First, my previous post had a mistake. Input is low(zero) on high frequency and high(one) on low frequency. I forgot to account for the inverter that does the final TTL normalization.

The other good news is that I think I’ve been able to get a partial read of one of Mark Arnold’s recordings. It’s not surprising to me that I can’t get an entire read of a copy of a 40 year old recording. This is a move in the right direction, because it gives me a good look at how the read card recovers a proper SCELBI recording. In fact, the TTL normalized signal looks different than what I get from a recording coming out of my system. I need to look closer at my write card operation.