boostrapping a mars colony

Discuss life, the universe, and everything with other members of this site. Get to know your fellow polywell enthusiasts.

Moderators: tonybarry, MSimon

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Re: boostrapping a mars colony

Post by happyjack27 »

    There's also the problem of available resource on Mars. While we can do chemical reactions and we have then only issues of exothermically and endothermic and catalysts, silicon is an element.

    Short of turning iron to gold, where would we get the silicon?

    It strikes me that there is mention of "silica" on Mars. Presumably for making glass for greenhouses. But perhaps also - for making semiconductors. Now we just need to dope it - with P or N, or "Metal Oxide".

    happyjack27
    Posts: 1439
    Joined: Wed Jul 14, 2010 5:27 pm

    Re: boostrapping a mars colony

    Post by happyjack27 »

    paperburn1 wrote:Another thing to consider is to drop back a level of technology. This is a time proven tech for other exploration of our planet. Go back to the era when chips and transistors were easier to work on and repair of boards was done the old fashion way with a meter and O-scope. bring back the tube even.
    As per one site I am interested in being evolved with
    http://opensourceecology.org/wiki/Globa ... uction_Set
    for when things go really south you can rebuild. After all we all can not go to Mars

    two min video explaining it all
    https://vimeo.com/16106427

    I don't think vacuum tubes are viable.

    Bear in mind that when we did vacuum tubes we didn't know about "electron holes".

    If we had the same level of technology when we had vacuum tubes, but we knew about semiconductors...

    Semiconductors, when you know the physics, are not that hard to make, and on earth, at least, far cheaper.

    It took Mr. Cray - yes, that Cray - a bit of convincing to switch to silicon. But need I remind you the first transistor - which had NO production line , was not only smaller than a vacuum tube, but faster switching and dissipated less energy. (Heat)

    The chronology of a Mars colony need not follow that of an Earth one. If we have anything to "bootstrap" it, it's knowledge.

    The Greek tragedy "Prometheus" is named after the main character, whose name literally translates to "forethought". Perhaps in this endeavor we need precisely the opposite: retrospect.

    I wonder, if Prometheus had taken into account the whims of Zeus, might he not been bound?

    Tom Ligon
    Posts: 1871
    Joined: Wed Aug 22, 2007 1:23 am
    Location: Northern Virginia
    Contact:

    Re: boostrapping a mars colony

    Post by Tom Ligon »

    happyjack27 wrote:
      There's also the problem of available resource on Mars. While we can do chemical reactions and we have then only issues of exothermically and endothermic and catalysts, silicon is an element.

      Short of turning iron to gold, where would we get the silicon?

      It strikes me that there is mention of "silica" on Mars. Presumably for making glass for greenhouses. But perhaps also - for making semiconductors. Now we just need to dope it - with P or N, or "Metal Oxide".
      Mars is lousy with silicates. Overall the composition is not that dissimilar to Earth's rocks, except that Mars is mostly igneous or "weathered" igneous minerals instead of metamorphic forms present on Earth unique to plate tectonics and our abundant water. We have some ores here produced by life and by hydrology that won't be present there. Our banded iron ore is an example. The elements are there but may not be pre-processed quite the way we're used to.

      https://upload.wikimedia.org/wikipedia/ ... 121203.jpg

      To my mind, that puts power and plenty of it as a high priority. Solar? Too wimpy for serious industry, such as extracting minerals from marginal ores. In keeping with the theme of the forum, Polywells would be perfect. And I probably should remind you folks that they are, in fact, vacuum tube technology. They use electron emitters and those obey Child-Langmuir law.

      paperburn1
      Posts: 2484
      Joined: Fri Jun 19, 2009 5:53 am
      Location: Third rock from the sun.

      Re: boostrapping a mars colony

      Post by paperburn1 »

      Way back in the salad days of digital computing (the 1940s and ’50s), computers were made of vacuum tubes — big, hot, clunky devices that, when you got right down to it, were essentially glorified light bulbs. This is why early computers like the ENIAC weighed more than 27 tons and consumed more power than a small town. Later, obviously, vacuum tubes would be replaced by the solid-state transistor — which would allow for the creation of smaller, faster, cheaper, and more reliable computers. Fast forward to 2014, though, and the humble CMOS field-effect transistor (FET) is starting to show its age. We’ve pretty much hit the limit on shrinking silicon transistors any further, and they can’t operate at speeds much faster than a few gigahertz. Which is why NASA’s Ames Research Center is going back to the future with its new vacuum transistor — a nanometer-scale vacuum tube that, in early testing, has reached speeds of up to 460GHz.

