bad explanation that gets you most of the way for an PBX. Add an arduino such that if a party line is in use, then other calls can't be made. There are some chips that is made for the matrix switching, such as 8 rows by 8cols or even 16 lines. This video and schematic has got me thinking!

Nice! How did you manage the part where only 2 ppl could speak to each other? Can't wrap my head around it.

@@BurkenProductions Thanks. I'm not sure I understand exactly what you're asking, but let me summarize how my system worked. Each phone had its own series "line relay." Initially all the line relays received voltage from the supply. When one phone went off-hook, that removed voltage from all the other line relays; so other phones could not initiate any actions. Thus only one conversation was possible at any given time. (Given that we had only 7 or 8 extensions, and only 4 people were normally involved, it was unlikely that we'd have a need for simultaneous conversations.) The "active" phone also controlled "B" and "C" relays, which in turn controlled the stepper switch. When the phone dialed some other number, the stepper selected the "called" phone. If the called phone was (presumably) on-hook, ringing voltage was sent (via stepper switch contacts). If/ when the called phone went off-hook, ringing voltage was disconnected and called phone was connected to calling phone. That allowed the conversation. I honestly don't remember whether the two phones had a DC connection path, or whether there was a series "talk" capacitor connecting the called phone back to the calling phone. In the former case, the system would remain "tied up" until both parties hung up; in the latter case the system would be controlled by the calling party. In retrospect some other features would have been nice ... maybe some sort of audio tone injected into the talk circuit if a *third* phone went off-hook and wanted to use the system. Then when the first two parties heard that tone, they could hang up, freeing the system for the third phone. Etc. But this was 1965, when there no such thing as consumer-level logic chips, and I was spending my allowance to buy surplus relays. If that didn't answer all your questions, let me know and I'll try to recall more details. Cheers!

IMHO that's a much better solution than the one presented here. Better waveform, fewer HF harmonics so less problems with transformer impedance, etc. The challenge is finding a transformer that can pass 20 Hz with reasonable efficiency.

Yes, you need a coil in serial with the power, otherwise you will just kill the audio with the power supply, but here they use a resistor, will work but not that well as a coil.

Agreed 100%. Think about a TELCO central office subscriber line circuit. A big mother 48v battery (0 ohms AC impedance) in series with two relay coils which produce a *high* impedance to the audio.

It's not clear to me exactly what you want to do. However, the TELCO phone line provides a maximum of about 50-75 mA, not nearly enough power to run the ring generator and other electronics in this system.

The system doesn't recognize when the phone receiving the call has picked up, but only rings for as long as the caller is pressing his keypad button, so a quick press should be enough for the called phone to realize someone is calling them, as long as someone is in the room that is.

Just use a HT502 ATA?

@@tieline1333 An open hardware solution would be nicer. Also, for a lot of individuals, they'll still need the Raspberry Pi or whatever to run Asterisk on.

You don't need an Arduino. Just use a BCD-to-decimal decoder, e.g. 7445, to drive the relays. Interface to landline using a relay and 600:600 ohm audio transformer, and you'll need some logic to keep that landline circuit latched on until the calling local phone goes on hook.

Obviously no such provision, since all the phones are normally connected together on the "party line."

It wont.

Yes it WILL work but you'd need a pulse to tone converter as well.

Let's start with the design parameters for a POTS telephone system. About 48v DC (very well filtered, but regulation not too critical) when all the phones are on hook. Then you need enough series resistance so the voltage *at the phone* drops to around 10-15 volts when the phone is off hook (even a bit lower voltage is OK). This will make the tone generator and talk network work correctly. The series resistance is important because it isolates the telephone audio from the zero ohm AC impedance of the power supply. In a real telephone exchange, this resistance comes from the windings of the coils on the subscriber's line relay. The coils have inductance, which provides even more isolation than a pure resistance would. Many local office systems (PBX = Private Branch Exchange) will start with 24v DC, rather than 48v. No problem, you just need less resistance in the relay coils, to achieve proper working voltage at the off-hook telephone. My point is that you do not need a constant current supply. You can achieve good operation with a random DC supply (must be well filtered!) and series resistance, or better yet series inductance. You do want to end up with around 10-15 volts at the phone *when* the phone is off hook. The constant current supply is OK. When one user is talking and that one phone is instantaneously drawing less current (negative audio peak), the current will be forced into the other phone, and vice versa ... that's how the audio signal is transferred between phones. Personally, I prefer filtered DC *voltage* supply with series resistance or inductance ... that's just me, what's what the phones were designed for.

I've thought about the same thing. In fact the Arduino could also accept input from the DTMF decoder IC, so it would work with either type of phone dialing.

