Inside a 20-Watt Traveling Wave Tube Amplifier from Apollo | Hacker News Reader

Inside a 20-Watt Traveling Wave Tube Amplifier from Apollo

righto.com

145 pointschmaynard36 comments

Author here if anyone has questions on Apollo communication...

More generally, curious about what your plans are for a 400Hz power supply. There are so many interesting avionics, aerospace, etc. equipment which would be interesting to experiment with. Short of spinning a generator, have you found a VFD or something else OTS that can supply suitable power? Tnx!

CuriousMarc got a VFD box that he thinks will power this, as well as some gyroscopes. I don't know the particular model.

Always enjoy reading your blog. Very informative and entertaining. Thank you so much!

Do you have any I-V plots of the wave tube?

I'm not sure that an I-V plot makes sense for a traveling-wave tube. It's not like a regular triode tube where you're controlling via a grid. Instead, you put in an RF signal and the signal comes out amplified.

Love your articles Ken! Interesting content, straightforward but not dry writing, helpful diagrams and images and footnotes.

Do you know anything about the heart rate telemetry they did?

I'm blown away by the description of the helix that slowed down light by 90% so it could interact with the electron beam. Who figured that out? Brilliant.

The biomedical telemetry was a bit complicated (as was everything in Apollo). When astronauts were outside, there were 7 channels of analog telemetry transmitted by FM to the Lunar Module. The Lunar Module converted these channels to an AM signal which was transmitted over VHF to the Command Module. The Command Module converted these to PCM data, which was then sent to Earth over the unified S-band system.

For details, see "Unified S-Band Telecommunications Techniques for Apollo" page 11. https://core.ac.uk/reader/80678138

> Who figured that out? [traveling-wave tube amplifier]

It was Andrew Haeff at RCA in 1933: https://patents.google.com/patent/US2064469

Haeff was something of a prodigy, looking at his other patents. E.g., a 3D volumetric laser scanner from 1964: https://patents.google.com/patent/US3636250A

Dude didn't let any grass grow under his feet, tech-wise.

Hmm, I wonder if that's where late 60s Alan Watts got this idea:

"with the aid of laser beams, we can translate a physical object, let's take a complicated one, let's take a dandelion flower gone to seed a dandelion clock. You take a dandelion clock about so big, it can be turned into a formula passed through channels, enlarged to any size, say this big, and with laser beams, cut in solid plastic in a matter of moments. You can get this reproduction of a three-dimensional object. But the transition between the two was handled simply in terms of pulses."

"Rudolf Kompfner in 1942 —notably after his public announcement in 1946— when he was secretly working on microwave vacuum tubes for the British Admiralty at the University of Birmingham during World War II. But the history of this device is more complex because the traveling-wave tube was, consecutively, discovered thrice independently"

https://hal.archives-ouvertes.fr/hal-01754885/document

That epoxy-like plastic the components, posts and wires on the boards in https://static.righto.com/images/sband-twt/transformers.jpg are covered in - looks like it might be a way to perform vibration-proofing? Seems like it would be very robust.

Know anything more about this method?

The Juno II rocket launch in 1959 spectacularly failed when the rocket did a U-turn after launch. The cause was two diodes touching due to vibration. After that, rocket components were often potted in plastic to prevent similar problems.

Video of launch: https://www.youtube.com/watch?v=885as-VE50g&t=1446s Discussion of diodes: https://youtu.be/885as-VE50g?t=1688

62 years ago. Just letting that sink in. (Thanks for the video links!)

How does one knew at that time that is the issue … must be a great detective work!

Same for plenty of vehicles, early ECUs were often potted and the ones you see today sometimes still have a resin coating on the component side to harden the device against vibration.

My dad said he referred to those potted assemblies as “integrated circuits” when working on TV bandwidth compression for the phone company in early 1960s australia.

> Most communication went over the "Unified S-Band", which sent voice, data, telemetry, TV, control, and ranging through one unified system.

We'd probably call that "USB" today.

