Do you agree that since the description of pin 3 is "SPDIF in" that the signal is no different than if it was directly from a SPDIF cable rather than via a Toslink connector which then translates the pulses of light to a TTL voltage and presents it as SPDIF to the PCB?
No, I donĀ“t. You are mixing up different levels of the OSI Layer model. S/PDIF describes the data structure of serial audio data. Coax and TOSLINK are two different ways to physically transmit it. Closely related is the AES/EBU way to transmit serial audio data - the physical interface is slightly different to the consumer coax connection, but the data representation is close, but not equal, as many tapers found out to their disappointment.
The coax S/PDIF output signal is an AC signal. The LSI chip expects a voltage <0.8V as low and >2V as high (if it is TTL-compatible). This is a minimum required voltage difference of 1.2V. A conformant S/PDIF coax output is required to deliver 0.4V peak-to-peak as a minimum. See the problem?
Again:
http://www.epanorama.net/documents/audio/spdif.htmlThe TOSLINK receiver and transmitter in my PCM-D50 connect directly to the audio LSI chip - I expect a similar solution in the D100. Wiring a coax S/PDIF input without ESD protection and pulse forming straight to the LSI chip is a sure way to disaster.
But - its not my recorder. Feel free to experiment. Report back if it works - or not.
Greetings,
Rainer