TSMC Nanke 14 Factory Production Interruption Could Affect NVIDIA and Others

Zarathustra[H] said:
It's almost difficult to believe they don't test their chemicals for identity and purity before use.

In biopharma itis a regulatory requirement to as part of incoming inspection to do identity and purity testing in a QC lab using some sort of HPLC or LCMS method.

Sure these things cost money, but they are WAAAAAAY cheaper than losing several lots of silicon.
Click to expand...
But if they got say 200 barrels of some chemical, do they test each barrel or just 1, or does the company that supplies it do a test on the batch and then submit those results. Most metallurgical companies will do their testing on a sample from each batch then submit the results of those tests on the batch that way you know the purity of any compounds you are purchasing before hand. Granted after this they are most certainly going to investigate this and they better implement stricter testing on their supply chain before use. So with out knowing how they were doing their material QC it is all speculation, but I know there were similar incidents happened to Rio-Tinto last year when one of their plants that supplied a lot to China suddenly had far more silicon dioxide in it than the batch tests indicated which led to a very poor quality yield and set them back months, while the Chinese state used that to push some over to their own state owned plant that was able to meet their demand. Same story with a Japanese Steel manufacturer that was supplying China with some finished steel a year or 2 ago and again it was used to push a campaign getting local manufacturers to buy from a local state owned steel mill while giving a competitor a black eye. I just think the timing is a tad sketchy...
 
Lakados said:
But if they got say 200 barrels of some chemical, do they test each barrel or just 1, or does the company that supplies it do a test on the batch and then submit those results. Most metallurgical companies will do their testing on a sample from each batch then submit the results of those tests on the batch that way you know the purity of any compounds you are purchasing before hand. Granted after this they are most certainly going to investigate this and they better implement stricter testing on their supply chain before use. So with out knowing how they were doing their material QC it is all speculation, but I know there were similar incidents happened to Rio-Tinto last year when one of their plants that supplied a lot to China suddenly had far more silicon dioxide in it than the batch tests indicated which led to a very poor quality yield and set them back months, while the Chinese state used that to push some over to their own state owned plant that was able to meet their demand. Same story with a Japanese Steel manufacturer that was supplying China with some finished steel a year or 2 ago and again it was used to push a campaign getting local manufacturers to buy from a local state owned steel mill while giving a competitor a black eye. I just think the timing is a tad sketchy...
Click to expand...


I can't speak for every industry, but in biopharma (I work in medical device, but I did a "drug-device combination device" project once so I had to learn some of this stuff) there is a requirement that identification and purity are tested every time custody changes, so you cannot rely on vendors test results and Certificates of Conformance.

As for what sample strategy you use, I can't remember the details there. It's definitely somewhere in 21CFR Part 210 or 211 or the ICH harmonized guidelines. The systems I have set up test every delivery of ingredients, and if that delivery contains more than one lot of ingredients, it also tests every lot. Then there is a sampling strategy in place for how to handle multiple containers. You obviously are not going to test every container if there are thousands of them, but there is a statistical sampling method that instructs how to sample them.
 
Zinn said:
RTX 2080 is 12nm and this shutdown appears to only affect 14nm, so wouldn't this just affect Tegra chips (eg. Nvidia Shield or Nintendo Switch) ???
Click to expand...
I did some digging because you kept me thinking, ExtremeTech "but this problem is supposed to be specific to the 14/16nm product line." So, it appears you were correct. My bad.
 
dangerouseddy said:
you'd think with critical chemical components they'd test each batch and make sure its up to spec.
Click to expand...
This...

How can a state of the art, multi-billion dollar foundry not test every step of its process? One speck of dust is the end of a die, sneeze and that a whole wafer gone... How can there be no checks on the chemicals they use?

And to those talking about nVidia RTX failures... I thought that was traced down to being Micron DRAM?
 
Stimpy88 said:
This...

