The Globus INK: a mechanical navigation computer for Soviet spaceflight | Hacker News Reader

The Globus INK: a mechanical navigation computer for Soviet spaceflight

righto.com

146 pointspicture32 comments

> The name Globus (Глобус) seems to be a nickname, and I suspect it's more commonly used in English than Russian.

Actually, "глобус" is just the Russian word for the familiar model of the globe (like those found in classrooms), which is what this thing is. No mystery there :)

Wow! An analog dead reckoning computer with a beautiful display. I'm especially intrigued by the cam based CVT (constantly variable transmission.) I'm glad to know this exists. Thank you for sharing the details.

Author here if you have questions about the Globus or other space hardware.

This is a naive question, but how does one even begin to conceive of a device like this? What does a "mechanical computer" development life cycle look like?

Not so different from software I would hazard a guess. Document requirements (accuracy over intended usage period, human interface, weight, size) and identify potential subsystems and interfaces. Obtain industrial base with both practical and theoretical expertise in mechanical engineering. Brainstorm possible solutions and compare to requirements. Prototype and iterate. Test. Train.

I think it's basically a process of figuring out the equations you want, then how to "mechanize" them with gears, cams, and differentials. Then it is mechanical engineering to fit all the gears in a box. (The equations are generally much simpler than you'd implement in software, and you're not dealing with algorithms.)

Another thing is that the Globus went through multiple revisions so the original didn't have as much functionality. So it's kind of like any project where you make the minimum viable product and then incrementally add features.

Just like with electronics, it all started with individual components. It only looks so incredibly complex when all those individual units are composed into a big whole.

Underneath it there is order similar to the layering of software though.

Here are a few mechanical computing components explained, the context here was real-time in- and output to control ship guns against ground, sea or air targets, taking into account own speed, angles, angular speeds, distances and their changes, and other factors, like wind, if they could be measured.

https://youtu.be/s1i-dnAH9Y4 -- "Basic Mechanisms In Fire Control Computers - US Navy 1953"

So, the engineers coming up with the top level design probably did not think in "gears" but in higher level computing and transmission units.

Exactly, we think in terms of interfaces. The apollo moon mission technical history is a good example. At a high level there is a big diagram with things like transmissions and modules drawn as single blocks. We define how those blocks need to interact with eachother. Then teams of engineers work on each block, designing and making each block to do what it was defined to do on that big flow chart. Each part is tested to its interfaces (which can be mechanical, electrical, hydraulic, control, etc) and if each part does what it's supposed to the whole thing should.

No questions but thanks to all of you guys. That's one heck fascinating stuff you are working on there and big kudos for researching and documenting the tech history!

I remember seeing this Globus device in all Russian spaceflight photos since I was a kid and we learned about Gagarin and others (I am from ex-communist country) - and nobody has ever published much information about how this thing works, until now. Yet it was a staple of Soviet spaceflight until early 2000s and similar devices flew also on all Soviet space stations (Salyuts, Mir).

> nobody has ever published much information about how this thing works

I’m afraid you are mistaken on that. The instrument has a very extensive wikipedia page since years: https://en.m.wikipedia.org/wiki/Voskhod_Spacecraft_%22Globus...

I kind of missed or misunderstood how this device was supposed to know it's initial position. Was it preset on launch?

Launch is always from the same spot. So yes.

So it just starts ticking on lift off?

Sounds like it.

> The cosmonauts configured the Globus by turning knobs to set the spacecraft's initial position and orbital period. From there, the Globus electromechanically tracked the orbit. Unlike the Apollo Guidance Computer, the Globus did not receive navigational information from an inertial measurement unit (IMU) or other sources, so it did not know the spacecraft's real position. It was purely a display of the predicted position.

I’d imagine they could error correct via ground triangulation of their signals periodically. Or looking out the window for a large distinctive landmark.

Turning the big knob on the front rotates the globe so you can set the position. I would assume that they set the position once they are in orbit and periodically make sure it still accurate.

When I was locked up in the feds I worked at an electronics recycling facility in the prison and we would get old Russian space program stuff like this all the time from NASA to tear down and recycle. The coolest thing was something the tag from NASA said was a "Russian Piano Console" not sure exactly what it was and can't find any info on the net today about it.

This sounds super interesting. Would love to learn more about this device.

What year(s) was that?

It was very recent actually, around 2018 is about the time we were getting in the Russian stuff. I saw alot of very interesting antique items from various government agencies in the three years I was there, plus all the brand new and still usable equipment the government throws away due to the "use it or lose it" budget policy. Lots of waste.

> Although the Globus is mostly mechanical, it has an electronics board with four relays and a transistor

a transistor (!!)

The funny thing about this germanium transistor is that it looks like a TO-3 power transistor package, the large flanged metal. But the Russian packaging is different: the flange is separate from the transistor and clamps down the round transistor package. I'm not sure that makes sense, but here's an eBay photo: https://www.ebay.com/itm/203509582918

Archived for posterity: https://archive.ph/J9mEb

> I'm not sure that makes sense

Maybe this is done in order to decouple the orientation of the leads from that of the holes, so one does not force the other? (Just a guess.)

In fact, back in the day they were making electro-mechanical wall clocks which, too, had exactly one transistor in them. (The transistor controlled the tiny coil that kept the balance wheel moving and, at the same, served as the timing sensor.)

Excellent writeup as usual by Ken. The only additional aspect I wish was included is a video showing the unit in action.

I have a short video here: https://twitter.com/kenshirriff/status/1616849306480156678

No doubt similar systems were installed on the soviet ICBMs, probably still in use today.

> No doubt similar systems were installed on the soviet ICBMs

I very much doubt it. This is a display unit. Has zero purpose on an uncrewed system.

Furthermore ballistic missiles remain suborbital. This device is usefull to track your position on orbit.

Ballistic missiles don’t reenter using retro rockets, so showing the reentry point to the no-existing crew in case they activate the non-existing retro rockets has limited usefullness either.

In short ICBM’s don’t have this or similar systems for the same reasons they don’t have cuckoo clocks. They are not usefull there.

Not exactly, but 8K67 ( NATO: SS-9 "Scarp" and probably SS-19 "Satan" ) had so called ГСП (gyro-stabilized platform) - mechanical device that kept start position orientation. All flight calculations were made against that position. Very cool device inside.

And it was damned precise - allowed to hit targets with 100 meters precision (if I remember that right) on the other side of the globe.

Would this withstand an energetic EM pulse?

E.g., https://en.wikipedia.org/wiki/Nuclear_electromagnetic_pulse

This is pure awesome! It might be a bit of a stretch but I'm hoping the community is able to 3d model this much like the Curta Calculator. I'd really love to be able to 3d print and play with this

You can operate that device in vacuum grade gloves.

Good luck with that on your Tesla :)

Reply on news.ycombinator.com