Took an old 42U rackmount cabinet and turned it into a dehydrator, last year. Cleaned it out and used light diffuser grille for shelving, a little table fan for air movement and a simple speed controller. Ugly fruits make great dried fruit snacks.
Software may become involved. i'm playing with ideas for adding temp / humidity sensors and putting the fan's speed under the control of something that switches it based on time and those sensor readings. The in/out vent baffles could be actuated by servos. It completely lacks blinkenlights, which deficiency keeps me from showing it off more.
Cobuilding and comaintaining the whole family unit is a major accomplishment that we do not sufficiently value in the tech community, especially in the VFX & video game industries. Sure you raised some VC or had an exit or launched a game or got a paper accepted somewhere or have a popular open source project, but did you completely sacrifice having a family to do it?
I came here to say the same. And let me just say, that with special needs children, it truly is the hardest thing I've ever done. I remember older people saying that sort of thing before I was a parent and I was SO DISMISSIVE. Like, "oh yeah sure, you put you heart into it, it's important to you," whatever. But holy fuck. It's actually really really hard. Love is NOT enough. Strategy is NOT enough. Patience is NOT enough. It takes everything you've got.
Even though lots of people are parents, being a good parent requires a huge amount of work before, during, and after their childhood. We should celebrate people who actively give their children a good household and life. Your life is going to be pretty turned upside by having children whether you invest your energy in them or not - might as well make something great out of it.
I make them out of inexpensive wood scraps, so they end up looking rustic. Basically, they would be a prototype for someone with actual Woodworking skills.
They all include a substitution cipher message and code wheel. They are usually single movement puzzles. The "Sweet as an Onion" cryptex required multiple positions to solve, and the recipient was never able to solve it, so I lean into the simple designs unless my friend is an expert puzzle solver.
I usually go for hidden pocket puzzles, but my current favorite design is a variant of the snake cube puzzle that assembles into a functional code wheel to solve the message woodburned into the blocks such that if assembled in a second configuration reveal an encrypted message and instead of letters, I used Cistercian numerals.
Each puzzle also tries to embrace the onion philosophy, where each layer is easy to peel, and leads to making the puzzle easier to solve. Clues are usually hidden on the object to eventually guide solvers to solving it instead of trying to frustrate or defeat them.
Nice. I did a lenticular thing too. A couple years ago. A 4x6 inch 6 frame art thing. The printer was in Los Angeles and shortly thereafter he stopped doing business. Funny how hard it is to find a printer of such.
My father runs a car wash and I built a vacuum cleaner that accepted coins using coin acceptor from Aliexpress and Arduino. That was fun but it lasted only a few months until the device got too dirty from open weather to recognize coins properly.
I've looked at the stuff that gets built into car washes and that's a seriously hostile environment, but the weather by itself can also be pretty nasty. Quickest way to demolish a house without any effort: open the windows and doors, maybe a small hole in the roof. The weather will do the rest.
I learned the basics of electrics and rewired my house. While doing that i also learned how to "make" network cables and everything. That in turn has lead to some cool house automation for lights pool and more and now i can make my own homekit adapters too =)
The viewfinder isn't mounted in that picture. The two knobs on the right are from the lens to control the speed (125, 250, 500) and the aperture. The knob on the left is the shutter trigger. Shutter cocking is done by removing the ground glass from the back (which is how you preview focusing). Focusing is done by rotating the front side which is actually a giant hollow screw. The indentations are finger grips for rotating.
I have details for all my 3D printed cameras, unfortunately they're under my real name.
A titanium road bike frame. UBI in Ashland OR USA had a 2 week course during which you learn to and complete a frame. You didn't need prior experience but I took a welding course in a community college beforehand.
Very cool. I wanted a custom frame for my hard to fit wife. The bike shop (dragonfly bike shop in Davis CA) offered a custom frame or I could build one myself for the same price.
I ended up taking the semester class that met twice a week for an evening. Had an overview of bike design, light overview of drafting, and used a 1:1 frame drawing on paper. Handy to be able to hold up the frame you are working on up to the paper. It was fun to go from a pile of bike tubes to a usable bike frame. The meeting of 4 tubes at the bottom bracket had a surprisingly complicated interface. I'm glad I sprung for the powder coating at some bike frame aware place, maybe Cycle Art?
