I listened to management kvetch about how the technicians were overpaid. This in a factory that profits were in the millions of dollars per day. Get your MBA and get brainwashed, then export the jobs to China.
I've worked as a contractor at a chemicals plant doing unrelated maintenance. Day of the maintenance a Pipe needs closed off and the plant looses around £100000 each hour. All maintenance work for that sector isxplanned and choreographed to perfection, every risk foreseeable is planned for and expensive (almost never used) redundancies/workarounds paid for
I've been at a place like that too. $7m/day per train and we had 2 trains..when there was a shutdown (called a trip) it would take 24-36 hours to get back to full production.
It’s quietly muttered by many, it’s bad fab managers that yell it out. It’s like a power flick in a fab, as the turbo pumps slam back on shaking the building you hear everyone involved in the recovery sighing f::…
as someone who suffers from an incurable, painful disease, with no effective treatment, I yell swears all the time (not directed at individuals) . there have been studies (including one released from Harvard University) showing that swearing for pain relief is as effective as opioids.
The ASML field engineer only makes between 90 and 110 grand a year and has to travel all the time, often on short notice (but at least the long flights are business class). The job is really tough on your body, with long hours of standing and moving heavy equipment. And you have to do it all in a clean room environment. But on the plus side, you get to work with the latest and greatest technology.
i can understand how they accept the job, many times i've seen that being on the cutting edge or in a job position many would like, turns out this way. Game devs, tho not so cutting edge, have the same overwork/underpay issue
That's the same pay I had as a former field service engineer working on digital cutters made by the Swiss, mainly in the printing and packaging industry. It's not enough money and yes, I am totally ruined health wise because of it. Started when I was 23 and barely made it intermittently working to my 30s, late 30s now with disability due to many unfortunate events that were all in some way stemmed from my job.
I used to work for Hewlett-Packard analytical instruments and one of our customers had to pay a penalty of 25 thousand dollars per day if the report was not filed. We had a lot of pressure to get the machine back in line. I can only imagine the pressure the engineers at ASML suffer.
My work, we have a removed tertiary involvement in wafer production where we have different FSE's on site for our "tools"... there is pressure on them to get stuff back up. (I'm applying it) But they do what they can, get what parts they can when they can, then go home. Me, I oversee the entire building... if we have a total down, I don't have the buffer of being just a paid vendor. The FSE's don't feel the $250K/day loss like I do. Had a few of these days so far this year, let's just say I no longer have a functioning digestive track. While I'd never want to be one of these ASML FSE's, I'd really never want to be the fab manager in charge of that ASML tool... hats off to those guys.
hope u are doing better, noticed you were down last video, dont know what happened (if anything) but just wanted to let you know my (and many many more people) appreciation for your superb content, subtle humor and excellent information! :p
@@Asianometry It's ok to be sad from time to time. Thank you for sharing. I hope your tomorrow is a wonderful day. You make the day a little more bearable for so many people. And you fill the world with wonder. Thank you.
I would guess 99% of people have no idea how complex and coordinated (between many companies and countries) making chips is. Thank you for making some of us a bit more knowledgeable. Literally, a bit.
The overwhelming complexity and size of the body of knowledge always makes me think how fragile our modern world is. These chips are, more than anything, what makes our way of life possible. And the amount of expertise necessary to produce them is mind boggling. Losing just a small percent of these experts would grind it to a halt.
A shout out to all the Field Application Engineers (FAEs) who knows their stuff and are able to minimise downtime of such a valuable piece of engineering Marvel.
I worked customer support for warehousing equipment (big sorters, conveyors, etc. for some of the biggest distributon centers you can think of...) and they sure hate downtime. Was on a 14hr long call with one place once. We had to get support engineers from another company on the line, and there were VP-type folks from the company experiencing the downtime on the call. The support engineer said he'd have to escalate the issue to the other engineers on his team, and when asked how long that'd take, said "they guarantee a response within 5 days". I could practically hear the customer's executives' heads almost explode.
@@josgeerink1350 Being constantly exposed to mental stress, take its toll on the liver. You can end up with cirrhosis that'll permanently damage the liver and even cause death if not diagnosed on time. Stress and alcohol abuse do the same damage to the liver.
It ain't fun being a lead Engineer on support where your task is not complete unless it is 100% rectified, tested, commissioned and signed! Sure the money is great... but the pressure makes one feel life is getting shorter by the second. Glad I'm retired! LOL!
