Show older

Looking at the precautions for potting compound made me realize how incredibly toxic some of this stuff is. Especially the chemicals used in aerospace for high/low temperature tolerance

Some of this stuff will make your eyebrows fall out with one whiff

All of them need a respirator, long sleeves, and thick gloves to handle safety

Show thread

The Curiosity rover has 2 radiation hardened computers (primary + backup) clocked at a maximum 200MHz. The CPU alone costs $200K, without the mainboard and other support hardware

Plan is to make my own radiation "tolerant" CPU with similar specs, but in RISC-V RV64, in my living room for the same price as a PizzaHut dinner for 2

Piece of cake

Show thread

Found my starting point. The registers can be built with a bunch of these D-type 3 state flip flops:
74ALVCH16821

PDF Datasheet assets.nexperia.com/documents/

Show thread

According to my rough calculations, I need an O'Neil Cylinder built in Earth's Larange point 5, with a diameter no less than 1.14km and length no less than 7.61km, to support a renewable forest for firewood for a single a wood stove in a single orbit

Seasons like winter can be simulated with solar shading and moisture pumped through a central line of needle spray nozzles. The Coriolis effect on falling snow and rain would be remarkable

Maintaining this wouldn't require much processing power

Show thread

Finding bucket seats good enough (and cheap enough) to withstand a launch into orbit and reentry is a challenge

Show thread

This can probably be built in a weekend with a bit of spare stainless steel, some pipes, and maybe a gasket or two

How hard can it be

Show thread

TIL This is the "noise" captured by Voyager 1 in the interstellar medium. I can't be entirely sure there are no space whales

Wonder how practical it would be to build something approaching the Voyager probes' durability in my living room

www-pw.physics.uiowa.edu/voyag

Show thread

There's technically no rule against using bamboo inside a spacecraft, aside from weight considerations (and possible sensitivity to temperature fluctuations)

The only hazard is possibly flammability, but that has countermeasures

Show thread

A DIY space program needs lots of antennas

It's possible to build a fairly adequate parabolic dish (beam waveguide type so the base is stationary) with EMT electrical conduit. The dish just needs to support itself and EMT conduit is fairly easy to bend to the correct shape and is robust enough to survive outdoors even under heavy snow

Show thread

There's a lot of "X has more processing power than the Apollo lander" and similar phrasing and that's usually not a good comparison

Spacecraft hardware is highly specialized and extremely capable in the narrow field in which they're used

From a biological perspective, it would be like comparing brain size and neuron count. Behavior is a lot more accurate indicator of intelligence and capability

I can't even guess how smart an octopus is, but we don't typically use their size to compare

Show thread

And speaking of space hardware, being resilient is often more important than just capability and speed

In fact, most space-rated hardware is going to be slower than their consumer counterparts, but that's OK because the software is highly optimized to run on it

The trick to getting the DIY space program operational on that front is making sure the hardware is resilient and the programming optimized and actually fits the purpose more than increasing raw speed

Show thread

This is by far one of the better animations of a rocker bogie that I've seen in a good while. It moves the middle suspension bar to the rear, which frees up the middle for actual storage, power etc...

Via:
youtube.com/watch?v=49AJeDyDjm

See also:
en.wikipedia.org/wiki/Rocker-b

Show thread

This is by far the simplest practical breakdown of the rocker bogie mechanism that's easy to understand. It's one thing to read papers with angle calculations, pressure delta etc... but another to actually see just the moving components

I think this platform would work quite nicely for the simplest practical rover

wpirover.wordpress.com/2011/11

Show thread

Personal hygiene is important under the DIY space program. Of course, a spinning habitat would make things a lot simpler, but in the meantime, we'll probably go with the no-rinse shampoo too

reddit.com/r/Damnthatsinterest

Show thread

I was looking at the engine detail of the Redstone rocket engine, the heart of the PGM-11 missile. It occurs to me that this particular geometry lends itself quite well to the expander cycle

This particular engine had a turbopump powered by decomposed high-test peroxide

In theory, H2O2 is a pretty neat way to power a turbopump (also used by Copenhagen Suborbitals is using it for their engines) and as a monopropellant for RCS thrusters, but it's kind of a pain to handle and store safely

Show thread

Fiber optic cables don't seem to fare well without special shielding and different methods of construction in a high radiation environment. While they're resistant to transient EM spikes, they do tend to break down and attenuate more of the laser light over longer distances

This is a bit of a dilemma. Either go with copper lines for signaling and sensors and risk spikes and have them acting as antennas in a EM field, or go with fiber. Alternatively, I may have to make my own

Show thread

Also, I may have accidentally landed myself on some kind of list by searching for "radiation sources" too many times

Show thread

I'm both heartened and terrified that some of you are genuinely interested in this backwater space program

Please be patient about the first several (dozen?) explosions

Show thread

Speaking of explosions...

The number of accidents in rocketry is too damn high!