This one is physical and digital. I personally patented, designed, built, and deployed a novel ultrasonic inspection device that encoded position/orientation using IR (Just like motion capture in movies). Actually used it in the field at a nuclear plant. The value pitch is that is allowed for fast deployment (manual scanning), but could also record coverage ensuring nothing was missed. Fast deployment was important to avoid radiation and guaranteing coverage was important for safety. Ended up costing significantly less than automated scanning using a robotic system.
My absolute favorite was using the data to project a 3d model of the interior of the welds. Had to learn quaternions which was nearly the death of me, but I'm just glad it worked.
They still have the link apparently. The system used traditional manual scanning system with a VGA capture card on the back. I synced that with an off the shelf IR capture system and was able to record the ultrasonic scan into a video file using opencv. I would also record a USB cam at the same time as a sanity check. I could generate point clouds of data by reading the video file and the synced position data to project where the ultrasound was going and the linear response of the scanner. Was a ton of fun to develop and it was nice that I pretty much only had myself to blame when things went wrong. I also had to write the procedure for operation in case someone else had to use it.
That sounds really fascinating, could you post the patent no. or a link to more information? I bought a cheap NDT ultrasonic 'scope' to play with recently, to see if I can potentially measure specific gravity via it.
This correctly listed the parents. Hard part was linking all the systems together and keeping sync. Did it all in C++ way back when I actually programmed for work. Now I am a PM and code on the side for fun. I'm almost glad I cannot go back and see my code. It produced good data but damn was I a novice.
I do a LOT of custom stuff (mechanics, woodworking, house rebuilding, electronics, ...) but this is my favorite one yet, kind of sad I'm bad at showing the stuff I make :D
If you have any idea of what kind of job I could do to make use of all these skills I would be happy to have your thoughts !
I think it would be cool if you made a "player xylophone". I love the tonal sounds of wood and have thought to do the same. Similarly a plucking piano with bass strings that you play like a piano but sounds like a bass guitar.
Note that they're doing a rebuild of an existing house, but the general idea is the same - live in a trailer/old house while you build a garage with an apartment above it, and then move into that apartment and build the main house. This is quite common in more rural areas, such that you can see it even 40 years later; the garage will have an apartment above it, or the barn will have some rooms.
It's a very rewarding hobby, from making the mirror (manual labor + measurements with interferometry), silvering it (a bit of chemistry), designing in CAD and building the actual usable telescope, there are a lot of moving parts to design, and steps that you can either do or delegate. I think anyone who likes manual work can find a piece of a telescope that's enjoyable to work on.
The feeling you get after a few nights of hardware debugging, when everything properly works and you get a sharp focus in a clear night, that's quite a step above any of my other hobby projects to this day (:
Plus, you can always improve it with ergonomic modifications or additions to make observing sessions smoother.
Edit : here's a pic at first light of my last one (8" f/3.52)
I designed and made a 'cheapscope' out of PVC pipe, 3d printed parts, and cheap aliexpress 76mm mirror, mounted on a camera tripod.
It works well enough for the moon etc, and, as I'm based in a developing country, many people have never seen the sky except with the naked eye, so the gasps and oohs, etc are priceless.
I had intended to document and open-source the whole thing, but was busy at the time, and the PLA prints have since completely disintegrated in the hot humidity, lol. Maybe I'll get back to it again, with PET or nylon this time.
That's very nice. All DIY efforts to make visual astronomy more accessible are more than welcome as consumer products are priced for the global market.
That's my ultimate goal, an itinerant astronomy car to go in villages where there are no astronomy clubs. I also started teaching the few things I know about mirror making to my club and whoever asks.
I'm building public-compatible stuff now that I have my own stuff that is quite delicate to handle, after a few public nights where children grab onto a high-end eyepiece...
The 16.5" will be a very rugged scope to fulfill this purpose. I think John Dobson's approach to telescopes is the ultimate one, that is, the value of a scope is the number of people who looked through it
Instead of an eyepiece, i strongly recommend having a DSLR or other camera attached and showing the display on a nice screen. It makes it soooo much easier (although, it doesn't work great for nebulas or other faint objects). I've had a parent sort of utter "wait... we're looking at the moons... of jupiter... in real.... time?"
I've got an unfinished 8-inch mirror that I spent a month of evenings grinding... Someday I'll get around to polishing and building out a cargo-bike-collapsable structure for it.