"These machines, especially at the leading edge, barely work, and I mean that in the best possible way." Hoo... I once worked in a photography lab and we had a film roll developer machine go bad on us. It's a simple machine that's basically a few electric motors, heaters, and a timing system. A technician came on site to replace the main controller motherboard, which had a small burn on its surface for an unknown reason. When the new board was in and turned on, we heard a small pop, and about 3 seconds later a massive bolt of lightning shot out the side of the machine and blew a hole several inches wide through the new main circuit board and scorched a ceiling support beam. The poor technician yanked out the main power cable to stop the fire and I thought he was going to faint. The guy was lamenting how a $5,000 board just went up in smoke, and I was just thankful the poor guy didn't have his hand burned off. We found out it was one little frayed control wire in the loader hatch, far away from the circuit board. Somehow it caused a short circuit deep in the machine. Given the amount of power going through these ASML machines, I can only imagine what happens when you have a catastrophic failure on a cutting edge machine that fills a room and "barely works" under normal operation. They must have a pretty strict "no frayed wires" policy.
Standards for ASML machines are *probably* several dozens of leagues higher than those for a film roll developer. Although, the highest tensions are probably not found in lithography machines but in ion implanters. Most of those I work with can go up to 250kV, I don't see why lithography machines would need to go anywhere near that.
A Scanner down creates a lot of havoc on the track side as well. It's a rare opportunity to get some work done. Pretty crazy how much info you got in here. It'd be interesting to see what you could say about track dispensing purity and health
Imagine something that makes not a single machine, but an entire cleanroom go down. Like a malfunction in the air ventilation system, or a fire alarm that gets triggered. Being a simple maintenance guy at a fab must be insane.
This hits home hard as a former field service engineer. I worked pretty much exclusively on a Swiss made digital cutter line and it was very much like this too, these machine were always the bottleneck and key equipment in all the businesses that had them, so when they went down it was most often a big deal for the customer, every minute was costing them and their own customer's money.
Heh I did my undergrad dissertation in Comp Sci on simulating and correcting optical aberrations in human eyes. Made some nice wavefront visualisations based on Zernike polynomials.
Oh my nightmares are back. Get all the way into the Fab and realise you forget something back at the office. Three hours later you back through and the customer is screaming.
Get in the fab , gown , speak to client people solve problem, talk to fab people leave fab , degown go back to desk have an call/email waiting saying they need you back in the fab.
I dunno anything except that 300 wafers/hour is astonishing, incredible, fantastic and amazing and all the other superlatives.... To me it appears that these machines are like the absolute peak of terrestrial technology...
@@luminousfractal420 Bet less than 35% of the US population missed Interstellar.... It's like mentioning the Titanic or something, we ALL watched it lol
As a principal data engineer working on the flow of data coming from these machines, I can say I learned a thing or two from your presentation. Great video and keep them coming.
I remember sitting next to an ASML employee and fellow Dutchman once on a flight from LA to Amsterdam. And even though he was candid about his workplace and I have a mechanical engineering background and an interest in computer hardware in general I didn't get much wiser than that he did something with software for his employer. This was not too long after the IP theft case in San Diego became news, so I could imagine that they were extra careful already back then.
Although drilling an oil well may not sound particularly high tech, it is. A dynamically positioned drill ship may be drilling in water 1500 meters deep with a hole 1500 meters or more below the sea floor. All this relies on many things working properly. When somethings breaks, like the systems that keep the ship on station, all hell breaks loose. It's a million dollars or more per day to operate the ship, and the ship may be 200 miles from shore. The technicians on board are the best of the best. They have to be.
Those dynamic positioning systems are so critical. There was a diving support vessel years ago that got caught in a storm and its DP system failed at the worst moment, began drifting while divers below were still tethered to the vessel. Led to one of them having their gas and hot water lines severed leaving him thousands of feet alone at the bottom of the sea with no air. Crazy story, he survived and was rescued for the record, nobody knows exactly how he survived so long without air. Name is Chris Lemons for anyone who wants to check the story out.
I'm in field service for semiconductor vacuum pumps. When customer calls saying an EUV tool is logged down for an unscheduled down that tool immediately jumps to the top of the priority list.
@@dougdimmadimsdale9571 Yikes, I hope not in a high cost-of-living state. And unpaid overtime!? I understand for a salaried office-worker but how is that even a thing for an FSE?
