What kind of diode is this?

[MaximRecoil]

It doesn't seem to have any writing on it. The PCB it belongs to has 7 diodes in total, and all but this one have writing that identifies them. It has:

Four 1N4004
One 1N4742
One 1N4148

And one mystery diode.

I've attached a picture of the PCB (the mystery diode is in location CR7), which is a DTMF tone generator from a 2500-type telephone. It uses a TCM5087 chip along with a 3.579545 MHz crystal (which is the NTSC TV color-burst frequency) to generate the tones. It isn't broken but I want to reproduce the circuit for use in an old broken payphone that I converted to a regular home phone, so I need to know what all the components are. I've been able to identify everything except for that one diode.

Also, if the diode can't be identified, how important is it to use the exact same type of diode? Is there any reason I couldn't substitute one of the other types that are on the PCB for it? It looks similar to the 1N4148.

[MarkT]

Also, if the diode can't be identified, how important is it to use the exact same type of diode? Is there any reason I couldn't substitute one of the other types that are on the PCB for it? It looks similar to the 1N4148.

It could easily be a zener, using a 1N4148 would then over-voltage part of the circuit perhaps - best to identify - can the circuit be powered up and voltages measured?

[MaximRecoil]

It could easily be a zener, using a 1N4148 would then over-voltage part of the circuit perhaps - best to identify - can the circuit be powered up and voltages measured?

Yes, where should I test the voltage at? I'm guessing that it's only powered when pressing a key / generating a tone.

[wasedadoc]

See if you can find an Application Note for the IC.  If it is specifically intended as a DTMF generator in a phone the AN is likely to show a typical circuit which has probably been used with little modification. 

[Kim Christensen]

If the diode is connected across pins 1 & 6 of the TCM5087, chip then it's probably a zener... Though, I guess you've already ID'd that one (1N4742). It might be a varistor which limits the max volume in the ear piece. Best to trace out what it's connected to, and then look at an application note like this one:

[MaximRecoil]

If the diode is connected across pins 1 & 6 of the TCM5087, chip then it's probably a zener... Though, I guess you've already ID'd that one (1N4742). It might be a varistor which limits the max volume in the ear piece. Best to trace out what it's connected to, and then look at an application note like this one:

I've attached an image showing what it's connected to: pin #6 (VSS) of the TCM5087, as well as wire #3 (white) which connects to the GN terminal, and wire #4 (red/green), which connects to the R terminal, both of which are located on the phone's network (Western Electric 425E).

The volume of the tones in the receiver seems to have something to do with wire #1 (orange/black), which connects to the C terminal on the network. If you don't connect that wire to the C terminal, the tones are too loud, and the handset's transmitter (microphone) doesn't work.

[Kim Christensen]

One thing you could try since you've taken it out of circuit is to test it.
A curve tracer would tell you a lot about it. (Have to be careful you don't ruin the "diode" by keep the voltage and current low)

Another option is to take a 1K resistor and a small variable DC power supply. Put the resistor in series with the "diode" to limit the current and connect them to the power supply. Then slowly ramp up the voltage from zero to 10V while measuring the voltage drop across the "diode" and then the resistor. Voltage drop across the resistor will give you the current through the diode since they are in series. (Each 1V = 1mA @ 1K )
Then plot the IV curve of the "diode". Do this with a reverse polarity from the power supply as well. See if it looks like a varistor curve.

[MaximRecoil]

I don't have a curve tracer nor a variable DC power supply.

I know that the network already has a varistor(s), starting with the 425B network that was introduced in the early/mid 1950s (earlier model 500 phones with the 425A network used a separate filament-based "equalizer" to accomplish the same thing, i.e., make the volume the same regardless of your distance from the central office). The receiver element in the handset also has a varistor (originally the 44A in a metal can package with the U1 receiver, and later, a small plastic 104A with axial leads in the U3 receiver) to reduce the volume of any clicks/pops in the receiver.

I have another tone generator PCB in another phone which also uses a TCM5087 chip but has a different circuit design (picture attached below). It only has 6 diodes and only 1 of them looks similar to the mystery diode (small glass package). It has places on the PCB for 2 more diodes (CR7 and CR8), but they were left unpopulated:

Four 1N4004
One 1N5348

And unfortunately, its diode that looks similar to the mystery diode may also be a mystery diode. It does have writing on it, arranged like this: 

SC
83
OF

But I can't find any relevant search results for SC83. The closest I can find is "HSS83" - https://www.amazon.com/FFIME-Original-high-Voltage-Switching-HSS83TD/dp/B0CKB8QNSG

[MaximRecoil]

What are the possibilities? I don't mind trying different ones to find one that works.

Someone mentioned that it could be a Zener diode. Is there any other type of diode it could be?

I wonder why it has 2 extra diodes compared to the example circuit in the TCM5087 datasheet that Kim Christensen attached in reply #4. I checked the datasheets of 2 other compatible tone generator chips (S2559 from AMI and UM9559 from UMC); the AMI datasheet didn't include an example circuit, and the UMC datasheet (attached below) shows an example circuit that's similar to the Texas Instruments one, except it suggests a 1N4742 for the Zener diode instead of a 1N4743.

I'm guessing that the 1N4742 at location CR5 on my PCB is serving the function of the 1N4742 / 1N4743 in those datasheets, so I don't know what the purpose of the 1N4148 at CR6 and the mystery diode at CR7 is.

The image I attached in reply #3 shows the complete circuitry (it's a single-layer PCB), but I don't know anywhere near enough about designing circuits to be able to look at it and say, "Such and such a diode would make sense / work for location CR7."