EEVblog #854 - B52 Bomber Automatic Astro Compass Teardown

[EEVblog]

What's inside a Kollsman Instrument Corporation MD-1 Automatic Astro Compass mechanical analog computer from a B52 Bomber?
Part of the star tracker inertial navigation system that calculates the altitude and azimuth.
Thanks to the http://thegeekgroup.org/

[Skimask]

Oooooo.....
There's an old friend (or is it mortal enemy?) I haven't seen (or worked on) in a loooooong time.

[RobertoLG]

man, that was cool!!! teardown, i just wanted to se that thing working, really nice 

[Someone]

The synchro is an early way of communicating an angular position to a remote location:
https://en.wikipedia.org/wiki/Synchro

[rrinker]

 That is some serious electro-mechanical porn right there. Amazing piece of engineering - all those fine tooth gears, those beautiful helical ones on the one shaft that was being turned.. wow. This turned out to be a MUCH better teardown that I originally thought it might after seeing it in the Mailbag. I may be an EE by degree but I do love me some high quality mechanical bits like this - goes back to tearing apart old clocks when I was a kid and manually turning the gears to figure out how the escapement movements worked. I think I would have been there all day myself turning the various gears and seeing what interacted with what on this thing.

[Lightages]

It looks like it is computing the differences between two spherical coordinate systems.

[ez24]

Questions and comments


What happens to items like this?  Will they end up in a museum some day?  Seems Sidney could have a nice electronics museum some day from the stuff people send you.  "Dave's Electronic Museum"

If there is a Keysight scope giveaway - my vote goes to whoever sent this to you.  They deserve it.

Enjoyed this video.

[rrinker]

 That's what those synchros are for, getting input from the MD-1, and probably also operating the display. In searching around I've found some other systems where the mechanical bits were mounted directly behind the panel driving the altitude/azimuth display via direct mechanical linkage
 I'm thinking that springy bit that winds alternately between the two spools is some sort of integrator, or other math function. Might be useful to measure the sizes of the two spools with calipers and see what if any ratio exists between them - they are definitely not the same size. And in the time lapse you can see it winding up on one spool then reversing and going back to the other one.

[jh15]

I'm going to look up some co-workers from Kollsman. The inventor of the first practical aircraft altimeter.

I worked there during this time, and being in EE land, I was amazed at designs coming from experts in other fields, especially mechanical genius.

I bet some of this was headed by S. K. just for privacy stuff.

Hint, later in life he was quite deaf, and used to shoot any more than 4 squirrels at his bird feeder from the kitchen window with a shotgun. His wife didn't appreciate it.

I have the solder tape hanging over my bench as I speak, and is easy to remove with a big ass soldering iron.

We used helium for leak checks, but believe most of our instruments were just nitrogen charged.

We also sold very expensive infrared environmental chamber windows made from turkey oven wrap and aluminum rings designed by this guy to middle east country.

Now I am sure to be put on watch list.

[Monkeh]

If there is a Keysight scope giveaway - my vote goes to whoever sent this to you.  They deserve it.

There isn't a single person there who'd know what to do with one.

And it'd probably walk off like the last, much less valuable scope.

[jh15]

I'd sure know what to do with one, I hate buying Tek 7000 series scopes just to have something nice.

[edy]

From:  http://www.glennsmuseum.com/bombsights/everything.html

B-52 MD-1 Astro Navigator

Before GPS, a primary navigation method for ships, aircraft, spacecraft and missiles was celestial tracking, also known as astronavigation. The basic approach is to sight on the sun or a particular star and calculate the position of the aircraft, missile or ship from the sighting angles to the star. The calculation is based upon highly accurate data about the positioning of the sun and certain reference stars at various time and locations. For modern applications, the star tracking location is used to calibrate and update an inertial navigation system.
Early Air Force aircraft such as early B-52 bombers relied on a human using a sextant through a bubble in the top of the plane. Some time in the late '60s, manual sighting was replaced on the E and F models by this automatic astro tracker mechanism and its associated analog computer. At some point, the astro navigator was replaced by GPS. Our device has a maintenance tag dated 1983, so the system was used at least until then.

Similar "star trackers" were and are used on many other planes such as the SR-71 and RC-135 reconnaissance aircraft, the B-58 bomber, and the P-3 Orion patrol plane. Of special interest is the fact that the most modern USAF plane, the B-2 bomber, still has a automatic star tracker navigation system. Many missiles, including some currently operational ones, also use automated start tracker navigation: the Polaris, Poseidon, Trident, MX, SM-62 Snark and the AGM-28 Hound Dog, and others.

It looks like you would press some buttons which would help you line up sites with some stars and while you were doing this the same inputs would be delivered to the inside of this mechanical computer. The internal mechanism would line up as well in a certain way and the resulting position of cogs and wheels and such would give you some numerical output that represented your position?

That curved part that slides along the other curved track and bobs back and forth is almost like a sextant itself. Here is a good primer:

http://astronavigationdemystified.com/

Fascinating!

[mtdoc]

That was a really cool teardown! 

I need to show this to my dad and get his take on it.  He worked on the Inertial Navigation System for the MX missle in the 70s and 80s. Cool stuff.

[dexters_lab]

great video Dave 

would expect there should be a mix of motors and syncro resolvers in there to get get the data in and out of the unit, this would have to move to and sync with the actual position of the aircraft and then lock in

https://en.wikipedia.org/wiki/Resolver_(electrical)

at a guess it looked like it should be mounted with the connectors (top) facing up, could be the resistance you noticed when it's in the wrong orientation might be deliberate, there could be a clutch arrangement to lock the system up if it sees too much pitch and roll?

wonder too if there could be multiple inputs, ie. would you not have to fall back on dead reckoning / gyros if you cant position from the skymap?

i would be tempted to remove the hacked up base, mount it and use an arduino or something to make it slowly operate as an ornamental piece.

