https://hackaday.com/2022/05/16/can-you-hear-me-now-lunar-edition/
Despite what it looks like in the movies, it is hard to communicate with astronauts from Earth. There are delays, and space vehicles don't usually have a lot of excess power. Plus everything is moving and Doppler shifting and Faraday rotating. Even today, it is tricky. But how did Apollo manage to send back TV, telemetry, and voice back in 1969? [Ken Shirriff] and friends tell us part of the story in a recent post where he looks at the Apollo premodulation processor.
[...] [Ken] takes us through each module. The voice and data detector module extracted voice on a 30 kHz FM subcarrier. There's also a bi-phase modulator, voice clipping, and a relay module to pass signals from the lunar module back to Earth.
(Score: 2) by dwilson on Sunday May 22 2022, @08:37PM (2 children)
In other words, the same way we'd do it today: With a bunch of specialized hardware spec'd out and hooked together specifically for the task at hand. Except it was 1969 hardware, not 2022 hardware.
Neat to read and see how it was done, but the summery was crap. 0/10 would not RTFS again.
- D
(Score: 0) by Anonymous Coward on Sunday May 22 2022, @09:50PM (1 child)
Care to write a better one?
(Score: 0) by Anonymous Coward on Sunday May 22 2022, @10:52PM
He would, if he knew how to write.
(Score: 0) by Anonymous Coward on Monday May 23 2022, @12:16AM (11 children)
someday somebody is going to explain to my why optimal circuitry requires heat generating elements, called resistors?
(Score: 5, Interesting) by RS3 on Monday May 23 2022, @01:05AM (4 children)
I don't know the name of the logical fallacy, but your question is based on a false assumption: that resistors always generate heat.
Resistors are generally used to limit current, control impedance, gain, produce a voltage based on the current through them.
In engineering we rarely need to account for every microwatt. The current through most of those resistors would be so small that the power dissipation would be essentially negligible.
The power supply voltage regulators in linear (resistive) mode would generate heat, hence their use of switch-mode regulators. They didn't have transistors with the switching speeds we now have, so the regulator would not be as efficient as we could achieve now, but it was still much better than linear / passive regulation.
Bottom line: resistors don't necessarily generate significant heat.
(Score: 0) by Anonymous Coward on Monday May 23 2022, @05:54AM (3 children)
I'm not sure about GP, but you moved the goal posts from the absolute "always" to the relative "significant" right after accusing them of committing a logical fallacy. /pedantry
(Score: 2) by RS3 on Monday May 23 2022, @07:09AM (2 children)
Pedantry is okay, and you've posed a good question.
Often when you read something you have to read the somewhat unwritten message. OP asked: "why optimal circuitry requires heat generating elements, called resistors?"
Yes, I added "always" but I think it's implied, because, otherwise who cares about the resistors?
IE, I'm saying that if the heat generated is insignificant; is below some threshold, then the resistors and their heat are okay.
Stated yet another way, the circuit can still be considered optimal even with resistors, as long as they don't create any significant heat.
Does that explain it? If not, I'm willing to try some more. Clear communication seems to be becoming more and more important and difficult.
Somewhat aside: the line between "significant" and "insignificant" would need to be defined, and should be obvious that it would be defined by external parameters. IE, system designer would tell "black box" designer what it must do, how much power it can dissipate, etc.
Yet another wording: "optimal" does not necessarily mean "perfect", or maximum efficiency, just within the design parameter requirements.
One could argue for higher efficiency.
BTW, in analog / linear "class A" circuits, you're usually going to have power dissipated. It can be in a "load" resistor, or a transistor (or vacuum tube / valve). If the power dissipated becomes significant, we often move to "class AB".
(Score: 1, Insightful) by Anonymous Coward on Monday May 23 2022, @08:23AM (1 child)
Oh, I understand and agree. It's just stupid-o-clock in the morning and it's making me cranky. Cheers and good night.
(Score: 2) by RS3 on Tuesday May 24 2022, @12:09AM
Your point was well made and well taken. I've always struggled with communication- thank you for requesting clarification. If that's you when cranky, you're a gem. Thanks, cheers, and good night to you too.
(Score: 2, Insightful) by Anonymous Coward on Monday May 23 2022, @06:10AM (5 children)
Every element in the circuit produces heat. Thermodynamics doesn't allow any useful work to be done without it. Resistors limit current flow through other components that would otherwise produce even more* heat and possibly destroy themselves. An optimal design is about balancing costs, and sometimes you have to spend a small amount in one place to save more somewhere else.