      The main problem,with old style tubes was that the cathode had to be heated up by a filament so that it would emit electrons — and where there’s heat, there’s a lot of power consumption and a lot wear and tear. As you may know, it wasn’t unusual for old tube-based computers to break down every few hours whenever a tube burnt out.
      So, anyway, the high running costs and frustrations of operating a tube-based computer were eventually assuaged by the discovery of processes that allowed for the cheap and plentiful production of integrated circuits with solid-state MOSFETs. And, in the last 40 years or so, we haven’t looked back. Until now.


      NASA’s Ames Research Center has been working on-and-off for many years on the vacuum-channel transistor, which is essentially a vacuum tube that can be fabricated using conventional CMOS techniques. Instead of a gate sitting between the source and drain, there is… nothing. A vacuum. By a method known as field emission, electrons are drawn across the vacuum from the source to the drain when a current is applied to the gate By using field emission rather than the thermionic (hot) electron emission, these vacuum-channel transistors don’t require a heat source. Importantly, they also don’t require a vacuum — instead they use helium (it’s sparse enough that the electrons have almost no chance of bumping into helium atoms while they traverse the few-nanometer gap between source and drain). The electrons also traverse the air gap a lot faster than if they had to pass through the gate electrode.

      All in all, these vacuum-channel transistors, crazy as they sound, are surprisingly feasible. They can be fabricated using existing processes. The helium packaging is tougher, but the NASA researchers think that the techniques used to package current microelectromechanical sensors (gyroscopes, accelerometers) should be suitable for their unorthodox transistors as well. In early testing, a vacuum-channel transistor was able to operate at 460GHz — on the order of 10 times faster than a conventional, silicon-based FET There’s no word on whether these vacuum-channel transistors will allow for the creation of small, modern audio amplifiers that sound like original tube amps — but maybe!
      Cut and pasted but you get the idea, tubes might be the way to go if mars has not lost all of its helium over the eons Oh wait the polywell reactor can make that for us. :D
      I am not a nuclear physicist, but play one on the internet.

      zapkitty
      Posts: 267
      Joined: Fri Apr 09, 2010 8:13 pm

      Re: boostrapping a mars colony

      Post by zapkitty »

      Tom Ligon wrote:Upside: No danged lawyers. No environmental laws. The environment is toxic anyway... product that's high value per unit mass and noxious to manufacture on Earth... isolated biological test facilities.... agricultural pests or genetically modified organism, and ways to deal with them... Mars is a natural for running tests too risky to do on Earth.
      And so much for the brave new frontier of Mars.

      First Musk declares a policy of genocide against any pre-existing Martian life that might get in the way of Manifest Destiny and now the fourth planet is to become a fourth-world toxic waste tax haven for transplanetary elites... :twisted:

      vasimv
      Posts: 15
      Joined: Mon May 16, 2011 7:20 am
      Location: Moscow, Russia

      Re: boostrapping a mars colony

      Post by vasimv »

      kunkmiester wrote:You'll need power transistors too then. Fpgas are not the only silicon based parts you'll need.
      I've mentoined igbt/mosfets already. To be fully correct, need some different stuff:

      1. FPGAs and microcontrollers
      2. SDRAM and FLASH memory chips (that "spartan 6" does have some onboard but not enough for some memory hungry stuff)
      3. IGBTs and MOSFETs (ranged from small up to high power ones)
      4. Surface mount resistors and capacitors (well, they are really small and light, one can bring them in millions)
      5. Inductors for DC-DC converters
      6. LEDs and some LCD/OLED screens
      7. Wires with insulation (most heavy stuff in the list)
      8. Connectors
      9. Copper foil (to create PCBs by etching or engraving) or some kind of printing material to print it, until they will able mine copper
      10. Solder paste (with the lead, to prevent mass disasters because "Pb free" stuff is too fragile)

      Last 4 items could be produced right on the Mars but will require working metallurgy and chemical factories.

      With these parts you can build literally any electoronics.

      Effective motors are quite hard to build. Although, after some time they should able to create primitive motors and generators (with electromagnets). But at the start you'll need some with good rare-metal magnets to deliver from the Earth.

      Btw, we have "printed electronics" technologies already (like printed LCD/OLED screens on flexible films). But i don't think you will able good MCU/CPU with these yet.

      Tom Ligon
      Posts: 1871
      Joined: Wed Aug 22, 2007 1:23 am
      Location: Northern Virginia
      Contact:

      Re: boostrapping a mars colony

      Post by Tom Ligon »

      paperburn1 wrote: Fast forward to 2014, though, and the humble CMOS field-effect transistor (FET) is starting to show its age. We’ve pretty much hit the limit on shrinking silicon transistors any further, and they can’t operate at speeds much faster than a few gigahertz.
      Are they still building CMOS processors? I thought they were showing their age by the 1990's. The last time I paid any real attention, most of the really fast ones were gallium arsenide, and to overclock them for more speed they've stated making them water-cooled. Now they're talking seriously about quantum computing.