Ah, good question that I didn't cover.. phones are equipped to handle a horrific barrage of voltages and transients that could appear on the line, including ring voltage when off hook. It just makes a clicking noise on the receiving end. If I had used an off-hook detection circuit and maybe a microcontroller, then I could prevent this from happening. But, I wanted to keep it relatively simple so that anyone could build it. Thanks for the excellent question.

@@AmRadPodcast Still a very good coverage and nice way of explaining them. Hope to see more of your videos here!

Can you send that to me as I am very interested in this

I am also interested in this project

@@sam462m github.com/WRFleete/PABX-project has the code + schematics + gerbers, havnt tested with the RJ45 connectors on that schematic etc but I did have individual ports for each phone. Code may be a bit of a mess and I may comment on it a bit. The code has basic support for pulse dialing as well and the ability to connect to a POTS/PSTN phone line (and dialing to get the "outside line" and "hold" feature)

@@rossherbert1591 posted a link in another reply to the git repo

DTMF is actually analogue audio.

Just add a stepping relay, plus "A," "B" and "C" relays to control it.

Talk to the electronics shop teacher at your local high school. He will set you up with one of his students who would love to build this and would appreciate a little cash in return.

@@gregm.857 I don't think we have hardly any shop program at our local HS anymore. A lot of it was shut down in the 90's after I graduated and went to college.

@@looselycollected7505 Where I live, most of the shop programs have been moved out of the local high schools and into centralized vo-tech schools. Maybe your local HS administrative office can direct you to a nearby vo-tech school. Alternatively, try the local amateur radio community. Admittedly, times have changed, and now a lot of "hams" do NOT know how to solder and build their own gear; they just buy a radio online and start talking. But there will be a few "old timers" who are proficient at hardware construction and might be eager to help you with this project.

There is nothing the least bit complicated about this project. It's the heart of simplicity. In fact you don't even need the constant current supply (TELCO systems don't have one in the speech circuit) ... it could be replaced by a filtered supply and a series resistor.

You miss the entire point. The point is to find a "fun" use for some vintage wired telephone sets.

@@gregm.857 You have your own agenda as you clearly didn't read what I posted...

@@Fifury161 I did read what you posted. You said this project was "too late" because now you can use a lot of modern apps (instead of telephones). My reply was that this project is not meant to be the "latest and greatest" but rather it is to intentionally use some thing that is vintage. I don't know what agenda that indicates.

That's why he connected a multivibrator to a transformer.

Not so sure about that. The ‘text headers ’ may be one direction, but I’m speaking about the buttons. Columns of buttons going vertically, rows of buttons horizontally.

Caleb Bennetts yes!!!

What is that ASMR ?

True this is definitely an option, and maybe a subject for a future video, but in this case I wanted to focus on old analog phones. I do have asterisk running on RPi which I use for daily communication - and to be honest, it's a bit sluggish.

@@AmRadPodcast you can add old analog phones to asterisk by adding a gateway.

@@AmRadPodcast npstn.us -- Uses Asterisk Systems linked together to call into vintage telephones and switching equipment, You could even connect your circuit you built into asterisk using a Gateway, Then configure Asterisk to dial onto the circuit, dial a number and connect a call. Asterisk can do Everything Analog & everything digital, It's awesome!

You can't have amps without volts.

@@gregm.857 my point was that a 10000 V charge with 1 amp will not really hurt you but a 12V charge at 100Amps will fry your - toast.

@@larrypriest5789 I'm not sure I understand why you made the comment in the first place, and I surely do NOT agree with your conclusions. 10,000 volts will overcome your skin resistance and will go through your body. It takes only 100mA (0.10 amps) to kill a person. So 10,000V and 1 amp certainly WILL hurt you! The usual cause of death is ventricular fibrillation where the current causes the heart to stop beating normally. It also usually stops respiration so asphixiation is a possibility, and at 1 amp there will be a lot of burned tissue. On the other hand, 12V is not enough to overcome normal resistance of dry skin, although if your skin is wet with perspiration (basically salt water) or you're standing in a puddle of water then 12V will go through you. And 100 Amps literally might boil some of your internal organs due to resistive heating. If you want a simple experiment, use your car's electrical system. It is all negative-ground. So first put your hand on the positive terminal of your 12V car battery ... you won't feel a thing. But if you pull back the insulating cap on one of your spark plugs and put your hand on the connector inside (around 12,000 V or more) you will get a hell of a shock. And BTW I do *NOT* suggest doing this because it *MIGHT* be lethal even if you are not grounded. The situation that you said "will not really hurt you" has 10,000 V times 1 A = 10,000 watts of power. The situation that you said is worse has 12V times 100 amps or only 1,200 watts. So your "safe" situation has eight times more power and can potentially cause eight times more damage. I definitely do NOT agree with your conclusion.