How can a state of the art, multi-billion dollar foundry not test every step of its process? One speck of dust is the end of a die, sneeze and that a whole wafer gone... How can there be no checks on the chemicals they use?

And to those talking about nVidia RTX failures... I thought that was traced down to being Micron DRAM?
Click to expand...

Check a few posts up, I explained it as I lived it.
 
Not every manufacturing facility "tests" things to the same level. I worked at a brick manufacturing plant where impurities in the mud, dirt used, dyes etc could lead to bricks that would fail in a few years rather than 25+. Sucks for those homeowners in a few years time, and TBH is much more costly to fix than a video card.

The sampling for the finished product was 2-3 samples per batch, but the chemical testing prior to making was time consuming when compared to keeping the line going to keep with the housing boom. Bottom line is, I doubt every single chemical is tested enough to catch every scenario. Some industries (such as food processing and medical processing are subject to strict and rigorous testing at almost every level), yet they still manage to let slip batches of food/medical products which require recalls. While I doubt this is the case for TSMC, the current situation's lack of a clear cause to the effect leads to believe that the cause is quite out of the ordinary.
 
bobzdar said:
Ran into this when I worked for a memory manufacturer - we had polonium contamination in our phosphoric acid causing occasional stray gamma rays that would damage thin oxides - so it'd wreck an occasional capacitor in our dram. It was such a tiny amount that it'd basically knock out one or two cells in ten thousand chips (which had 8 million cells each) - the worst kind of needle in a haystack you've ever tried to find. We finally traced it back by segregating the line and sending lots partially processed at our other locations except for certain steps on our line only - to where we finally narrowed it down to the wets area and then to the phosphoric acid steps early on in the process. Problem was you'd have to wait for the polonium to degrade so it'd take a few weeks for any cells to fail. I figured out that heat accelerated it a little (before we knew what caused it) so we'd test the wafers, then stick the lots in a high temp oven for a week and then re-measure. Cost us tens of millions and took almost 6 months to figure out, but luckily for us only affected one production line and didn't affect graphics ram (due to the high speed nature of it, it didn't need as much capacitance). Even when we traced it back to the phosphoric acid, we couldn't figure out why it was a problem right away as we had the same supplier in all of our fabs. But, it turned out that for our US fab they had phosphorus mined in a different region than the EU and Asian fabs, and that was the difference. We're talking completely undectable amounts of contamination, though we did figure out later how to detect it.

It wouldn't surprise me that they're dealing with something similar, we're talking indetectable amounts of contamination that would only show up after long periods of time over thousands of chips, and at a very slightly elevated levels compared to background radiation- but enough for customers to see a slightly elevated failure rate. I won't go into everything we tried to isolate the issue, but one of them involved flying wafers on coss continental flights to expose them to higher levels of outer atmosphere radiation to see if it accelerated the issue, thinking possibly something about the way the chips were manufactured were more susceptible to background radiation. Turned out to be wrong, but we had the right idea in that radiation was causing it.
Click to expand...


I thought Polonium was primarily an alpha emitter?
 
This makes sense. They had to come up with a reason to slow down production, and Samsung already did the old "fire at the power plant" excuse. Using "bad chemicals" is the next one on the list.

Next? Well, "train derailment interrupted critical process and ruined many batches" is low down on the list. Can't say "my dog ate it", but you CAN say, "an endangered species is nesting in our factory and must be, delicately, relocated".
 
motomonkey said:
I thought Polonium was primarily an alpha emitter?
Click to expand...

Yes, you're right. We thought it was background radiation that we tried to accelerate on the planes thinking there may have been something about our buried strap oxide that was more susceptible to it. It turned out to be a radiation issue, just not what we originally thought.

Also turns out somebody patented a way to clean up polonium contamination in phosphoric acid - just dip silicon wafers in it - that literature search is what finally got us on the right track to solving it....

Point was it's not always something simple like testing your chemicals.
 
let me guess, next to be impacted by one thing or another will be DDR4 ... again
 
You must log in or register to reply here.
Back
Top