Some of that is the inherent danger of the fuel + oxidizer and the complicated plumbing involved. I have neither the intelligence or patience to grasp all of that mess so if I build a rocket, it will probably use the expander cycle

The plumbing is *far* less complex with fewer moving parts. Which means fewer places for things to go wrong and fewer areas of concern to monitor (less work for the computer)

Show thread

I know the new hotness is Methane right now, but I'd go with Dimethyl Ether as the fuel instead. It's very cheap and can be made from biofuel (possibly on Mars via a different process as well)

As the oxidizer, I'd use Nitrous Oxide. Laughing gas, basically. It's also very cheap and made by the millions of tons. Best of all, it can be decomposed into a very hot gas at 577 C

Dimethyl Ether auto ignites at only 350 C so there's the startup process solved. No need for pyros, sparks, etc...

Show thread

No nonsense that can fail explosively. Just a tiny electric pump to get the N2O going over the catalyst, which is more than hot enough to ignite the mixture

Also, the expander cycle limits the maximum size of an engine, which means you have to build more smaller ones anyway. While that's extra work, you get spares if one or two shutdown mid-flight

And expanders don't have complicated turbopumps which need high temp/high speed moving parts and seals so they're cheaper to build

Show thread

Hmm...

An expander cycle rocket engine with a radiation resistant computer and sensors will probably cost more than the $2.41 I found in my couch this morning

I'm gonna need to engineer this to within an inch of its life

Show thread

Going with Dimethyl Ether and Nitrous Oxide as fuel and oxidizer also because they can be kept liquid at similar temps

-141 C To −24 C for Dimethyl Ether and −90.86 C to −88.48 C for Nitrous Oxide. Which means, they can share a bulkhead in the tanks

This significantly cuts complexity and reduces weight

Also, both have enough expansion at warmer temperatures that I can use "Autogenous pressurization". Same as SpaceX's Starship (though mine will probably be made of glue and aluminum foil)

Show thread

ProTip: The oils on your skin can be just as bad as battery acid depending on the material you're touching. Especially if you can't clean it ever again and will remain in operation for years or decades

Also, welding gloves are better than oven mitts for everything you'd need ovens mitts

Show thread

Technically, there's no rule that you can't use a turbo charger used for a typical small block V8 in a rocket engine as a turbocharger (with a few modifications)

As long as it doesn't blow up immediately, it should work

This plan is flawless and brilliant

Show thread

Apparently a turbocharger from a V8 isn't actually designed to funnel liquid propellant at -89 C at turbopump tolerance. Or survive reentry down from orbital speeds (because reusable rockets or GTFO)

My plan to use a rice cooker to make homemade high-temp composites is probably gonna have issues too

I hope lockdowns lift a bit so I can go trash diving. I'm sure someone's thrown out an orbital grade turbopump somewhere

Show thread

ProTop: Always separate utensils used in your rocketry experiments separate from your food cooking stuff

Graphite tastes terrible and epoxy is toxic

Show thread

I think I'll also be using stainless steel for my rockets so there's basically no difference between me and SpaceX

Show thread

Oh wow. Titanium is kinda expensive. Like much more than I thought it was. Well, it's probably OK to use plain ol' steel as long as it doesn't reach softening/melting temperatures

Show thread

Although there are significant drawbacks with a multi-nozzle rocket engine design, I think for small scale development, the benefits outweigh them. Biggest benefit being combustion stability, which is greater in smaller combustion chambers (see Apollo's F1 issues). The biggest drawback is the additional manufacturer of nozzles and chambers, but once you get the design finalized, they're relatively straightforward to do repeatedly

Show thread

So far, I've been unsuccessful in finding an appropriate spacesuit design that I can make with my own sewing machine and materials from the craft store. Probably will have to settle for a pressure suit made of heat resistant material for now, until I get better at it

I'd like to investigate non-pressurized suits, particularly compression types which use "lines of non-extension". The materials will still be expensive, but at least this will be old research

Show thread
Follow

Today in gloriously-impractical-yet-want, here's an actual core memory unit including the Gerber files for the circuit board and schematic

mikesmods.com/mm-wp/?p=556

Spacecraft don't use these anymore since there are better options, but part of me feels like there's still some value in here

I also wonder if it's feasible to have core memory backed SRAM

So the system reads/writes to SRAM in normal use, as a fast buffer, while it slowly updates the core memory

Might just be crazy enough to work

· · Web · 4 · 9 · 13

I thought I was being very clever when I thought of combining PCBs with core memory

Basically, you include an iron coated ring in one of the inner layers (or maybe several rings) while a via through the middle takes care of the connections to sense and address leads

Well, someone already thought of it before

Back in 1961

But it does mean this may be a feasible way to create modern core memory without resorting to fiddling with tiny ferrites. Maybe even mass-produce it by iron-plating rings

Show thread

If we take "ring" or "ferrite" to just mean "iron loop", then printed core memory may actually be fairly straightforward (hah!)