*Heat is I²R, but current is V/R, so increasing resistance actually results in less heat being produced because it limits the current flow through the circuit. The other option is to lower the supply voltage, but that isn't always possible since many components have a minimum operating voltage.
(Score: 0) by Anonymous Coward on Monday May 23 2022, @01:37PM (3 children)
ohhok, yes. i suppose the electronic components all have resistance and are not super conductors.
however, all but the resistor, elements provide a "function". the resistor only provides heat.
not electric phd here but methinks 99.999% of electric circuitry are build from, starting from a energy source. it just magically exists. so resistors are okay.
however, we now have components that are energy harvestors, like the one translating light energy to electricity or heat differences to electricity or ambient radio waves.
well, nevermind, i don't know what my point was anymore :/ guess batteries empty.
(Score: 0) by Anonymous Coward on Monday May 23 2022, @07:56PM (2 children)
I think your attempted point was based on the misunderstanding that resistors only produce heat without any other benefit or purpose. Maybe RS3 and I haven't done a very good job of explaining it, but resistors do actually provide the necessary and useful function of limiting the current that flows through other components. I hope that makes it clear enough. Cheers.
(Score: 2) by RS3 on Tuesday May 24 2022, @12:47AM (1 child)
How about this: resistors are pretty much a necessary evil. Or better stated, they're the best compromise to provide the necessary functionality- again, limiting current, or providing a voltage drop for a given current.
I think (surmise, deduce,...) previous AC's point is that in theory, things, like resistors, that produce heat are an inefficiency in electronic circuitry, and it'd be great if there was a way to do the same job without the energy loss.
In a linear (analog) circuit you can't get away from some heat loss. Basically, to amplify or otherwise process an analog signal, you generally use transistors (same applies to vacuum tubes (valves)). They need to be "biased" in the ON / linear operation region to get a clean (minimal distortion) output signal. We generally call this Class A [electronics-tutorials.ws] circuits.
There are circuits, like most modern op-amps, that minimize the quiescent [merriam-webster.com] power by carefully running the circuits in Class B (or AB) [electronics-tutorials.ws] mode. There are op-amps which consume microwatts when idle.
https://www.analog.com/en/technical-articles/low-power-op-amps-only-20ua-for13mhz-240ua-for-30mhz.html [analog.com]
https://www.mouser.com/new/texas-instruments/ti-tlv854x-op-amps/ [mouser.com]
Again, the total energy loss is quite tiny and in most cases negligible.
For larger power uses like large audio power amplifiers, motor controllers, etc., more and more switching ("Class D" and up) circuits are being used. They're much like "switching" (switch-mode) power supplies.
Some numbers: Class A circuits are theoretically 50% efficient, but in practice more like 30%. Class B, which is typically implemented in a "complementary" or "quasi-complementary" circuit (referenced above) can be 75% efficient in theory, and in practice ~ 60 - 70%.
Class D amplifiers can be above 90% efficiency, and in high-power sizes, will weigh far less.
(Score: 1, Interesting) by Anonymous Coward on Tuesday May 24 2022, @05:41PM
"The best compromise" is a good way of putting it. Engineering is fundamentally about making trade-offs, because everything has a price.
It sounds like a nice idea on the surface but ideals based on misunderstandings or ignorance rarely end well in practice. All too often trying to do X without Y ends with the discovery that Y was needed for something important. A fun example is to image what the world would be like without friction.
There is actually a class of circuit that doesn't consume any energy when operating, called a reversible computer [wikipedia.org]. This of course ignores the energy cost of cooling the superconductors, but the computation itself is perfectly efficient because no information is lost. The flip-side is that conventional non-reversible computing must consume energy because it loses information. There is some deep physics in that because it implies that heat is lost information.
(Score: 2) by RS3 on Tuesday May 24 2022, @09:32PM
You're generally correct, but a significant contrary example is an induction motor. If you reduce the voltage into a motor running a moderate to heavy load, you'll increase the current drawn. Better said, the motor will draw more current than proportional to the voltage drop. Because, the motor becomes less efficient, and due to the way induction motors transfer energy to the rotor, at the same time being a motor, you'll cause "slip" to increase, and more I²R losses in the rotor and stator.