[Smalltalk-80]

What would be the reason for doing this by mechanical means? In the 60s this would seem trivial to do with transistorized op amps. It would be more reliable, lighter and faster.

[HAL-42b]

What would be the reason for doing this by mechanical means? In the 60s this would seem trivial to do with transistorized op amps. It would be more reliable, lighter and faster.

Solving multi - vector equations in realtime was far and away from the newfangled transistors in those days.

The most difficult thing was to make the system work over the poles. The primary mission of B-52 was to fly over the pole and deliver nukes to Russia.

[janoc]

What would be the reason for doing this by mechanical means? In the 60s this would seem trivial to do with transistorized op amps. It would be more reliable, lighter and faster.

Probably because it works and was flight certified. The space shuttle used a variant of the flight computer used on the B52 and it still had core memory right through the 1980s.

B52 was designed in the 50's before transistors were usable (they have been just invented in the 1947!). The transistors didn't become meaningfully usable until sometime in the early 70s or so. Usable solid state opamps were still a long way off.

Also don't forget that this was designed to not only deliver that nuke on target but also to withstand the blast and bring the bomber back home. The EMP blast from a nuke would probably obliterate the period solid state electronics, leaving the bomber blind.

It was good enough and worked sufficiently reliably so that they didn't feel the need to replace it. There has been plenty of similar gear on the period planes - e.g. the famous WWII Norden bombsight or the remotely controlled defensive gun turrets. All driven by mechanical analog computers, using motors and selsyns (synchros).

[Smalltalk-80]

The M1 was only introduced in the mid 60's. Until then it seems sekstants where used, to which this seems to be a direct equivalent. By then electronic transistorized analog computers doing stuff like this was pretty trivial tech.
The Space Shuttle core memory was because it was non violate first and foremost, and then also because it has natural radiation hardening. You'll have to link me to somewhere saying that the Shuttle also used a mechanical flight computer. That wouldn't make any kind of sense. Especially since it is a research vessel and can use radio reliably for the same means.
The bell around the mechanical computer was already plenty shielding for the heights the B52 flew, and could have been made smaller and lighter if electronics where used.
The thing with EMP could be a probable explanation but there are still plenty of electronics in a B52 that would have been affected. Also sheilding could have been employed.

Edit: Minuteman 1962 https://en.m.wikipedia.org/wiki/D-17B#
And that was a digital unjamable computer, for a missile.
I still think an analog electronic computer would be better in a B52 since it is essentially a stream processor and have to react very quickly and reliably for many years.
Analog computery stuff is still used in many many places. It's just not fashionable to talk about. ;-). An example is the tracking in a harddrive. A digital system would be to slow and inaccurate to react there.

[HAL-42b]

The D-17B was probably very highly classified and even the Apollo mission wasn't allowed to use it and had to come up with the Apollo computer.

Also D-17B was purely for inertial guidance.

Besides it was large and heavy and the Apollo mission needed three (minimum two) separate computers so this called for the development of a separate computer from scratch which won't expose any of the details of the nuclear missiles. They used the first integrated circuits in any computer.

As B-52 was very likely to fall into enemy hands it was unthinkable to put the latest technology on it. In case of nuclear exchange not many of them were expected to return anyway. Besides, it was expected to accumulate very many flight hours because they were loitering in the air over Canada 24/7/365 ready to set off for Russia at moment's notice. In this environment a mechanical system is easier to maintain by the existing personnel maintaining the other flight instruments.

[LPaul]

That was a really cool teardown! 

I need to show this to my dad and get his take on it.  He worked on the Inertial Navigation System for the MX missle in the 70s and 80s. Cool stuff.

Looks impressive. Crazy that they still rely on that in the 21th century. Guess there is not that much information available on this ?

[Arne]

Highly interesting ! Cant wait to see it calculate 


The Link provided by Roberto (http://www.prc68.com/I/MD1.shtml) offers an overview of patents relating to the compass. This one (I guess you all already have it printed and hanging over your bench) looks just like it describes the computer you got:

„ Pat. 3042296 Celestial Data Computer, V.E. Carbonara & E.D. MacDonald (Kolsman Inst Corp), Jul 3, 1962, 235/61NV“

https://docs.google.com/viewer?url=patentimages.storage.googleapis.com/pdfs/US3042296.pdf


Maybe you can relate the input/output sections of the patent to those of your computer (Input: “Input shafts of the data computer 150 mainly: L.H.A. (local hour angle) shaft, latitude shaft 152 and declination shaft 158”; Output: Altitude and Azimuth (!) – hence “Computer Altitude/Azimuth”)

So there should be three main input motors (?) … can’t see it on the video unfortunately, but it looks like you’ve been turning the output shafts …

All the best from Germany
Arne

[Kjetil]

Beautiful machinery.

This video explains many of the mechanical/mathematical functions:

[liviuo]

Beautiful machinery.

This video explains many of the mechanical/mathematical functions:

Thank you for sharing the video link !

[Mashpriborintorg]

I have already seen a star tracker dome for sale on ebay, can't remember the final price it went for. And Dave speaks about the sonobuoys in the video, I already had the opportunity to take one apart, it was a old one form the 70, in very bad condition, but still, something you don't have the chance to take apart every day ! 

[rrinker]

  Great find. These old training films are always good information. It seems like a mess of gears and shafts until you see how it breaks down into simple easily understood bits as shown in the video.