@@dougdimmadimsdale9571 Is this current? Because I know that last year they offered 80-85K for new grad production engineers in the northeast US. It's hard to believe that field service would be less than that.
Fault prediction reminded me of 2001: A Space Odyssey. An error in fault prediction was the thing that clued the astronauts in that the Hal 9000 had a mental health issue.
Imagine being the driver for the spare part. You get pulled over by the police and try to explain to them that every second they waste costs 30$. What a fever dream
They're simultaneously the most simple thing, a glorified relay, and the most complicated thing we've ever made because we put billions of them into tiny areas.
@@myne00 The computer is the greatest achievement of human knowledge and the most complicated single thing. Just think of how many different disciplines are required to make one, and to make it do something useful, quickly.
@@andrewallen9993 that's something someone not actually deeply familiar with computers would say. No single person will ever be able to completely comprehend all the hardware and software parts of a modern computer.
@@blink182bfsftw I am deeply familiar with computers having started with icl1900, ibm360 and hp2100. Steve Wozniak was however the last person in history to ever design the hardware and os of a computer by himself :)
Broke a boat load of 4" wafers at NMOS 4, Ed Bluestein.... too much coffee! Luckily at the 03 level, just added poly contacts.... not as bad as 07 after metallization. Used to run a Leitz wafer microscope, inspect 1000 wafers a night... I think they put meth in that coffee.
Ex-ASML employee here, actually the most pressure came from tsmc, once upon a time there is one machine with down time long as 3 days lol, no one knows how
@@nixic_ unexpected down time for three days and no one knows how(on CS team, in production), the longest down time I have experienced is in installation phase, unexpected down for about 3 weeks, it was pure pain
Y'all got any information on the overhead track systems. What the hell goes into scheduling and moving the foups into different process modules. Especially rerouting if a piece if equipment goes down
I'm a cnc machinist studying EE and was hoping to get a fab job. This is super interesting to me, you don't see very much on day to day operations in the fabs. Thanks!
Great video. This push for super efficiency and performance to recoup costs and maximise ROI reminds me of London Heathrow airport. A plane full of passengers takes off or lands every 30 seconds... until something happens (too much wind requiring greater safety distances, a single plane aborting landing and having to go round to rejoin the queue amongst many others). This single exception ricochets through all the flight schedules across European airports as the exception causes a slight delay which causes further delays through the day until the airports close at night and some attempt at recovery can be made overnight. As a seasoned flight passenger, the learned lesson of 'stay at an airport hotel overnight and get the first flight out' is a sure-fire success strategy! Perhaps Asianometry could make a video about the tuning of airport and airspace efficiencies?
In my younger days, I worked on CD measurement SEMs. The drag with those is there was usually just one per product line, so when it went down, the whole line was down. It felt like all eyes were upon me as soon as I entered the fab. I knew if the problem was in the electron gun, it would take an overnight pump down to reestablish vacuum. I used to think those were fairly complicated systems, but compared to modern litho, they seem simple.
Thanks Jon, This gives a whole new set of dimensions to one of my standard quips (when someone wants to jump into a new or untried/untested technology): "Can you fix it when it breaks?" [usually the answer is 'no.']. Thank God for ASML support!
Excellent overview of a little-observed support system and procedures. Great! Are you ever likely to do an analysis of how Philips has essentially disappeared from the forefront of electronics and other technologies, and how it contributed to the establishment and growth of other international companies?
All the technicians and engineers keeping our world running deserve maximum respect. Being up to the tsak and bearing all the responsibility plus hard work on their shoulders is mind bugling. You couldn't pay me enough to be in that position.
Wow, just read up on the practice. Never knew this was a thing. I wonder if there's a clash between the guai guai proponents and boring inspectors that insist that a food item has no place in a clean room environment.
Mostly watch these videos since I got interested, when I was doing research prior to buying the ASML stock. It has turned out to be one of my best investments, and the lithography process is pretty fascinating. I have a history in the medical industry, building machines that sort proteins. The sorting was done using monochromatic UV spectroscopy.
@@coincrazy3563 paying channel membership gives early access. Going back to 2000 every fab has been able to generate similar charts with a cd sem and overlay measurement system. The ASML onboard tests were similarly impressive for stage or lens testing.
@@tomarmadiyer2698 Exactly. BTW, I think the value for support of this channel is very good. It's mostly early access, so everyone eventually gets access to the content which is a nice way to do it.
