Looking inside 70's military avionics: navigation computer controls

[D Straney]

Here's the display & control section for the AN/ASN-112 navigation computer, used on the C-5A military transport plane back in the 70's.  Haven't seen any sign of the navigation computers themselves, but the UI sure is interesting to look at on its own!



Made by Northrop:


The displays look like 7-segment digits, with incandescent filaments?


Let's pull off the covers: top/back...

...and bottom:


Power Supply
Turns out, the power supply is in a separate detachable module:




There's a transformer, followed by a couple linear regulators, plus a bulk cap and a smoothing inductor on the back side:


There's two separate LM723 military equivalents in the metal cans, which are general-purpose regulator controllers:

There's also a bunch of stud-mount devices, 3 of which are diodes and 2 of which seem to be the output rectifiers for a center-tapped transformer secondary.

One of them is an SCR.  I originally thought this might be a phase-cut pre-regulator, but tracing out the connections, it looks like it actually is a crowbar across the bulk capacitor.  I don't know if this is over-voltage protection against power spikes (esp. with the series inductor possibly resonating with the bulk cap?).  If it were about giving the digital logic a clean power-down, I'd expect there to be a much easier way to do that in the linear regulator control.


PCBs
You can see some PCBs through the holes behind the power module, so let's pull them out of the bottom (after admiring the really nice hand-wiring):



With the PCBs removed, we can get a better look at the display modules, with a bunch of 5447 BCD-to-decimal decoders/open-collector drivers mounted on bent flex PCBs:


One of the main boards connects to the external connectors & controls...

...while the other board connects mostly just to the other board, and the displays:


The "comms board" has some special metal packages which I'm guessing might be line drivers/receivers for the interface back to the navigation computer.  The outputs, after going through some inverters, seem to go to the other board.  The rest of the ICs here are misc logic & resistor arrays, so this probably has something to do with reading and encoding switch states from the front-panel controls:


The "logic board" has a set of latches which perfectly match what's needed for each of the displays: 4 bits for each digit.  The rest is less clear-cut, although there is some kind of serial interface used for control.  There's 16 bits worth of shift registers that probably have at least something to do with the loading the display registers, and another 4-bit shift register right next to a 4:10 decoder, which might be used to select the "latch enable" inputs on the display registers, or might be used in keypad row/column scanning.


It's interesting to see how the conformal coating is full of little bubbles; definitely less controlled than the more modern stuff I've opened up before:


Controls
With the PCBs removed, you can also get a better view of the controls and wiring on the inside, all nicely connectorized:


It's impressive how much hands-on work went into these overall, between all the hand-written part numbers & serial numbers...

...and the really nice wiring, even outside of the all-hand-wired power supply module:


Wrap-up
Next step is to figure out the pinout for the display module connectors.  I'd like to power up the displays and be able to put numbers on them to use as a readout of some kind (they probably look great when lit, too) but without having to figure out the entire power supply and logic board workings or a 400 Hz power source (which is what's still holding me up on powering up that Teledyne aircraft CDU).  Luckily that should be pretty simple as it's just a bunch of BCD inputs, 1 for each digit, along with power supply and ground - now a matter of doing continuity checks with sewing needles.  Will post some photos when I get that working.

There's another of these left on eBay if anyone's looking for a similar reverse-engineering-and-reuse journey; was able to talk the price down.  There's more photos as well - full album is here.

[coromonadalix]

aaah  seeing theses ... that should have taken a long time to assemble by today standards,  forming wires  etc ..     

thks

[TimFox]

I like the two important knobs in the lower-right of the front panel, with the lever arms opposite the arrow, which would be easy to turn even while wearing gloves.
Many years ago, an engineer who built marine communication equipment wrote into R A Pease' column, describing how he always tested his panels when wearing gloves, due to expected conditions on the bridge of a maritime vessel.
He pointed out that as technology progressed, the space behind the panel got emptier and emptier with smaller components, but that the front panel had to stay large.