A cool thing about DIY telescopes: You can get better results than a commercial mirror, because you can turn your time into quality at a rate which is not economically feasible for manufacturing.
Telescopes are often built simply (and rightly) as functional tools or instruments, and that's great of course, but I wanted to compliment you on making a non-ugly, aesthetically pleasing telescope; a bit of craftsmanship goes a long way, looks great man.
I once made a "therapeutic bartender" machine in a rapid prototyping class. It showed different drinks on the screen and used computer vision to estimate your happiness with each option (how big did you smile?). It then pumped and stirred a drink from a cooler full of ingredients.
A fence on a ridge. High tensile wire, pickets and some other fancy stuff. All hills and transitions. 1 pedestrian gate and 1 wideish chainlink gate. Wood posts and tposts. Box braces and such.
* high variance in teacher quality (we had such a bad French teacher that after 9 years of lessons, I knew less than those who had 4 years at a public school)
* long hours
* since the school didn't believe in failing students, everybody was kept in class, and the teachers had to do lots of repetitions to accommodate unmotivated students who remembered nothing from the past two years of classes. Very boring and demotivating for others.
Speaking of guitars, I am currently restoring a Korean superstrat I have since I am a teenager that seriously needed love. I am learning a ton of things like fret work, electronics and light wood work.
I have a Vintage V62 ICON that I modded heavily. All new electronics (including pickups), replaced the bridge with a Babicz full contact, and I had to do a lot of work on the neck, especially fret sanding, aligning and polishing.
It's really cool to see your work improve an instrument. That's now my favourite guitar to play. Pretty low action with zero buzz, pickups can give me everything from full dynamics and great range, to easily distortable power solos.
I'm now in the process of building a new guitar with a 3D printed body. I bought an unfinished neck, so I'm also learning about applying lacquer which is interesting.
A battery-powered sky camera I put together using a raspberry pi pico that connects over wifi to send photos every hour. Ended up going crazy and making a custom PCB for it. Need to do more work on the battery life + camera exposure levels, but overall I'm pretty happy with it, especially how reliable it is given my minimal C skills.
693 comments
[ 4.2 ms ] story [ 365 ms ] threadSoftware may become involved. i'm playing with ideas for adding temp / humidity sensors and putting the fan's speed under the control of something that switches it based on time and those sensor readings. The in/out vent baffles could be actuated by servos. It completely lacks blinkenlights, which deficiency keeps me from showing it off more.
It's just not that particular one that's biblically famous.
https://youtu.be/o3SEndDnVXg
https://shkspr.mobi/blog/2020/06/building-a-minimum-viable-l...
First time I've ever built and sold something physical. Amazed at how many people liked them.
[1]: https://en.wikipedia.org/wiki/Fairchild_K-20
The viewfinder isn't mounted in that picture. The two knobs on the right are from the lens to control the speed (125, 250, 500) and the aperture. The knob on the left is the shutter trigger. Shutter cocking is done by removing the ground glass from the back (which is how you preview focusing). Focusing is done by rotating the front side which is actually a giant hollow screw. The indentations are finger grips for rotating.
I have details for all my 3D printed cameras, unfortunately they're under my real name.
https://youtu.be/te130KONPP4
I ended up taking the semester class that met twice a week for an evening. Had an overview of bike design, light overview of drafting, and used a 1:1 frame drawing on paper. Handy to be able to hold up the frame you are working on up to the paper. It was fun to go from a pile of bike tubes to a usable bike frame. The meeting of 4 tubes at the bottom bracket had a surprisingly complicated interface. I'm glad I sprung for the powder coating at some bike frame aware place, maybe Cycle Art?
Recommended if you get the chance.
My absolute favorite was using the data to project a 3d model of the interior of the welds. Had to learn quaternions which was nearly the death of me, but I'm just glad it worked.
Also agree quaternions are the worst. You should check out n-vectors. Solves most of the same problems and way easier to work with.
They still have the link apparently. The system used traditional manual scanning system with a VGA capture card on the back. I synced that with an off the shelf IR capture system and was able to record the ultrasonic scan into a video file using opencv. I would also record a USB cam at the same time as a sanity check. I could generate point clouds of data by reading the video file and the synced position data to project where the ultrasound was going and the linear response of the scanner. Was a ton of fun to develop and it was nice that I pretty much only had myself to blame when things went wrong. I also had to write the procedure for operation in case someone else had to use it.
