Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Home Chip Fab

sam.zeloof.xyz

Post reply on HN
Print view View original

71–80 of 136 posts

Re: Home Chip Fab

#71

Post by bogomipz »

Junk_Collector wrote: post #40
bogomipz wrote: post #3

This is really fascinating. I apologize if this is a silly question but under the Fabrication section there is a pic titled "physical structure(Texas Instruments, 1997.)" My question is why do all of the components always look so "puffy" or "fuzzy" in these chip zoom-in pics? I guess I'm always surprised that such a precision process produces something with such imperfect looking shapes. Or are these slightly amorpho…

As mentioned in other posts, the process of making these types of devices tend to produce rounded structures. This is a byproduct of using Ion beams, deposition, lithography, etc in manufacture. BUT, the image is also made worse by how the image is captured. The picture you are looking at is a FIB (Focused Ion Beam) cut cross-section of a transistor which is then imaged by a SEM (scanning electron microscope). Probab…

>"As mentioned in other posts, the process of making these types of devices tend to produce rounded structures. This is a byproduct of using Ion beams, deposition, lithography, etc in manufacture."

Thanks for the detailed response. Might you have a link to one of those other posts or some other resources that talks about how these processes tend to produce rounded structures?

Re: Home Chip Fab

#72

Post by LargoLasskhyfv »

hollerith wrote: post #61
LargoLasskhyfv wrote: post #59

Earlier quoted context omitted.

Nikkei Article from December 2016 mentioning 0.8 micron, while aiming for 0.35 micron in 2018. So that's it, i guess? [1] https://asia.nikkei.com/Business/Biotechnology/Minimal-fab-t... According to wikipedia the following CPUs were built with that: [2] https://en.wikipedia.org/wiki/350_nanometer Which i think of as more than sufficient to finally being able to implement something like SCED, WAM, CHERI, Applecore, wh…

That's cool, but your [1] seems to say that the machine uses a wafer 12.5 mm in diameter, which is is probably less area than any of the CPUs in your [2].

Should suffice for my needs, see for example here

[1] https://en.wikipedia.org/wiki/R4200

under R4300i which mentions 45mm² for the die.

Also mentioned here [2] https://bits-chips.nl/artikel/small-series-of-chips-profitab...

are 0.25-micron to be released this year, with 190nm and smaller on the roadmap.

Which leads us to [3] https://en.wikipedia.org/wiki/250_nanometer at least.

Maybe not comparable in die size for all the chips mentioned there, i don't care so much, because i don't want to clone or emulate them. I want to go simpler. Rebranch from the 70ies so to speak, to take all the roads not taken since then. Just to see what's there :-)

Re: Home Chip Fab

#73

Post by bcaa7f3a8bbc »

faissaloo wrote: post #66
dymk wrote: post #51

Earlier quoted context omitted.

A highschooler don't end up with a garage full of equipment like that unless their parents are already in the industry.

Nope, Zeloof got most of the stuff on Craigslist and Ebay. The big stuff is usually pretty old, and alot is hacked together (e-beam stepper) or custom built entirely (sputtering chamber, or whatever the plasma machine is called I forget).

He also purchased a decommissioned, broken electron microscope from a university lab on eBay, and repaired it by himself.

Re: Home Chip Fab

#74

Post by a1369209993 »

hollerith wrote: post #61
LargoLasskhyfv wrote: post #59

Earlier quoted context omitted.

Nikkei Article from December 2016 mentioning 0.8 micron, while aiming for 0.35 micron in 2018. So that's it, i guess? [1] https://asia.nikkei.com/Business/Biotechnology/Minimal-fab-t... According to wikipedia the following CPUs were built with that: [2] https://en.wikipedia.org/wiki/350_nanometer Which i think of as more than sufficient to finally being able to implement something like SCED, WAM, CHERI, Applecore, wh…

That's cool, but your [1] seems to say that the machine uses a wafer 12.5 mm in diameter, which is is probably less area than any of the CPUs in your [2].

