Comment by rossdavidh | Hacker News Reader

Comment by rossdavidh | Hacker News Reader

So, it's been a while since I worked in semiconductor manufacturing, but I know a little bit about this from personal experience. The community colleges, yes maybe; the universities, no not particularly.

Once upon a time every fab needed lots of engineers, because they were designing the manufacturing process (because each company had only one fab, or at least only one modern one). As they grew, they kept hiring engineers for each fab, mostly out of habit and custom. Then they put huge bureaucratic obstacles in the way of changing anything, because the manufacturing process was already designed (at another fab), and you didn't want every engineer who wanted to make his mark on things to screw it up.

The thing is that if you make a change in step 87 out of 450, and that change turns out to screw things up, you can have (450-87)/450 = 80% of your factory is full of dud semiconductors. It takes a few months for a wafer to get from the beginning to the end of the line, so that's a couple months of production from your multi-billion dollar fab, that you can throw in the trash. So mostly, they don't want the engineers to do any design, because there's way more downside than upside. Only at the R&D fab is anything like real design done.

Now, to run the (rather complex) equipment, you probably want a lot of community college graduates, although frankly a smart high-school graduate could do it well with a few days training. But the only reason they "require" so many engineers at production fabs, is that they don't realize it's not necessary. It's a bunch of people who worked for years to get an engineering degree, put in a place where they are very nearly forbidden to change anything.

If the multi-billion dollar fab is made, they will run it. If there are not enough college graduates to staff it, they will miraculously discover that smart high school graduates will do just as well, as long as there are engineers at the R&D fab.

It's also been a while since I stepped foot in a fab, but my impression is that for modern processes there are no people in the fab proper. It is fully automated, people sit outside the fab or they service machines in drop-down bays underneath the fab floor, accessible only through airlocks.

What the technicians are there for isn't to operate the robots...they're supposed to analyze the performance of the robots constantly and summarize anomalies and flag them to the next shift. So a work shift might involve metrology, statistics, and communication skills; I wouldn't expect that skill set from any but the most exceptional US high school graduates, and those folks are bound for the university, not a job with 12-hour days wearing clean room suits and limited bathroom breaks.

Some of the more fussy processes like EUV require regular cleaning of excess tin in the machine as well to keep yield up. afaik it's a trade secret as to how this process works, but I know one of the big delays in rolling out EUV is that the cleaning process was hard, and that lead to a lot of downtime on the line. People on both side of the Pacific were able to make chips with EUV for a while, but they just couldn't make money while making them. TSMC cracked that nut, but I haven't read any articles describing exactly what their magic was to get the uptime so high. However, I suspect it might involve a platoon of PhD-level experts that also work like an F1 pit crew.

Anyways, the days of carrying boxes of wafers from machine to machine and pressing buttons are long gone. Everything that can be automated has been, what's left is debugging and servicing robots, and searching for early signs of degradation in an otherwise fully automated machine.

So, the real value of a highly trained technician in this case is they can look at bumps in the data and go "huh..." and flag it to management so that you don't have 80% of the fab full of trash wafers. In this situation, you want someone who actually understand the physics of the process going on inside the machine, and not just someone who is following an SOP. Any analysis that could be easily automated probably already has been.

Of course, you need even more highly skilled people to set up the robots in the first place, and that process also consumes hundreds if not thousands of engineers, depending on your schedule, scale and the complexity of the machines. A lot of that can be pushed back onto vendors, but you still need an on-site team who actually understands the physics to get your process yields up to several nines and hold it there with almost no downtime.

There are still plenty of people in the fab. They're not usually running product around in a modern automated fab, but there are always equipment issues to work through. When you have fleets of multiple tens of tools, it's always an exercise to keep them matched and in-line with each other. The majority of modern cleanrooms are "ballroom" style, where the tools all sit out in the open. Each tool has a class 1 mini-environment inside which keeps the product clean. Outside the tools is a cleanroom, but it's not as clean as inside.

Remote access has helped out a lot, but the process engineering folks will still spend a bunch of time every week on the line. The equipment engineering folks will spend time every day on the line working on things.

True that, but if they didn't have enough engineering graduates, they wouldn't just not run the fab. They would suddenly discover that a few engineers at the R&D fab, plus some motivated and smart high school graduates at the production fab, can totally make things run. The fabs will not sit idle due to a lack of college graduates.

A lot of what needs to be done definitely is not rocket science, and much of it can be trained -- at least when things are going well.

One thing I do appreciate is that there is a pretty good meritocracy at the second tier fabs working on legacy nodes. I know a fair number of folks who have worked their way up into upper management with only an associates or bachelors degree.

Yes, it's mostly the continual equipment maintenance and the occasional process engineer that goes actually looking at the equipment (mostly in drop bays) as wafers go by in boxes. Most of those are associates degree positions, because you need to understand enough (cleanliness, electronics, a little chemistry/optics, and some chemistry) to debug and replace components without breaking more stuff. Most of the equipment is manufactured, updated, and maintained/fixed by european, american, and japanese FAEs anyway.

That said, I do still hear that when there's an actual yield problem (or a delay in achieving yield), then there's a lot of overtime for the process engineers getting it up and running. Usually some engineer has to be on call all night in each area (photo, diff/dep, etch, implant), but in an office nearby.

Oh, and also there is all the process model extraction/control that has to be done for each fab/process. That's a whole group of serious engineers to create the PDKs.

Reply on news.ycombinator.com