Maybe costly while down, however they earn a lot of money when its working. Before customers had back up computers and only relied on one, IBM used to guarantee to get parts to the customer with in 2 hours if the part is in country and 4 hours if it was not a part that very rarely fails.
I worked in an automobile final assembly plant. Every minute a vehicle rolls off the line, so when we're down, every minute is one vehicle's worth of profit lost. Depending on the plant, that can be $2,000 or $20,000 or $50,000. So its interesting to me that one machine can capture the same value as one whole assembly plant.
I used to be involved with integrated circuit manufacturing. It's incredible how the technology has advanced in the last forty years. It seems like Star-trek to me now.
When will Focus come out on Audible, I wanted to download it right away but its not in the bookstore just yet. I definitely am interested in listening to this while and work and when driving through traffic. Thanks.
Yup that EUV takes a lot of energy to generate and is very harsh on optics exposed to it. The lasers tubes that drive it and optics would be consumable at the rate it fires in these machines.😮
With such costs per hour of downtime, I kind of would've expected a team being at the ready at all times to jump into "GO! GO! GO!" mode, working at pit stop speeds to change lens elements or whatever.
When an ASML machine goes down, I get paid well. I word in repair. Motion Control repair is a Great STEM field for those who are able to learn and be effective in troubleshooting. The process of overlay is referred to as Registration, where one layer is registered to the prior layers.
Another one of your videos I'll keep bookmarked to show the interns + new hires (also if you say "well-ee" vs "well-uh" the Zeiss guys will bully you lol)
Great Video as Always!.. Would it be possible for you to do a deep dive on how the chips are separated from the wafer and then processed into a finished chip? I've seen a lot of your videos and if I have missed that you have already done a video on that I apologize. Love Your Videos!
At the computer chip company I used to work at there were people I regarded as 'million dollar engineers'. They were so exceptional that an entire project/product depended on their ability to debug critical issues. But they were never paid a million dollars.
Downtime… is a horrible thing if you work in a plant that runs 24/7… Good production management has a method to include scheduled downtime in their plan… Unscheduled downtime is what they work hard to avoid…. Paying more for good quality machinery… pays for itself! Nice work Jon, you sound better today. 😃
these machines are so unique that their many modes of failures are unknown and unexplored. these machines also becomes obsolete rather quickly so they don't have time to mature and have all the kinks worked out. This will always be the perpetual problems with cutting edge technology. it takes decades or even centuries for technologies to mature to a point where you can just take it to a garage to fix it.
Offtopic: I remember how much of a "black magic" it was to produce plastic model kits in the 1980s (that is: the molds). You could watch a printed catalog and buy the kits in the store, but how those were made in the first place remained a complete mystery to the average customer. In some way to me this resembles the chip production of today. Thanks for the video.
I was field engineer servicing production printers. Never really liked making decision whether it is time to replace problematic part or try extend its life by servicing it. And stakes were far lower. I woukd not be able to handle pressure making decisions on this kind of machine.
This is quite different from when you are servicing a 30+ year-old device, the developers of which no longer exist and if something happens, there are only 2-3 maintenance personnel and those spare parts that are in the next room.
An interesting thought is that the iron ore processing hub I work at loses 1 million an hour of downtime which is prehistoric compared to this complexity of these machines
I was engineer on a line that cost a million dollars a day to be down. What a job that was. Talk about motivation to keep ahead of problems.
He sounds happier today❤
that is crazy.
I listened to management kvetch about how the technicians were overpaid. This in a factory that profits were in the millions of dollars per day. Get your MBA and get brainwashed, then export the jobs to China.
I've worked as a contractor at a chemicals plant doing unrelated maintenance. Day of the maintenance a Pipe needs closed off and the plant looses around £100000 each hour. All maintenance work for that sector isxplanned and choreographed to perfection, every risk foreseeable is planned for and expensive (almost never used) redundancies/workarounds paid for
I've been at a place like that too. $7m/day per train and we had 2 trains..when there was a shutdown (called a trip) it would take 24-36 hours to get back to full production.
"A wiki of sad times" is such a fantastic quote.
Agreed.
I liked the quote but couldn't understand it fully. Can you elaborate? 😊
@@RJ.Mangal It's a knowledge base filled with issues, so it's a wikipedia page, but filled with only problems. Hence, wiki of sad times.
Yeah homie is witty & funny in these masterclasses =P
I bet you when it goes down there's always that one guy that screams "fuuuuuuccck."