Slip: the amount the rotor turns more slowly than it theoretically should based on the input frequency and number of "poles" (synchronous speed).
https://blog.isa.org/autoquiz-slip-ac-induction-motor [isa.org]
even more fun:
https://www.engineeringtoolbox.com/electrical-motor-slip-d_652.html [engineeringtoolbox.com]
(Score: 4, Interesting) by janrinok on Monday May 23 2022, @06:29AM (5 children)
I'm not sure why, but I imagined that the various boxes in the numerous systems that went to the moon wouldn't have looked the same as the stuff I was building around the same time. I had always imagined that they would have been neater, or more compact, or something...
OK, there are differences - I didn't use spot welding for a start! But my soldered joints always did the job and the rat's nest of wires and components, which were very similar to those in the diagrams, also seemed to work.
I look at how it is done nowadays. I cannot even manipulate the ultra small components that are essential for many circuit boards. A simple sneeze and they would be gone from my bench top with little chance of them ever being re-identified even if I could recover them.
(Score: 2) by RS3 on Monday May 23 2022, @07:23AM (4 children)
I have a great Nikon stereo microscope, some Swiss tweezers, micro soldering iron, etc. Yes, the tiny stuff is getting ridiculous. It kills me to see a pc board, like in a phone, resistors the size of a pinpoint, and large empty spaces on the board. I guess if you're geared up to do pick-and-place parts that small, you just use them for everything (except power dissipation stuff).
Maybe I'm not seeing it correctly, but it looks to me like some of the components are soldered to the posts.
I've done some cobbling, but never anything quite that cobbled! I've always used "perf board". I'm not sure why they build the circuits that way. Seems inefficient. But totally how they used to build tube (valve, okay? :^) radios, TVs, etc. Who figured out which parts to put in first? Yikes!
That said, ever look closely at the plumbing on a jet engine? Or worse yet, a rocket engine?
(Score: 3, Interesting) by janrinok on Monday May 23 2022, @08:37AM (3 children)
Some of the joints appear soldered to me too - but then again I'm not sure what I would see if they were using some form of spot welding for components as reported in the TFA.
I also use perf board of various makes and designs fairly extensively. The advantage in the move to ever small components is that it has meant that buying 0.1w resistors etc has become a bit cheaper for me. And the bulk packs of transistors are almost at give-away prices and certainly good enough for testing ideas before investing in better quality parts for the final product. I've stopped working on radio and TV, and spend much more of my time on robotics, and SBC projects.
I was crew (EW specialist) on several military aircraft so we should swap horror stories someday.
(Score: 3, Interesting) by RS3 on Monday May 23 2022, @08:54AM (2 children)
Oh, that's awesome, much honor and respect to you.
I've never worked on one, just seen them. That said, I do some pretty complex stuff so I'm sure I could learn it. I was referring more to the crazy plumbing- all over the place. Seems random, Dr. Seuss-esque.
https://static.righto.com/images/apollo-pmp/wiring3.jpg [righto.com]
Black resistor, upper left, lead wrapped around and soldered to upper center post. But many other connections are the spot weld- no solder. I guess I'm curious why they went to the trouble to spot-weld, if solder was okay.
I don't remember ever coming across circuits that looked like these. It would be interesting to learn about the design / construction philosophy.
(Score: 1, Informative) by Anonymous Coward on Monday May 23 2022, @09:34PM (1 child)
If you have the equipment then spot welding is faster and more reliable (less chance of a cold joint) than hand soldering. Considering the application the weight of the solder might be a factor as well. I'm not sure why that one is soldered but if I had to guess I'd say they needed more mechanical strength for that joint due to the longer span the resistor bridges. It looks like they tried gluing it as well. There is another solder joint in the bottom left that appears to be a ground connection, so it seems to be a case of welding what you can and soldering what you can't.
It's point-to-point wiring [wikipedia.org]. I don't recall seeing anything that high end before, but it was common in period electronics when they wanted to save space.
(Score: 2) by RS3 on Tuesday May 24 2022, @08:52PM
All excellent points and info, thanks.
Not disagreeing, just posing a possible alternate:
That one soldered resistor looks to be a high-precision one. My hunch (educated guess?) is it was selected to tune/balance/bias the circuits later in the build/test/calibrate process; and maybe potentially changed out more than once after other ones installed the same way. Most applications would use a "trimpot" (variable resistor for those who don't know that term) but any potentiometer could become "flaky" at a bad time, so no pots allowed in flight.