Mystery analog board from old computer

[D Straney]

Got this little guy recently with a couple old flatpack ICs, and have been trying to figure out what it is:



There's not much identifying info on the ICs, just "WC 161", and I don't know enough to connect that to anything specific:


I'm not at all a vintage computing expert, but from taking a quick look through some photos, the connector/form factor/lack of handle doesn't really match anything obvious.  The only thing I saw with a similar connector is an early Univac military aircraft computer, but doesn't match that or other Univac systems in any way, as far as I can tell.
The same eBay seller has more boards that look similar to this one: https://www.ebay.com/itm/126231708839
...plus a whole batch of others that look like they're from the same system, and use Fairchild DTµL DTL logic chips, just like the Honeywell 316 (but I can't find a link right now because I think they got sold?).  See attached image at the end.

Using DTL logic suggests that, like the Honeywell 316, it wasn't part of some top-of-the-line pushing-performance-boundaries computer, but instead something meant to be more on the cheap (and maybe reliable) side.

Anyways, the far more interesting part is: what exactly does this card do?  Because of the small number of parts, the low density, and the single-sided routing, this one was supremely easy to trace the circuitry: didn't even have to do the normal "take photos from each side and overlay them" trick.

My best guess is that this is a 2-stage amplifier for pulses (judging by the AC-coupled input), for something like a readout from a ferrite core memory array, where a low-amplitude pulse is amplified until it's clipped, and drives a digital output (open-collector, with pull-up).  The "WC 161" chip probably has a couple cascaded differential pairs of transistors, probably a PNP diff-pair driving an NPN-diff pair (collectors directly connected to the output pins), judging by the external connections.  I'm not confident in this interpretation, and I'm not even 100% sure that the direction of signal flow here is correct, but this is the only way that the schematic fully makes sense to me.

It's interesting to see how the same chip is used in different ways in the 2 different stages, with sets of pins that aren't used in each case.  I think the first stage might be used as a linear amplifier, considering what looks like a DC-biasing feedback path plus some compensation components to stabilize the amplifier, while the second stage is driven to max. gain and used as a comparator (and therefore also doesn't need any compensation).
I'm not sure where pins 9 and 10 connect internally, but it seems to be using a zener-regulated voltage below the analog supply (maybe 5V) to clamp or bias some internal points, maybe as an amplitude limiter, or to keep the internal transistors from saturating when used in this non-linear-comparator-mode to avoid slowing down their turn-off times.
The digital output also gets a pull-up resistor to its own (likely-identical-5V) zener-regulated voltage.

Let me know if you have any better guesses or think I'm wrong about something.

[magic]

Feeding OUT- to IN- would be positive feedback, but it does look like some sort of NFB amplifier, maybe a filter?
Maybe OUT- is an internal emitter follower buffering a common emitter stage at OUT+.

[D Straney]

Oops, you're right, I meant it to be interpreted as negative feedback.  I wasn't paying any attention to the polarities of inputs and outputs (since I have no way of knowing about absolute inverting vs. non-inverting for the whole signal chain), so the "+" and "-" in that schematic are essentially meaningless.

I'm not sure about the emitter-follower theory.  It works on the left-hand chip, but not on the right: the reason I figured both outputs were NPN collectors from a diff. pair is that OUT- is connected straight to ground on the right-hand chip, which is fine for a current-mode output like a collector from a diff. pair, but would cause some problems if you tie an emitter follower straight to ground.

[magic]

Guess you're right, I concentrated on OUT+ pullup and missed the GND connection.
It also struck me that 270Ω would be suspiciously high bias for the CE stage.

[floobydust]

"WC" Wang Computer?  The circuit looks like a tape-drive head amp from the 1960's. Is their a date code anywhere, like on the black capacitors?
Wang 4000 has smaller 15-pin edge card cons and axial electrolytics. 1970 Mod 10 or Mod 7 tape drive used Motorola and Fairchild.
Later tape drives 1977 used diff amp NE592 followed by ECL 10116.

[D Straney]

Oh that's a good guess!  About date codes on the caps:

Larger polarized (tantalum?) ones show:
KEMET P(star)
(capacitance, voltage)
606P10%
...or for the 47µF, the last line is "647AE 10%".
Sounds more like case size or series to me?

Smaller non-polarized ones show, on the side that isn't just capacitance and tolerance:
200
6648 (could be a 1966 date code)