Barry worked hard to create well designed circuits that added little to no noise and did not add any color of their own. Cadac studio consoles were one of his many projects. He died before digital had become practical. He was an analog giant. I for one would like to learn more about him and his work.
Assuming that the PCB that I receive this week works, I would have no problem making the gerbers or even fabbed PCBs available to anyone that wants them.
Huh.
Well, here's hoping I don't jinx myself: Yr 'Umble Narrtr is working on a Cello-like of my own, with a twist: variable center frequencies plus or minus an octave from spec on a seven position switch. For example, the third band will contain 600 Hz as per the original spec, but also 300, 400, 500, 800, 1000, and 1200 Hz, whereas the 2000 Hz band would have 1000, 1200, 1600, 2000, 2400, 3200, and 4000 Hz. Note the overlap: if the 600 Hz band is tweaking 300 Hz and you also want to touch up 1200 then the 2000 Hz section can be enlisted to help.
Why a switch instead of a control? This one will have one set of controls for left and right so a control driving a state variable filter would need four gangs for two channels, and those things in anything but carbon track (ick) are rarer than hen's teeth. So it's gyrator time for this little puppy, although we'll avoid the standard one op amp gyrator which has a host of problems such as noise. Instead we'll use the Antoniou architecture which offers better performance at the price of requiring two gain stages rather than one. Since opamps are relatively cheap, it's no big deal.
The astute observer would note tuning only the inductor means the Q changes drastically across the range. However, the frequency selection switch can also choose a different capacitor to go with the "inductor" (I have a possible circuit on my other computer but the margin of this post is too narrow to contain the proof). Also, since the various frequency sections of the Cello have different Qs and +- gains, the individual tuned sections can be arranged to merge smoothly from one band to another. Doing the sums involves tedious and fiddly arithmetic, however, so I bunged together a spreadsheet to help with the number crunching. It's an unpretty piece of Excel and documentation is sorely lacking, but it serves my purposes.
Given my usual glacial pace of progress I expect this might make an appearance late next year, fair winds and following seas permitting.
Oh, right, the motivation for all this: the EQ is in a distinct box from the rest of the preamp, with a 25 foot (8 metre) cable running to the EQ which will sit beside the listening chair. And it'll be quasi-balanced, with the signal on the hot, circuit ground to neutral, and both lines going to a differential amplifier so it can cancel any interference. Here's a tiny subtlety: hot might want to feed the inverting input to avoid common-mode distortion. Since interference should be relatively small, common-mode interference on the zero-volt send to the non-inverting input will be presented at a lower level than the main signal on the inverting input and thus be undistorted so it can be properly cancelled. I had to keep a gimlet eye on power issues, avoiding ground loops by powering the EQ from the preamp and waving away line impedance via local regulation. Even at +-12 volts any competent opamp has more than enough dynamic range to completely saturate anything downstream.
Anyway, this is all speculative until the soldering iron warms up so I thought y'all might be amused at
the presumption. What will the thing sound like? As I tell my hiking partner before we embark on a tough one, it will be a Voyage Of Discovery. It's a 49-49 proposition: either really good or truly lousy, with a small chance of being somewhere in between.
