Gregory's ServerThis absolutely fascinating. A "lost" and possibly accidental method of making a solar panel 40 years earlier than previously believed and significantly more sustainable to build as well as totally recyclable.
https://solar.lowtechmagazine.com/2021/10/how-to-build-a-low-tech-solar-panel.html
kelbot ◖⎚∠⎚◗
31 comments
Billy Smith
This would be an interesting project to crowd-fund. Depending upon the location, i can arrange workshop space to do the making at London Hackspace. :D It would probably be a good idea to start with a small group to build a prototype first. :D 
Nelson Chu Pavlosky
@BillySmith @kelbot One aspect that stood out to me is "The melting point for Zn4Sb3 is 570 degrees Celsius, while it’s 1,400 degrees for silicon." Could a hobbyist reach these temperatures using just solar heating? Maybe just a huge magnifying glass? https://solar.lowtechmagazine.com/2011/07/solar-powered-factories.html
Billy Smith
We were able to do aluminium casting ~640-660C using a furnace that we made from a broken vacuum cleaner and a bucket with a hole in it. :D https://wiki.london.hackspace.org.uk/view/Project:GingeryFurnace Using a smaller 24V fan as the air input, with a renewably-charged battery, and sustainably-farmed charcoal, the whole system was carbon neutral :D So 570C for Zn4Sb3 is perfectly feasible. :D 
Alex@rtnVFRmedia Suffolk UK
@BillySmith @skyfaller @kelbot its worth a try - at very least if the cells work you should be able to get enough solar power to charge batteries that could run all the routers / networking / telecoms hardware in the hackspace, independently of UK Power Networks (good for resilience if there are power cuts)
Billy Smith
It would be interesting. :D I've never worked with German Silver before. :D I read how the recipe for the metal was nearly lost due to the anti-German feeling after WW2, but i've never played with it yet. :D Plus the press-fit design for the plug-caps could be pressed from sheet metal. The hardware and jigs to do that are really simple, and they've fallen out of patent decades ago. :D
Billy Smith
If they produce enough power, then attach them to meshnet nodes, and put them everywhere. :D
Billy Smith
That article reminded me of this OSHW project, https://web.archive.org/web/20160622174305/http://www.poc21.cc/solarose/ Using similar techniques to the version installed in Egypt in the previous article. 
Jasper
@skyfaller @BillySmith @kelbot Anyway, you'd use nichrome wire electric furnace, don't need to get all fancy while figuring out how to do this stuff..(which complicates safety too)
Billy Smith
It's definitely another topic. :D Captain Tangent strikes again. :D Also, I sent you links in the other thread of another approach towards solar reflector design. :D This is also why i have it written at the top of my profile: #adhd I get far too sidetracked by shiny ideas. :D
Billy Smith
Yes, using an electric furnace makes life a lot easier. The link to the Gingery Furnace, was meant to show that metal-casting could be done in a low-tech manner. :D The fact that we built that furnace and got it to work is a way of showing that lack of access to the tools can be worked around, using simpler methods, so that should never be a bottleneck. :D 
Andrew Cassidy
@skyfaller @kelbot sounds like it wouldn't be too hard, given you can get ahold of zinc antimonide
Andrew Cassidy
@skyfaller @kelbot though of course things might be too good to be true: it looks like Zinc Antimonide might be kinda toxic 
Arne Babenhauserheide
@skyfaller If it should work, that could be awesome, because the non-UV light could be reflected away to plants, using only the UV — and because some UV cuts through clouds. @kelbot
C24h29ClO4
1) yes it works. About as well as a water turbine on your showerhead. This video by an electrical engineer and pro solar YouTuber eevblog covers a system by Stanford that uses the same principle to "generate electricity at night" - https://youtu.be/Tdge8vEODeY - if you didn't try to bolt this to solar panels, it would 1/2
C24h29ClO4
@skyfaller @kelbot 2/2 work BETTER (and wouldn't destroy perfectly good solar panels like the stanford one) but I can't imagine you'd be able to generate anything close to regular PV cells. There are claims in the article that Zn4Sb3 also has some photovoltaic properties. I can't find a good source on that, but if true it MAY work better than a simple Thermoelectric generator. If it doesn't, I don't see this going anywhere outside of free energy scammers on YouTube.
C24h29ClO4
@skyfaller @kelbot if i was feeling snarky (I am), I would say that these 'solar panels' make about as much sense as running a website off of solar panels - theoretically a good idea, practically just a gimmick.
Nelson Chu Pavlosky
@c24h29clo4 We'll just have to disagree there. If my server is plugged directly into a solar panel, I can be confident of its energy use, embodied emissions, etc. If a data center claims to buy solar power from elsewhere, the situation is much murkier. I have to trust many entities who are often not transparent in an accountable way, and may not offer the info I want at all. Individual consumption is not meaningful to the climate crisis, but learning what to advocate for is.
Nelson Chu Pavlosky
@c24h29clo4 Beyond climate mitigation, resilience is another important reason to want to run your own server using an independent energy source. If you want to experiment with LoRa or mesh networking, to support your community even during extreme weather events or blackouts, you're going to want to be able to operate off-grid. It also presents the possibility of operating a mobile server that you can e.g. transport by bike to wherever it is needed.
C24h29ClO4
@skyfaller sure, and you'd be correct in thinking that way. My perspective is the actual webserver for a plain text website contributes very little to the actual power use when you consider the infrastructure required to route those packets to the device, and the device usage. Feel free to disregard that analogy and substitute your own appropriate "it technically works but it's not worth it under most circumstances".
C24h29ClO4
@skyfaller interestingly, running a microserver or similar power level device is an application that might make sense for Thermoelectric solar generators - if you don't have space constraints, prioritise environmental impact over cost, and can make one that'll last a decent time, I can even see it being a good option.
Mister Monster
@kelbot there's a write up somewhere on the internet that shows you the working principles and how to build one of these I read a year or so ago. They're pretty cool, I might build one one day
Andrew Cassidy
@kelbot yo that's so cool. I swear I've had the "could you make a metal-semiconductor solar panel" thought before
c+aue
@kelbot absolutely fascinating stands as a huge understatement.
i will have to read it more later on, as i see it a very long and dense read... but coming from someone who witnessed the boccara.org, this stuns me nearly as much as! 👽🥰 
Chartodon
Your chart is ready, and can be found here: https://www.solipsys.co.uk/Chartodon/108805143167206615.svg Things may have changed since I started compiling that, and some things may have been inaccessible. In particular, the very nature of the fediverse means some toots may never have made it to my instance, in which case I can't see them, and can't include them. The chart will eventually be deleted, so if you'd like to keep it, make sure you download a copy. |
@kelbot Important questions sadly unanswered by this article:
- Does this solar panel design actually work?
- Can hobbyists build one?
I really want to see a proof of concept and I hope that someone is inspired to make one.