That's 122mm², which is larger than some Pentium II dies, so if you're okay with designing to a circular die[0], you could fit a Pentium II-grade CPU on each wafer.

0: The main reason to use a rectangular die is that they tesselate better to fit many dies per wafer.

Re: Home Chip Fab

#75

Post by m0zg »

segfaultbuserr wrote: post #62
m0zg wrote: post #60

Earlier quoted context omitted.

Speaking of which, does anyone know where (in the US) I can safely dispose of ferric chloride, sulfuric acid, and hydrochloric acid? I used to etch my own boards years ago. And ever since then the spent stuff just sits in glass jars in a double plastic box in my garage. I don't etch anything anymore, and it's taking up space.

> safely dispose of ferric chloride, sulfuric acid, and hydrochloric acid The first step is easy: neutralize them with any base, such as baking soda. You don't need precision, just add a lot of it, and make sure the pH is close to 7 (or beyond, baking soda is not corrosive) at the end. After this step, the solution is no longer acidic or corrosive and much safer to handle. Unfortunately, the next step is tricky. The…

Is this what people actually do? I mean I'm aware I could neutralize them, but Cu is still going to be in there and I still can't pour this stuff down the drain. Then there's ferric chloride which is just nasty stuff that will likely stain everything it comes in contact with. I just want to get rid of it in a fully environmentally safe way.

Re: Home Chip Fab

#76

Post by Kirby64 »

bogomipz wrote: post #17
Kirby64 wrote: post #10

Earlier quoted context omitted.

You're referring to the fact that the edges aren't really sharp and are kind of 'soft', right? Like, everything isn't 90 degree angles and is kinda rippley in places? If so, that's not an SEM issue, that's just what it looks like. Photo exposure by it's very nature doesn't edge things 100% properly, diffraction or tiny misalignments causes stuff to end up being 'soft'. Also, keep in mind that was from 1997. Fab tech…

>"You're referring to the fact that the edges aren't really sharp and are kind of 'soft', right? Like, everything isn't 90 degree angles and is kinda rippley in places?" Yes to both of these questions. "Soft" is probably a better way to articulate this, thanks. I'm not following the explanation of this being caused by photography however. Is this specific to photography at this scale? >"I imagine if you were to make…

By photography, I meant the photolithography, as you mentioned to another replier. This inherently will cause softness as described since there's always going to be light leakage and the masks you use will never be truly perfect. Given the sizes though, they're pretty damn good.

As for 'modern tech' vs. old tech, I'm referring to if you made the same size silicon structure on newer tech. When you get down to 7nm or whatever, the same problems show up (or even new ones) because the sizes get smaller. I wouldn't be surprised if the image from 1997 is 180nm or larger nodes.

Also, unlike the other replier, I don't think there's any meaningful diffraction (causing softness) by the actual SEM images... they don't use glass lenses like traditional cameras.

Re: Home Chip Fab

#77

Post by segfaultbuserr »

m0zg wrote: post #75
segfaultbuserr wrote: post #62

Earlier quoted context omitted.

> safely dispose of ferric chloride, sulfuric acid, and hydrochloric acid The first step is easy: neutralize them with any base, such as baking soda. You don't need precision, just add a lot of it, and make sure the pH is close to 7 (or beyond, baking soda is not corrosive) at the end. After this step, the solution is no longer acidic or corrosive and much safer to handle. Unfortunately, the next step is tricky. The…

Is this what people actually do? I mean I'm aware I could neutralize them, but Cu is still going to be in there and I still can't pour this stuff down the drain. Then there's ferric chloride which is just nasty stuff that will likely stain everything it comes in contact with. I just want to get rid of it in a fully environmentally safe way.

