How hard is it to DIY a Czochralski puller...

>silicon wafers
>ASML makes billions producing devices so Intel, AMD, ARM-licensees can make processors
What makes you think you can do this in your basement?

> i am too retarded to use a search engine.
Most people don't actively highlight how brain dead they are, go you!

it's gonna be a 3D FGPA integrated with risc-v.

i mean, i think i got a good sense on lithography and crucible handling. my main lack of knowledge is circuit theory itself, hence why i'd source that to Jow Forums since you guys are pretty knowledgeable on logical things.

don't forget to clean your silicon

solardaily.com/reports/For_Solar_Grade_Silicon_How_Many_Ns_Are_Enough_999.html

Attached: dsa.jpg (1080x818, 91K)

i won't, thanks friend.

>lens-based projector
How many lenses do you plan on using to fix abberation? Zeiss or Leica?

16 zeiss lenses.

oh and to clarify they're water immersed for resolution boosts due to the ior being higher and in turn the aperture.

also, damn nine 9s for electronics grade huh. i can't find much on the seimens purification process in terms of reference books.

i also looked into this although i have no idea how to rationalize the design. i may need a lot more documentation.

euvlitho.com/2017/P23.pdf

Attached: Capture.jpg (896x689, 65K)

i found a book on this if anyone else is interested.

booksdescr.org/item/index.php?md5=A45E9F0060FF090BDC6428C792F7FD63

Its a symptom if autism

please don't call me an autist that's very rude :>

it's pretty easy bro, all you need is a toaster and a long stick. You can get sand for free at the beach!

Wyf is this? I'm only here for the video games and waifus

the numerical aperture*

standard liquid immersion photolithography

Dude, EUV is light-years ahead of what you're going to be able to pull off in your garage. It's also unnecessary unless you plan on making chips with CPU level critical dimensions. EUV is dangerous as shit and you won't be able to buy a laser capable of it as a civilian. Even DUV is too complex/regulated/dangerous for a homegamer. You'd need reactants like Fluorine to feed the laser, and all sorts of other shit. No way is this doable at home.

You want i-line. Nice normal UV mercury lamps. It's the cheapest simplest way to go, and it is still some level of spendy and dangerous.
Look to older tech for inspiration. Don't draw inspiration from $7,000,000 top of the line Intel-class scanners. Even steppers are beyond homegamer level.

Look into mask aligners. They used fewer optics, wider spectrum UV, and were dirt simple to work on compared to any of the new machines.
Perkin Elmer mask aligners are old now, and you can even find em on eBay sometimes. They're still being used by lower tech semiconductor companies, they'll fit in your garage, and you won't need to learn how to write recipes for a newer machine (that alone will save months of trial and error).
You can even modify your mask aligner once you get comfortable with it. They're easier to maintain, too. You need a team of professionals to maintain a scanner, and at least one or two experienced people to maintain a stepper. A drunken idiot can maintain a mask aligner. You can find most of the documentation for them online if you look hard enough.

Just outsource it. Will cost a lot less and yield something you can actually use.

wow thanks a lot. yeah i realized after some reading that the EUV designs are insanely complex for dimension constraints, to where i couldn't grasp how they managed to distribute laser sources anywhere in the design, or how to house the optics properly. i'll try to look into documentation for perkin elmer mask aligners then, are they an all in one system or only for the projection component?

ah disregard i'm an idiot, it's an aligner. a CVD isn't anywhere near this phase it's just for projection.

ebay.com/itm/PERKIN-ELMER-MICRALIGN-300HT-MASK-ALIGNER/361377542615?hash=item5423c7b1d7:g:7NQAAOSwHnFV49Vo

like this? damn i didn't realize they went so cheap

> Not just using a Czochralski Pusher instead

Why do it the hard way?

...

you can buy wafers cheap as fuck

backdoored by the chinks.

>Implying chinks won't bug your shit

Doesn't this have like a 30% failure rate when done by the only two companies that have managed to do it at all? I mean, that's why when you're buying a quad core processor, it's usually a six core with two failures, right?

That was invented more than 100 years ago, with 5000 $ budget you should be able to create this buying parts from alibaba.

Nevermind, Chang has you covered for far less.

Attached: asfeaf.png (1007x661, 185K)

>i mean couldn't i use a wire saw then polish and use weak acids to remove possible artifacts?
no. everything you mentioned would destroy the surface; you would have to do mechanical planarization afterwards.
wafers are cheap, user.
just buy wafers from a company that spent the billion dollars on the machines.

that's because it's proprietary information.