I’m likely interested in 1 or 2 sets of the A820 cards. From my perspective, the closer they can be to a direct plug-in replacement, the better.
The convenience of having multiple EQs and speeds easily available is a really nice feature which ideally would be maintained. However I presume that the digital select logic and multiplying DACs take up more board space than multiple sets of trimpots and a selector switch which is probably why you are thinking of going that direction. Plus doubtless it simplifies things. So if that capability has to be dropped, in favor of a selector switch on the card, it probably wouldn’t be a deal killer for me.
My strong preference would be to not simply bypass the headblock preamp. And if a replacement is created (likely required to get the quality you are after), it would be very nice if it was compatible with the current one so that headblocks are interchangeable between machines with the existing and your new reproduce amplifiers.
I presume it doesn’t make sense to also create replacement Line Amplifier cards too? While that would give you twice as much board space to work with, it probably also means that there’d be lots of additional stuff you’d have to interface with. I’m guessing that trade off doesn’t make sense?
Thanks very much for your interest and valuable feedback; you are absolutely correct regarding footprint taken up by the M-DACS and the associated NE5532A I/V buffer(s) on each; we will use the identical small multi-turn Bourns trimpots as used on our NextGen A80 repro card that Bruce Brown reviewed most favorably.
As far as the original headblock preamp is concerned, it is simply not a great design - even for its era - and makes numerous serious compromises, among them: the rather high noise and Ib/DC head current, mediocre BW/slew/rate/settling time and linearity (distortion) and it even compromises XTLK compensation by sharing just a single adjustment pot for both L-R and R-> L setting where there should be two (even the older A80 had individual L-R and R-L XTLK adjustments). All the sound you hear from the machine in repro starts with this headblock preamp, yet I would think the entire 2-ch headblock preamp could not have cost Studer more than $20-25 in materials, assembly, and test, this for a machine I believe cost more than $10K when released, and A820 machine up around $20K (someone please correct me if my recollection is wrong on these prices). So the headamp will either have to be bypassed or be replaced entirely.
As to the Line cards, certainly they also do not stand up particularly well, and I suppose if we designed a replacement Line Card we could pick up quite a bit of space on these cards as well as improve greatly performance - but how much could actually be utilized to extend the footprint of a repro card function is not entirely clear given existing backplane connectivity as you say, as well as software implications, and at the very least still require two new cards. If workable it would perhaps eliminate a new headblock preamp and require bypass the existing preamp.
Like you, I certainly prefer a new headblock 2-ch preamp IF it would allow us to use through-hole manufacture of the associated Repro Card; if it would not, and we need surface mount manufacture for the Repro Card with the built-in preamp, it may make more (economic) sense to bypass the Headblock preamp, but only once we verify (as I expect) both approaches will have identical noise floor and RFI immunity.
In our current NextGen A80 class A preamplifier (or Equalizer) stage we have 11 discrete transistors (jFets and bipolars), about 20 precision resistors and film caps, 2 trimpots (DC balance and XTLK) etc.
per channel. The current Studer headblock preamp has about 2 square inches footprint (and 1.5 cubic inches total) available for the two preamp channels, so we are looking at a new headblock preamp design with two stacked surface mount cards, to replace the current Studer 2-ch head block preamp design as the most likely solution, and one that is as compromise free of performance limitations as can be built today, at any price.
The Repro Card will contain the Equalizer, Rotary switch for EQ selection, Front-panel Breakout connector for the XLR and direct single-ended outputs, trimpot access, mute circuit, and a small toggle switch for Hi-Lo speed select (adjusts the head gap- and wavelength modeling- for the LS).
We are not at this time planning to support the Sync function (repro from the Rec Head), because of the additional space demands om the Repro Card..
But please NOTE that this new headblock preamp can simultaneously drive a second set of entirely outboard electronics (like our A80 NextGen cards less the redundant preamplifier stage) so there exists the option of retaining the original Studer Repro card with its Sync function and using the superior outboard electronics for the best repro performance. This adds another factor to consider.
Einstein famously advised "Things should be made as simple as possible, but no simpler."
I expect we will know quite soon which configuration
(1) new surface mount headblock preamp and through-hole Record Card; headblock returned by most users to us for installation and test;
or
(2) User-bypassed headblock preamp and new surface mount Record Card with inbuilt preamp;
makes the most sense from both an economic-, and and user- preference point of view, assuming both deliver identical audio performance (which will be verified). For now we still carry forward both approaches.
Users: please chime in if you have a strong preference.
Regards
Jeff