It’s quietly muttered by many, it’s bad fab managers that yell it out.
It’s like a power flick in a fab, as the turbo pumps slam back on shaking the building you hear everyone involved in the recovery sighing f::…
always a beaker 😂
Lol
as someone who suffers from an incurable, painful disease, with no effective treatment, I yell swears all the time (not directed at individuals) . there have been studies (including one released from Harvard University) showing that swearing for pain relief is as effective as opioids.
Yea you're mom
The ASML field engineer only makes between 90 and 110 grand a year and has to travel all the time, often on short notice (but at least the long flights are business class). The job is really tough on your body, with long hours of standing and moving heavy equipment. And you have to do it all in a clean room environment. But on the plus side, you get to work with the latest and greatest technology.
i can understand how they accept the job, many times i've seen that being on the cutting edge or in a job position many would like, turns out this way.
Game devs, tho not so cutting edge, have the same overwork/underpay issue
That’s horrible pay! Equipment engineers around here make up to twice that from what I’ve seen.
Now adjust for a country with policies that keep housing affordable.
@@myne00 ahahah holy shit 90 -110 grand is what they made 10 years ago, their wage hasn't changed at all, back then it was a REALLY good job eh
That's the same pay I had as a former field service engineer working on digital cutters made by the Swiss, mainly in the printing and packaging industry. It's not enough money and yes, I am totally ruined health wise because of it. Started when I was 23 and barely made it intermittently working to my 30s, late 30s now with disability due to many unfortunate events that were all in some way stemmed from my job.
“Right so the computer that makes computers for fixing other computers is broken”
😮😮😅
“But Have you tried turning it off and on again?”
Yes, very much broken like your cat that sleeps in its food bowl, rather than the bathroom sink.
I used to work for Hewlett-Packard analytical instruments and one of our customers had to pay a penalty of 25 thousand dollars per day if the report was not filed.
We had a lot of pressure to get the machine back in line. I can only imagine the pressure the engineers at ASML suffer.
My work, we have a removed tertiary involvement in wafer production where we have different FSE's on site for our "tools"... there is pressure on them to get stuff back up. (I'm applying it) But they do what they can, get what parts they can when they can, then go home. Me, I oversee the entire building... if we have a total down, I don't have the buffer of being just a paid vendor. The FSE's don't feel the $250K/day loss like I do. Had a few of these days so far this year, let's just say I no longer have a functioning digestive track. While I'd never want to be one of these ASML FSE's, I'd really never want to be the fab manager in charge of that ASML tool... hats off to those guys.
That's why HP charged 80 dollars an hour for our time.
45 or so years ago.
When you make die using a lens that's slightly out of focus you get fuzzy logic.
0/10 joke right here hahahaha
@@davidanalyst671 Nah, thats a great joke.
@@davidanalyst671 collapsed right into a 0
😵💫 Fuzzy logic! Fuzzy logic! 😵💫
hope u are doing better, noticed you were down last video, dont know what happened (if anything) but just wanted to let you know my (and many many more people) appreciation for your superb content, subtle humor and excellent information! :p
I think we all noticed and worried. He does sound more like his usual self.
I can't even watch the video right now because I have work tomorrow but I came here to say the same thing
This man is a treasure
I hope he is ok
I think he might of been just sick or tired
I am feeling a lot better. Yeah I’ve been a little sad lately. Thank you for asking
@@Asianometry
It's ok to be sad from time to time. Thank you for sharing.
I hope your tomorrow is a wonderful day. You make the day a little more bearable for so many people. And you fill the world with wonder.
Thank you.
I would guess 99% of people have no idea how complex and coordinated (between many companies and countries) making chips is. Thank you for making some of us a bit more knowledgeable. Literally, a bit.
Sadly though the high iq populations genetically capable of doing this complex endevaour are dying out due to their low TFR's
The overwhelming complexity and size of the body of knowledge always makes me think how fragile our modern world is. These chips are, more than anything, what makes our way of life possible. And the amount of expertise necessary to produce them is mind boggling. Losing just a small percent of these experts would grind it to a halt.
@@Norsilca to me, and perhaps Asianomitry might allow, it’s really beyond something that can be taught, or learned. It’s art.
The other 1% make chips
A shout out to all the Field Application Engineers (FAEs) who knows their stuff and are able to minimise downtime of such a valuable piece of engineering Marvel.