The iron loop can actually be printed as two iron-plated half loops on either surface of a 4 layer, double-sided PCB, with vias through the middle and on each half end of the loop to the one the opposite side

4 layers is also ample to handle addressing, sense, and inhibit lines

Coincidentally, 4 layers are the cheapest multi layer PCB type available from most vendors

Show thread

I think this is a Crew Return Vehicle concept. There's a lot of space dedicated to provisions and maybe experiments, but not a lot for fuel. I'm guessing there's just enough for a bit of maneuvering and a deorbit burn and nothing else

But that also makes it a feasible build of an orbital spacecraft for a DIY space program that's a bit more exciting than a simple capsule. Geometrically, it's actually an elongated capsule than a "spaceplane", unlike the Shuttle

And you can mount it atop a rocket

Show thread

Oh that's actually not a Crew Return Vehicle, but a full on space shuttle replacement concept by Lockheed. And it's a lot bigger on the inside than the concepts imply. Which means, it's probably not going to be built any time soon (if at all)

The problem with larger Shuttle* concepts is that they're always going to be semi-reusable as long as rockets are involved

Ironically, the SpaceX Starship is probably the way to go as it's the least "plane-like" of all the reusable systems

Show thread

This is a pretty nice video of on building a solar cooker with a parabolic reflector. The form is actually made of concrete so you can make many reflectors repeatably with very good accuracy

If I'm building my own radio observatory, I'd probably also use this technique to build hundreds of parabolic dish antennas

youtube.com/watch?v=9Q_VzJ3P4w

Show thread

Going back to the idea of DIY core memory, I came across this page many moons ago, but lost it. Sadly, the original page is gone, but luckily the Internet Archive has a snapshot

It's a very good primer on the operation of core memory and has a set of demo circuits to operate a small grid of cores (expandable to as many as is practical)

web.archive.org/web/2006120523

Show thread

@cypnk Wow! That's pretty cool. I've actually thought about how to make modern core memory and this looks excellent.

The #CollapseOS crowd will probably want to know about this, maybe even #osnews.

@profoundlynerdy Indeed. No matter what, we've been making tiny iron rings for centuries, so worst case scenario, there's room to store at least a few kilobits with some effort

@cypnk Right. That's an 8×8 matrix. Assuming no bits are reserved for parity: 1024 of those will get you 64kB of RAM.

That is plenty of RAM re-start the information age. Relatively fast (10's of MHz) CPU's can be made with scavenged TTL logic chips.

Punched tape as ROM can bootstrap an OS, phonograph cylinders can store a few hundred kB of program data. Long range communication is possible over #shortwave.

#retrocomputing

@profoundlynerdy @cypnk 1024 would give you only 8KiB (64Kib), not 64KiB (512Kib). Case matters. ;}

@vertigo @cypnk

Hahaha. I stand corrected, good sir. 😀 Such an easy mistake to make.

"…correct the wise, and they will love you." and all that.

@cypnk why core memory when you can use bubble memory

@cypnk @phooky Also, random access. Bubble memory is sequential, like a very fast tape drive.

@cypnk it seems sometimes like everything worth thinking up was thought up in the sixties

must have been a hell of a time to be alive

@fluffy Before the Internet, people actually had to work hard to send messages of discouragement. Of course, encouragement was harder too, but that also meant fewer distractions overall

@cypnk as boring as capsules are, they turn out to be the aerodynamically most stable solution when it comes to re-entry and landing on planets with atmosphere.

Size is the only reason not to use a capsule in my opinion.

That CRV shape remembers me of the early Starship drafts. They changed it got a few changes recently.

@zem There's also something to be said of design simplicity, at least when it comes to earlier spacecraft going back to the 60s. Hence favoring inherent stability

But I also think we can improve beyond seeking stability in favor of long term occupancy (since we now have computer modeling). Size for one, but also better options for maneuvering and landing on the ground. Splashdowns have their own hazards too

@cypnk hmm that's image is from 2009-ish?

Plane-like or not, sending all the thrusters to LEO and returning them might not be the best idea.

Having a high surface area to weight ratio is good? Having an upper stage return alone/just the engine, would help there..

This stuff is all very much not green no matter how you do it.. (pretty much need an orbital ring/launch loop and industry being diverted away from Earth to be that..)

@jasper This was from 2005-ish and was part of a number or Shuttle replacement concepts from various companies at the time. They were quite adamant to replicate the Shuttle back then

The only way I see are smaller, cheaper, scavenger spacecraft being used to mine asteroids and build infrastructure in orbit

That's a lot less damaging to Earth than launching everything from the surface

@cypnk
I have a book somewhere around here called Heaven's Flame by Joseph Radabaugh(? I think) about making solar cookers from cardboard and aluminum foil. Tried it. Didn't work here because we're too windy. Concrete might work.

@Tay0 I think the solar cooker idea in general might only work with a very sturdy surface, and then only in much warmer climates too. I don't know anyone in these parts who got one to work correctly either

@cypnk
Ours worked fine until it was gone with the wind.

@cypnk I wish I could sand like that. Admittedly, I've never tried to the William Tell overture.

@cypnk keep those fridge magnets away :D

Really though, thanks for linking that. I now got something interesting to read. Was getting kind of bored

Sign in to participate in the conversation
Mastodon

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!