      The venerable MOSTEK 6502 that was in the early Commodores and Apples was a remarkable little chip, capable of running on something like 2 volts at 2 mA, but the clock speed was flat out at a few MHz. Coupled with CMOS static RAM, you could build a computer that could run on a couple of D cells for a year. I don't recall CMOS ever being able to clock at GHz speeds, due to the high input impedance of the transistors. The gates were essentially low-leakage capacitors and their basic nature was slow, grossly impedance-mismatched for higher frequencies. Within the last few years I've been looking for a comparable technology, hoping to find an off-the-shelf single board computer with that low power consumption, to use in low-data-rate datalogging. I couldn't come up with anything.

      What, exactly, do you want with a computer that can run at half a terahertz for? Sure, if you want to model a Polywell at the particle level, a really quick supercomputer might be nice. If you're rendering 4K 3D video for an immersive movie experience, yeah. And I can see that Windows 15 will probably need that kind of speed just to parse thru the 20 TB operating system, and run its antivirus software. But to control your automated tunnel digging machine? A 6502 could do the job.

      93143
      Posts: 1142
      Joined: Fri Oct 19, 2007 7:51 pm

      Re: boostrapping a mars colony

      Post by 93143 »

      Technically the original 6502 was NMOS (as was the unlicensed clone in the Famicom/NES, if I'm not mistaken). The CMOS version was the 65C02, and it fixed a few bugs as well as tossing out the undocumented opcodes in favour of additional "real" opcodes.

      My experience, such as it is, is with the 5A22, which is a licensed clone of the 65C816 (a flexible CMOS 8-bit/16-bit hybrid with a degree of 6502 backward compatibility) with extra doodads. It was the CPU in the Super Famicom/SNES. Its main issue (aside from the difficulty of programming it, which I can't personally vouch for as I have little or no assembly experience with anything else) was that what with the accumulator-based architecture, the 8-bit data bus and the half-cycle strobe inherited from the 6502, it gave the memory bus a real workout, with the result that it needed 120 ns memory to run at 3.58 MHz. More powerful per cycle than the Motorola 68000, with far lower power consumption, but unable to run as fast... The fastest I've ever heard of a 65C816 running is 20 MHz (the Commodore 64 SuperCPU, released in 1997), though reportedly FPGA versions can hit 100 MHz, so I'm sure these things could be made to run faster if you really needed them to.

      It is kinda pathetic that modern computers require so much power for what should be (and in some cases are) extremely simple tasks. I believe it's called Wirth's Law...
      Last edited by 93143 on Sat Oct 29, 2016 4:22 pm, edited 1 time in total.

      kunkmiester
      Posts: 892
      Joined: Thu Mar 12, 2009 3:51 pm
      Contact:

      Re: boostrapping a mars colony

      Post by kunkmiester »

      Speed is not the driving force behind computer power anymore, last I heard. You could probably get plenty of power with a simpler architecture utilizing some of the newer techniques.

      Smt resistors and capacitors are thin film devices, I'm pretty sure they could be fabbed with a roll of carbon or Mylar and a proper fab.

      A microcircuit fab, component fab, discrete (power transistors and such) component fab, and a circuit board fab would converge on a pick and place that would put together whatever you're making. Pick when each part is shipped.
      Evil is evil, no matter how small

      paperburn1
      Posts: 2484
      Joined: Fri Jun 19, 2009 5:53 am
      Location: Third rock from the sun.

      Re: boostrapping a mars colony

      Post by paperburn1 »

      Like I earlier suggested and what Ligon and kunkmiester have said is basically true , dropping back a level to some of the older tech could be very beneficial. Taking care of basic housekeeping chores and like. At our work we have been playing with the raspberry pies as HLA gateways , Vpn tunnels and one process units(a inventory machine and a logbook, other things I can't mention). At seven dollars apiece the do everything a full blown PC does at 1 hundredth of the cost.
      Slowerer but fast enough to be real time for us punny humans.
      I am not a nuclear physicist, but play one on the internet.

      kunkmiester
      Posts: 892
      Joined: Thu Mar 12, 2009 3:51 pm
      Contact:

      Re: boostrapping a mars colony

      Post by kunkmiester »

      Honestly something like an 8088 can probably do everything except play the latest games, not even sure you'd need much for browsing the internet, especially once multi core and multi threading have been taken into account.
      Evil is evil, no matter how small

      vasimv
      Posts: 15
      Joined: Mon May 16, 2011 7:20 am
      Location: Moscow, Russia

      Re: boostrapping a mars colony

      Post by vasimv »

      Actually, you need high speed microcontrollers and FPGAs. Not for complex math calculations but to poll/process data from sensors, adjust output values (PWM on motors and such) and replace/emulate some stuff like serial interface driver. Slow MCUs will require a lot of additional stuff to be useful - ADC/DAC chips, buffer memory, serial interface chips and such. 6502/8088/etc aren't good examples as they will be quite useless in MCU's role (also plainly bigger, eat shitload of power and require 10+ support chips to work).

