The MN6042 is used in the MK2 versions of the SL-1300/1400/1500/150, the SP-15, SP-10 MK3, and SP-02.
Lewm: not millions by any stretch. The nominal frequency through the pitch chip is 262.080KHz. This is also where I'm seeing ~8Hz variation across the three samples I measured.
The output of the pitch chip is delivered to the DN860, which divides by 1/8, then 1/54 for 33RPM, and then 1/12 to the AN660 chip. The "quartz locked" frequency for 33RPM is 50.5555Hz, which is the frequency the AN660 uses to develop the reference and error control lines to the motor drive.
For perspective, the AN660 sees 50.5555Hz for 33RPM at 0.0% pitch, and 50.6061Hz for 33RPM +0.1% pitch from 262.080KHz. 262.074KHz is 50.5543Hz to the AN660, and 262.081KHz is 50.5557Hz at 0.0%.
Theoretically instability in the reference frequency should manifest as a voltage instability in the reference line to the motor drive. We'd then need to determine how the motor control would react to the degree of variation induced. My theory is this would work in concert with the offset voltage adjustment; tighter control of the reference voltage would allow the drive to be more critically damped.
When I get the next revision of boards in I'll put the motor on the bench and measure through the entire system. I haven't managed to find a test record anywhere close to the quality needed to reliably measure flutter below 0.02% JIS, so my conclusion on whether the increased stability of my chip replacement equates to anything of note at the platter may be based upon a bit of conjecture.
Worst case, it's designed to be a modern, resilient drop-in replacement for any circuit that uses an MN6042. Best case, we've made things a bit better in a meaningful way.
I do plan on making it available, as I'd love to have a shot at recouping some of the R&D costs. It's amazing how fast things add up. As the largest component on the board is a 4x4mm 20 pin package with no leads, I wouldn't offer it as a kit. Please PM for info.
JP