They actually called it USB at the time, which always causes me a moment of confusion.

These are my favorite blog posts on HN, what a wealth of history. The amplification method reminds me of how undersea optical cables have in-line laser pumped amplifiers. Obviously the underlying mechanism is a different one but it also allows power from the amplifier to be added to the signal in a very elegant and 'in medium' way.

> Surprisingly, this amplifier only produced 20 watts of power, not much more than a handheld walkie-talkie.

Humm I thought actual radio power of walkie-talkies were in the mW range (for regulation and power consumption reasons).

But yes, 20W of radio power is not little (actually it is a lot)

Walkie-talkies are mostly .5 to 5 watts, but you can get ones with 20 to 40 watts. I'm not sure how legal those are.

Whether or not they're legal, they're a bad idea to use next to your head.

Higher power handheld transceivers usually feature a corded microphone to allow increased distance between your head and the antenna.

If you're in the US and have an Amateur license, you can generally transmit with up to 1500 watts PEP (Peak Envelope Power). Some lower limits apply; for example a Technician licensee is limited to 200 watts in the HF bands, but may use the full 1500 watts in most VHF/UHF bands.

There is no legal distinction between a handheld and any other kind of transmitter. A transmitter is a transmitter.

There is a legal requirement to use the minimum power needed to carry out the desired communications, so you should not blast out 1500 watts at all times. Especially not next to your head!

http://www.arrl.org/frequency-allocations

Here's a traveling wave tube amp that does 5 kilowatts: https://www.arworld.us/post/5700TP12G18.pdf?44385.7040046296

The tube on the cover of this book is 1 MW at 90 GHz.

https://www.amazon.com/Tube-Guys-Norman-H-Pond/dp/0981692303

It's for an airborne ADS (i.e. pain ray). Good book if you want the detailed history of microwave tubes.

Platinum-cobalt magnets were used in performance-critical TWTs before the advent of rare earth magnets. Might be interesting to drill out a tiny sample of the TWT magnets for analysis.

They can explode if you try that. Those ceramic magnets tend to be quite well balanced and if you drill a hole in them you disturb that balance which leads to a lot of tension in the magnet body. I've had this happen once to me and it was totally unexpected (to me...). Shrapnel from the magnet embedded in the works surface. I was wearing safety glasses. I was cutting the glue layer under the magnet with a diamond dust coated wire to get it loose from its carrier when the whole thing went 'bang', I must have accidentally damaged the magnet to cause this. What remained of it wasn't enough to determine where it went wrong.

It would be interesting to learn more about these systems in the context of today's systems (say, a cell-phone receiving GPS signals), in terms of power received, bandwidth, signal:noise ratio, and so forth.

This kind of comparison can be really enlightening, especially to younger folks. (I'm not one of those folks ... I find it simply amazing how the tiny antenna in my cellphone can receive GPS signals at enough power to be useful for decoding position.)

You may know that GPS is built around pseudo-random sequences. By correlating the known sequence with what you receive, you can extract the signal from a lot of noise. By determining the time offset in the correlation you get the distance from the satellite.

The interesting thing is that Apollo used a very similar system for determining the distance of the spacecraft. They sent a pseudorandom signal from the ground and the spacecraft returned it. By correlating the sent and received signals, they determined the distance of the spacecraft.

It looks so well made, machined block, wow

I’m quite surprised that vacuum tubes could survive the vibration and G forces of launch.

Vacuum tubes can be ridiculously strong, to the point that they would be more reliable than transistors for some applications (vibration proof and higher EMP resilience, for instance). The problems are usually microphony (parts of the tube picking up vibrations from their environment and passing those on to the signal as though they are modulation) and the need to shed a fair amount of heat.

https://en.wikipedia.org/wiki/Nuclear_electromagnetic_pulse#...

I remember when somebody first told me about vacuum tubes in WWII-era proximity fuzes.[1] I thought they were pulling my leg.

[1]: https://www.historynet.com/proximity-fuze.htm

Reply on news.ycombinator.com