Your voltage divider R14-R2 reduces the loop gain and hence the bandwidth of the DC feedback loop and the extent to which it shifts the RIAA correction poles. As long as it doesn't take too long to settle at power on and as long as it can handle the offset and any deep subsonic input signals, it should work fine. Deep subsonic signals get attenuated by the fact that magnetic cartridges respond to velocity rather than excursion, so I guess that's a non-issue.
I guess that's a tad slow, isn't it? I figured, I could get the transient simulation in LTSpice to better show the behavior if put in a voltage source that starts by going from .9mV to 1mV in 100 ms and then stay at 1 mV that shows that you were quite spot on:
I think it should settle after 1/2 cycle of the time constant which is 2pi in this case so around pi-seconds + 100ms for the initial raise
cutting down that RC (R6 * C3) constant to 0.62 sec (1.6 Hz) makes it settle in a little over 300ms and still leaves the RIAA at its original accuracy down to 20 Hz. So I think that would be a good compromise...
Hi,
rather forget the ´overall´dc-servo.
Just servo the INA.
see for example Calvin-Platina-RP1
jauu
Calvin
rather forget the ´overall´dc-servo.
Just servo the INA.
see for example Calvin-Platina-RP1
jauu
Calvin
Oops, that looks very similar to what this ended up with 'cept for where the servo is plugged in ;-)
In fact, I'm very glad that it does because I was able, with help and advice from you guys, to figure it out (mostly) on my own.