      Also there will be lack of programmers on the Mars to code and optimize stuff for slow processors. To bring just one average programmer will require 300+ kilograms Earth-to-Mars delivery, so much smarter to bring 100000 fast FPGAs/MCUs instead (that should be enough for many years).

      Tom Ligon
      Posts: 1871
      Joined: Wed Aug 22, 2007 1:23 am
      Location: Northern Virginia
      Contact:

      Re: boostrapping a mars colony

      Post by Tom Ligon »

      Scanning sensors on some of these industrial microcontrollers can be done using their TPUs, unloading the main processor. Timer Processing Units can handle things like high speed sampling, digital filtering, etc. Even models made in the 90's had no problem with jobs like UAV control.

      And you don't need nearly that for things like temperature or environmental datalogging. I specifically want something to take measurements a few times a day on beehives. Once an hour would be overkill. I presently have an industrial datalogger that can't be set slower than a 1Hz rate, which produces an insane amount of data for many of my jobs. Beekeepers typically don't have line power at their apiaries, so a logger that can run on batteries or a small solar panel is a real plus. But the lowest-power "PC" I've found on the market lately draws 5W. A CMOS system might be made to run on 5 mW.

      Where lighning-fast speed is essential is in controlling something like a Polywell.

      As for the programming being inordinately difficult, I think that's baloney. The present crop of dolts writing scripts may be in over their heads, but assembly language is not all that difficult to learn, and data acquisition is a snap. I learned it cold turkey. I wrote my first driver in a couple of evenings on a TRS-80 Model III, printed out the hex code, loaded it into high memory on a NorthStar Horizon, and it ran perfectly on the first try. It was a pretty tricky scan, too. So I've got little patience with people who think it is Too Hard. I also have very little respect for people who don't try to write compact and optimized code, but instead expect faster processors to make up for their slovenly habits.

      Also, the programmers don't have to be on Mars. Code travels at the speed of light.

      kunkmiester
      Posts: 892
      Joined: Thu Mar 12, 2009 3:51 pm
      Contact:

      Re: boostrapping a mars colony

      Post by kunkmiester »

      As Tom said, while an on site programmer migbt be nice, he mostly would be auditing code written on earth.

      Not sure how much power the Atmel chips the Arduino uses, but they can be done solar powered. Probably wouldn't need batteries- -use the drop during the night to register your day cycle, unless you need measurements during the night. You can have it log to an SD card too.

      A chip like that would be nifty though, for multiple purposes. While developed for Mars, with money invested by potential immigrants, and profits reaped on earth from such applications long before a rocket leaves. Just need to find a financial planner.
      Evil is evil, no matter how small

      vasimv
      Posts: 15
      Joined: Mon May 16, 2011 7:20 am
      Location: Moscow, Russia

      Re: boostrapping a mars colony

      Post by vasimv »

      kunkmiester wrote:As Tom said, while an on site programmer migbt be nice, he mostly would be auditing code written on earth.
      Of course, most of code will be written on the Earth in any case. But you will need a programmer on the Mars. Debugging is not possible with 6..44 minutes delay in response. And you can not count colony as "selfsustaining" if they can't create a piece of code for their own controllers.
      kunkmiester wrote:Not sure how much power the Atmel chips the Arduino uses, but they can be done solar powered. Probably wouldn't need batteries- -use the drop during the night to register your day cycle, unless you need measurements during the night. You can have it log to an SD card too.

      A chip like that would be nifty though, for multiple purposes. While developed for Mars, with money invested by potential immigrants, and profits reaped on earth from such applications long before a rocket leaves. Just need to find a financial planner.
      Uhm, atmega chips are new generation comparing with 6502/8088 chips. It uses 130 nanometer (atmega) vs 3..8 micrometer (8088 and 6502) technology. Btw, Spartan 6 FGPA is 45 nanometer technology, so if you are able to build factory producing atmega-like chips - means you have modern semiconductor factory already. I think it is out of question, you will have import MCUs/FPGAs from the Earth, tens of years at least. Even one hundred kilograms shipment is enough for many years, so why to spend resources in attempt to build weak processor on the Mars, that will require much more on-site programmers, hundred times more space on PCBs (with all support stuff), power and many times slower?

      Post Reply