Unfortunately this clock does not have the 3rd wire, just an ac power plug. But if I interpret the back label properly (see photo), adjusting is controlled by injecting 3510 Hz on the ac line.

 

Was considering using the transformer from a power adapter as the interface to the power line, and injecting the isolated audio into the low voltage side?

 

I have an IBM Type:S3 56-12 School Clock
Works on 110v ac (Regular ac plug - no third wire)
Uses 3510 Hz to set time
Would like to build circuit to inject audio on ac to set time.

Hey Charger,

Did you ever get a signal injector circuit to work? I can't find any protocol information on this. I don't know what voltage the pulse carrier needs to be, nor whether it needs to be injected at the zero crossing of the line voltage. What have you found?

 

Hey Charger,

Did you ever get a signal injector circuit to work? I can't find any protocol information on this. I don't know what voltage the pulse carrier needs to be, nor whether it needs to be injected at the zero crossing of the line voltage. What have you found?

Hey Charger,

Did you ever get a signal injector circuit to work? I can't find any protocol information on this. I don't know what voltage the pulse carrier needs to be, nor whether it needs to be injected at the zero crossing of the line voltage. What have you found?

Thanks for you interest and help. I used a 6v power adapter, a generator @3150hz, and an isolation diode. The audio energized the relay which advanced the time to the next hour.

 

Here is how I got mine to work.
MCU keeps time with 32.768 Khz crystal, every 60 seconds, the MCU charges up a cap to about 26 V then dumps the energy into the clock coil.
The voltage divider that samples the cap voltage is switched- so it doesn't waste precious energy until the MCU needs to read the voltage.

The whole thing runs about a year on good alkaline D cells.