This correctly listed the parents. Hard part was linking all the systems together and keeping sync. Did it all in C++ way back when I actually programmed for work. Now I am a PM and code on the side for fun. I'm almost glad I cannot go back and see my code. It produced good data but damn was I a novice.
I do a LOT of custom stuff (mechanics, woodworking, house rebuilding, electronics, ...) but this is my favorite one yet, kind of sad I'm bad at showing the stuff I make :D
If you have any idea of what kind of job I could do to make use of all these skills I would be happy to have your thoughts !
Here's a good YouTube channel with a father/son (experienced) team doing just that: https://www.youtube.com/watch?v=QiOwwhoiKug
Note that they're doing a rebuild of an existing house, but the general idea is the same - live in a trailer/old house while you build a garage with an apartment above it, and then move into that apartment and build the main house. This is quite common in more rural areas, such that you can see it even 40 years later; the garage will have an apartment above it, or the barn will have some rooms.
Also an e-ink display that shows my calendar (and home power usage):
https://www.stavros.io/posts/making-the-timeframe/
Also a LED stick that lets you add images to photos in the real world:
https://www.stavros.io/posts/behold-ledonardo/
And a toy bus that shows you when the next real bus is coming:
https://www.stavros.io/posts/bus-stop-bus/
A 3D-printer-etched PCB: https://www.stavros.io/posts/make-pcbs-at-home/
It's a very rewarding hobby, from making the mirror (manual labor + measurements with interferometry), silvering it (a bit of chemistry), designing in CAD and building the actual usable telescope, there are a lot of moving parts to design, and steps that you can either do or delegate. I think anyone who likes manual work can find a piece of a telescope that's enjoyable to work on.
The feeling you get after a few nights of hardware debugging, when everything properly works and you get a sharp focus in a clear night, that's quite a step above any of my other hobby projects to this day (:
Plus, you can always improve it with ergonomic modifications or additions to make observing sessions smoother.
Edit : here's a pic at first light of my last one (8" f/3.52)
https://camo.githubusercontent.com/5939ba170acf1e9d6462d8268...
Now moving on to do twins 16.5" and 16" mirrors and scopes with a friend who is joining me into this adventure :-)
It works well enough for the moon etc, and, as I'm based in a developing country, many people have never seen the sky except with the naked eye, so the gasps and oohs, etc are priceless.
I had intended to document and open-source the whole thing, but was busy at the time, and the PLA prints have since completely disintegrated in the hot humidity, lol. Maybe I'll get back to it again, with PET or nylon this time.
https://miscdotgeek.com/3d-printing-the-hadley-114mm-newtoni...
It's done now, and has produced some nice images, but I need to spend more time with it to get planetary images etc.
https://www.youtube.com/watch?v=XCrJ3NflOpE
I'm building public-compatible stuff now that I have my own stuff that is quite delicate to handle, after a few public nights where children grab onto a high-end eyepiece...
The 16.5" will be a very rugged scope to fulfill this purpose. I think John Dobson's approach to telescopes is the ultimate one, that is, the value of a scope is the number of people who looked through it
A cool thing about DIY telescopes: You can get better results than a commercial mirror, because you can turn your time into quality at a rate which is not economically feasible for manufacturing.
Made only from components locally available in Zambia.
http://fleen.org/fences
* a bowl made out of copper, with a fitting lid
* a three-legged stool
* a very fine pair of salad servers, plus a wooden letter opener
* my own apron (sewn on a mechanical sewing machine)
* a bowl made out of cherry tree
If you count perishable items, probably my semolina soufflee :-)
* long hours
* since the school didn't believe in failing students, everybody was kept in class, and the teachers had to do lots of repetitions to accommodate unmotivated students who remembered nothing from the past two years of classes. Very boring and demotivating for others.
https://originalfuzz.com/collections/peruvian-guitar-straps
It's really cool to see your work improve an instrument. That's now my favourite guitar to play. Pretty low action with zero buzz, pickups can give me everything from full dynamics and great range, to easily distortable power solos.
I'm now in the process of building a new guitar with a 3D printed body. I bought an unfinished neck, so I'm also learning about applying lacquer which is interesting.
https://foundrytechnologies.com/relay.php
About 50% of the work on this was software, to be clear.