Don't ask me, I don't know, as I'm asking the same question. I'm waiting for some answers to my question: https://news.ycombinator.com/item?id=20659004

Re: Home Chip Fab

#78

Post by ChuckMcM »

LargoLasskhyfv wrote: post #64
ChuckMcM wrote: post #45

Earlier quoted context omitted.

That is exceptionally cool. They have taken existing best practices in semiconductor fabs and migrated them to an interesting place. What is missing of course is what their feature sizes are and the part at the end where you cut the die and package it. It is those things that would define the set of things you could put on a single chip.

Have a look at [1] https://bits-chips.nl/artikel/small-series-of-chips-profitab... under "E-Beam" for 0.25-micron node and roadmap.

Ok thanks for that. So 500 nm today, moving to 250 nm by the end of the year or early next year.

So your 12.5mm wafer can have a 8.8 x 8.8 mm square inside of it, or 78.125 mm^2. If I did the math right that is on the order of 156M transistors given a 4t ram cell that is about 39 million bits of RAM. So basically a pretty useful amount of space for "jelly bean" type applications. A synchronized fab line with a median processing time of 1 minute can produce 60 dice per hour. Assuming a physical plant cost of $8M US (that is "several million Euros + the office space to hold it) and a depreciation cycle of 12 years that is about $500 / day for the machinery we can add another $500 / day for staff + electricity, figuring 8 hour days, that's $125/hour to operate for 60 chips is a bit more than $2/dice.

Well the pencil math works (with all of those assumptions) but even assuming its off by an order of magnitude, $20/dice isn't a deal breaker for your own custom chip that does your special thing. You'll also notice that the 50 weeks a year 40 hours a week assumption. I'm guessing you can get better utilization than that which would offset your depreciation costs.

Re: Home Chip Fab

#79

Post by kragen »

segfaultbuserr wrote: post #70
asynchronous13 wrote: post #65

Earlier quoted context omitted.

pretty easy really. Add acid to water, then add baking soda to the acidic water until it's neutralized. (no more bubbling), then pour it down the sink with more water. Use appropriate precautions, gloves, goggles, well-ventilated area, etc. Once the acid is neutral, it won't hurt anything in the pipes.

But is it acceptable to dump wastewater with Cu+/Cl- ions inside? I've read that according to some regulations in some areas, they are considered pollutants and should not be dumped directly to the drainage. Or is it something you can just ignore, because at the end of the day everything goes to a wastewater treatment plant, and your volume/concentration is too low to be considered hazardous, and actually not more ha…

Cl ions are fine; dissolved table salt consists of them, plus some cations. The copper is more of a problem, and I don't know the official answer. I'd think that if you could oxidize it to copper monosulfide (covellite), cupric oxide (tenorite), or even fully hydrated copper carbonate (malachite) if you don't have acid rain, that would adequately protect it from weathering and thus allow you to dispose of it safely. But it might be more practical, as well as legally safer, to electrolytically reduce it back to copper and sell it for recycling. Copper, as a semi-precious and semi-noble metal, is recycled actively everywhere the humans live.

I think the biggest problems for chip fabrication waste (not circuit board etching) are HF and nonpolar organic solvents. I'd think neutralizing HF with chalk would yield fluorspar, which is resistant to weathering even over geological timescales. But again I don't know what the official answer is. Maybe dumping fluorspar in your yard will get you arrested.

Re: Home Chip Fab

#80

Post by 6nf »

SkyPuncher wrote: post #22
hammock wrote: post #20

Please don't dump your solvent down the drain like every Silicon Valley fab did back in the day, poisoning the entire region for decades to come. Just look up how many EPA Superfund sites are in the South Bay due to trichloroethylene, TCE contamination. Dispose of it properly and safely.

> Dispose of it properly and safely. I see a lot of people say this, but I rarely see any actionable advice. How does your average residential person find and dispose of chemicals like this? Seems like most people end up pouring them down the drain simply because they don't know how to actually find a better means of disposing them.

Your local rubbish service will point you in the right direction.
Post reply on HN