Cortelco network simplication of WE?

[dsk]

I have a 554 and the network has some different components. No varistors, but 2 and 2 (zener) diodes in series. No resistor in series with the transmitter...

Anybody out there who knows more? diagram?

I can not see any number on the network, but the core of the transformer is stamped 88710


dsk

[unbeldi]

Well, perhaps you should show what you have.
Did you take the network apart ?
How did you identify the zener diodes ?   If there is anything printed on them, then it would be easy to find their vintage and origin. Probably not from WECo.

The 425F had no transmission circuit.
The 425J used resistors for equalization instead of varistors.

[dsk]

I am not able to read the small text, but here are some pictures:

dsk

[unbeldi]

The four diodes D1,...,D4 are most likely polarity guard for a TT dial.
It should be relatively easy to draw a complete circuit diagram of the transmission unit.

[Alex G. Bell]

The four diodes D1,...,D4 are most likely polarity guard for a TT dial.
It should be relatively easy to draw a complete circuit diagram of the transmission unit.

I agree that a complete circuit diagram would be much more revealing but looking at the photo of the solder side of the board, it appears that the diodes are connected as back-to-back pairs rather than in a bridge config.

[Alex G. Bell]

The four diodes D1,...,D4 are most likely polarity guard for a TT dial.
It should be relatively easy to draw a complete circuit diagram of the transmission unit.

I agree that a complete circuit diagram would be much more revealing but looking at the first photo of the solder side of the board, it appears that the diodes are connected as back-to-back pairs (D1 & D2, and D3 & D4 as IDed in the last photo) rather than in a bridge config.

Years before the ITT to Cortelco transition, ITT introduced an L-C mechanical keypad dial with built in polarity guard.  The last ones I know of: 42 and 46 types, digital with conductive elastomer keypads, have built-in PG too.  It seems unlikely they would move this function back to the network. 

Perhaps they transitioned to use of amplified electret transmitters that don't need the same transmission regulation carbon transmitters did.  It might be useful to know more about the handset used with this network, especially the transmitter.

[dsk]

I did ask becauseI had not seen or heard about this before. Ihave measured a lot, and tried to make a diagram, and this is what I believe is the right. Terminals with no connection to components are not "mentioned"

dsk

[dsk]

The red ovals are then the varistor replecements:

[Alex G. Bell]

I did ask becauseI had not seen or heard about this before. Ihave measured a lot, and tried to make a diagram, and this is what I believe is the right. Terminals with no connection to components are not "mentioned"

dsk

Good work!  The silicon carbide varistors normally used in 425B networks and their descendants have particular dynamic resistance and dynamic impedance characteristics which I'd expect to be quite different from those of typical zener diodes, which have a very flat voltage characteristic and low effective impedance above the zener voltage. 

Consequently it seems worthwhile to try to characterize these zener diodes at least with respect to DC characteristics if you can do that.  You do not need to remove them from the network to do so.  You need to apply a variable DC voltage through a high enough resistance to avoid damaging them (no more than 100mA should flow through them), measuring and recording the voltage across the back-back diode pair or individual diode as the current is increased.

The characteristics of the 312 type SiC varistors is well documented in an early BSTJ article about the 425B net. published in the early 1950s.

BTW, ISTR that there were some simplifications of the network design in late type WECo Trimline sets, modification of the sidetone balancing ntwk.  Perhaps some Cortelco Trimline clones also have a circuit similar to what you found in this full size ntwk.

[poplar1]

Western Electric also produced a simplified network for 500s, 2500s, etc. in the early 80s. It was coded #4293.

[Alex G. Bell]

Western Electric also produced a simplified network for 500s, 2500s, etc. in the early 80s. It was coded #4293.

Here is what Ralph Meyer said about its circuit:
"Several shortcuts were taken in the No. 4293 network.  First, a much smaller and lighter weight coil was used.  Second, the condenser in the sidetone-balancing impedance was eliminated.  Third, the DC blocking condenser in the receiver circuit was reduced to 1.5 microfarads. Fourth, the 22-ohm resistor usually used in series with the transmitter was omitted.  And resistances of the remaining two resistors were also changed again.  These changes resulted in a sacrifice in performance, but they reduced cost and weight." ... "This circuit is identical to that used previously in the early Trimline telephones, where shortcuts were necessary to get everything inside the handset."

I don't think I have a 4293 that is not riveted down to a telephone set base but I infer from Ralph's description that the 4293 continued to use the SiC varistors rather than glass silicon devices.

[poplar1]

I don't think I have a 4293 that is not riveted down to a telephone set base but I infer from Ralph's description that the 4293 continued to use the SiC varistors rather than glass silicon devices.

The 4293 networks are not riveted down. Rather, they are attached with supervisor's tape (sticky on both sides).

[Alex G. Bell]

The 4293 networks are not riveted down. Rather, they are attached with supervisor's tape (sticky on both sides).

Do you have a loose one, or a phone that you're sure contains one?  I don't know whether I do and don't have the time to start opening phones to find one.

[poplar1]

If you have a Princess tel set with no lamp socket, or any WE set marked "1PTY", it will most likely have a 4293 network.

[dsk]

Good work!  The silicon carbide varistors normally used in 425B networks and their descendants have particular dynamic resistance and dynamic impedance characteristics which I'd expect to be quite different from those of typical zener diodes, which have a very flat voltage characteristic and low effective impedance above the zener voltage. 

Consequently it seems worthwhile to try to characterize these zener diodes at least with respect to DC characteristics if you can do that.  You do not need to remove them from the network to do so.  You need to apply a variable DC voltage through a high enough resistance to avoid damaging them (no more than 100mA should flow through them), measuring and recording the voltage across the back-back diode pair or individual diode as the current is increased.

The characteristics of the 312 type SiC varistors is well documented in an early BSTJ article about the 425B net. published in the early 1950s.

BTW, ISTR that there were some simplifications of the network design in late type WECo Trimline sets, modification of the sidetone balancing ntwk.  Perhaps some Cortelco Trimline clones also have a circuit similar to what you found in this full size ntwk.

I,m sorry, I do not have a suitable powersupply.  The way tha voltage of the diodes was determined was just by using a laptop power in series with a 1k Resistor, and measuring the voltage across the diode. 

As you see this network is even simpler than the 4293 someone mentioned.  The picture show a PCB with room for several other components.

dsk