sundance-communications.com/forum Forums Other TDM Phone Systems 1A2 and Comkey Telephones & Systems Common Audible Signaling with Mixed 6/10/20 Line Sets

Now that I have a 584C panel and a 515 KSU to play with (more on the 515 later), I've run into an issue and figured I should check with the sagely 1A2 wizards that reside here. I looked at BSPs 518-010-108 and 518-215-100, but I'm still scratching my head a bit.

I'm going to have a mix of 6, 10, and 20-button sets on my KSU with Common Audible Signaling for the CO/PBX lines. (I'm only using 400 KTUs.) At minimum there will be 13 CO/PBX lines.

This is what I had in mind:

- Lines 1-5 ring everything
- Lines 6-9 ring on the 10 and 20 button sets
- Lines 10 and above only ring on the 20-button sets

I can't see a way to do this without a diode matrix. Without a diode bridge it looks like there would be a sneak path from the higher-order lines to all ringers, due to the common ring path. Am I missing something or is a diode matrix really the only way to do this? I have to imagine having mixed sets was a common occurrence.

Yes, you will need to use a matrix, but a fairly simple one since you'll only need three outputs from it; one for all of the six button sets, one for the ten button sets and one for the twenty button sets.

You'll need one diode to serve just lines 1-5 on the six buttons;

You'll need two diodes to serve lines 1-5 and 6-9 on the ten buttons;

You'll need three diodes to serve all three groups of lines for the twenty buttons.

Don't forget to bypass the capacitor in the sets by either shorting terminals A and K together or by moving both wires on these terminals to either end on the same terminal.

Looking into BSP 518-010-108 again, I found the reference to BSP 518-310-401 which talks about the 227B KTU. Rather than modifying the stations, I think I can build something in the spirit of a 227B KTU.

To elaborate on my previous post, this is what I'll probably end up doing:

1) Swap the audible signaling supply on the KSU/584C panel side to be B Battery.

2) Wire three groups of common audible signaling; one for lines 1-5, one for lines 6-9, and one for lines 10+.

3) Connect each of those common audible signaling lines to a separate relay. In this case, I've got a bunch of 24VDC coil industrial relays floating around (they're Potter Brumfield KUP-type relays, for the curious), along with some 1N4001 diodes to snub the turn-off voltage spike.

4) Each common audible group is connected to an NO (normally open) contact on the assigned relay, and 105VAC is connected to the common terminal.

Alternatively, if I could get a hold of a few more dead 400-type KTUs, I can do some fancy microcontroller tricks and make a super KTU. I'll probably save that for another day, but if anyone has a box of dead KTUs (or interrupters) let me know.

Relay logic is primitive, expensive, and inflexible compared to the correct way: diode logic. Why limit yourself to just the configuration that you now want? Using a commercially-available or a home-made diode matrix block is the way to go, since you are a hobbyist and you (one hopes) want to emulate what a real system once would have looked like.

Take a 66M50 block (the older version, not the newer so-called Cat5 version). Pry the back cover off. Shake out and save all the pins. Take a 66B4-25 or 50, and take the back off and discard all the pins.

Insert the pins from the "M" block in Columns A&B, and in E&F of the "B" block. Now you have a diode matrix block capable of providing spaces for 50 diodes. The diodes go across the area where the Column C&D pins use to be. The CA leads from each line get punched down on Column A, a diode is installed from each lead, and on Column F you strap all the diodes together from the desired lines to create a ringing lead for one telephone.

input lead from line card > pin pin ---< diode >--- pin pin > ringing lead to set

Repeat as necessary for all the lines and sets. Just to emphasize what the other dinosaurs have told you...take the ringing condensor out of the circuit in each phone. Be sure to aim the diodes in the correct direction. (Cathode towards the sets.)

All valid points, Arthur. There's a few reasons I'm favoring relay logic over diode logic:

(1) I've got everything on-hand already
(2) Using the relay approach means not having to bypass the set capacitors
(3) I can add in a microcontroller down the road if I want more flexibility
(4) Personal design preference

Nick:

While your relay idea would work in theory, there's still an opportunity for a back feed situation if two of these slave relays operate at the same time.

Now if you're willing to get creative with some normally closed contacts on the relays, you might be able to circumvent this.

I actually gave this some thought regarding using the ringers for the dial intercom. Wiring the signalling with B battery to operate a relay for each station. The station would come off the common pin on the relay, common audible on the normally closed pin and ring generator on the normally open pins. This way when the relays were not operated, the sets would ring normally. A signal from the dial icom would operate the one relay and momentarily apply ring generator to the one station.

Chris: Sounds good in theory.

Ed: Had to think about a scenario like that for a second, but you're right, I can see a situation in which two relays could operate and cause an issue. The fix, using normally-closed contacts (like you suggested), would be to make the lowest-order (in my case lines 1-5) relay lock-out all higher-order relays.

I'll draw up some schematics and post those.