I worked customer support for warehousing equipment (big sorters, conveyors, etc. for some of the biggest distributon centers you can think of...) and they sure hate downtime. Was on a 14hr long call with one place once. We had to get support engineers from another company on the line, and there were VP-type folks from the company experiencing the downtime on the call. The support engineer said he'd have to escalate the issue to the other engineers on his team, and when asked how long that'd take, said "they guarantee a response within 5 days". I could practically hear the customer's executives' heads almost explode.
Well, that's what support contracts are for
It takes lots of livers. A lot of livers.
?
Production go Liver me Timbers.
@@josgeerink1350 Being constantly exposed to mental stress, take its toll on the liver. You can end up with cirrhosis that'll permanently damage the liver and even cause death if not diagnosed on time. Stress and alcohol abuse do the same damage to the liver.
It ain't fun being a lead Engineer on support where your task is not complete unless it is 100% rectified, tested, commissioned and signed! Sure the money is great... but the pressure makes one feel life is getting shorter by the second. Glad I'm retired! LOL!
The money is great?!
@@yaboifet9058 IT is the irony of it all, that in the end it ain't worth it. Life is to be lived and money has nothing to do with it!
"These machines, especially at the leading edge, barely work, and I mean that in the best possible way."
Hoo... I once worked in a photography lab and we had a film roll developer machine go bad on us. It's a simple machine that's basically a few electric motors, heaters, and a timing system. A technician came on site to replace the main controller motherboard, which had a small burn on its surface for an unknown reason. When the new board was in and turned on, we heard a small pop, and about 3 seconds later a massive bolt of lightning shot out the side of the machine and blew a hole several inches wide through the new main circuit board and scorched a ceiling support beam. The poor technician yanked out the main power cable to stop the fire and I thought he was going to faint. The guy was lamenting how a $5,000 board just went up in smoke, and I was just thankful the poor guy didn't have his hand burned off.
We found out it was one little frayed control wire in the loader hatch, far away from the circuit board. Somehow it caused a short circuit deep in the machine.
Given the amount of power going through these ASML machines, I can only imagine what happens when you have a catastrophic failure on a cutting edge machine that fills a room and "barely works" under normal operation. They must have a pretty strict "no frayed wires" policy.
Standards for ASML machines are *probably* several dozens of leagues higher than those for a film roll developer. Although, the highest tensions are probably not found in lithography machines but in ion implanters. Most of those I work with can go up to 250kV, I don't see why lithography machines would need to go anywhere near that.
A Scanner down creates a lot of havoc on the track side as well. It's a rare opportunity to get some work done. Pretty crazy how much info you got in here. It'd be interesting to see what you could say about track dispensing purity and health
The lithography machine's mom went down!
Nice
to Georgia
Beautiful. Simply beautiful. 😂
muscle guy online 😂
Imagine something that makes not a single machine, but an entire cleanroom go down. Like a malfunction in the air ventilation system, or a fire alarm that gets triggered.
Being a simple maintenance guy at a fab must be insane.
This hits home hard as a former field service engineer. I worked pretty much exclusively on a Swiss made digital cutter line and it was very much like this too, these machine were always the bottleneck and key equipment in all the businesses that had them, so when they went down it was most often a big deal for the customer, every minute was costing them and their own customer's money.
Heh I did my undergrad dissertation in Comp Sci on simulating and correcting optical aberrations in human eyes. Made some nice wavefront visualisations based on Zernike polynomials.
Oh my nightmares are back. Get all the way into the Fab and realise you forget something back at the office. Three hours later you back through and the customer is screaming.
Get in the fab , gown , speak to client people solve problem, talk to fab people leave fab , degown go back to desk have an call/email waiting saying they need you back in the fab.
@12:50 "Like a traffic jam on the 405" - that's the 5 freeway right at my exit in Tustin California 🙂
I dunno anything except that 300 wafers/hour is astonishing, incredible, fantastic and amazing and all the other superlatives....
To me it appears that these machines are like the absolute peak of terrestrial technology...
I first read the title as ASMR machine goes down and was very confused.
Lol, i want that video😅
What strikes me most is how many different frikkin parallels have been drawn with Interstellar!
and how many people think everyone sseen it when its locked to a network.
@@luminousfractal420 Bet less than 35% of the US population missed Interstellar.... It's like mentioning the Titanic or something, we ALL watched it lol
How so?
As a principal data engineer working on the flow of data coming from these machines, I can say I learned a thing or two from your presentation. Great video and keep them coming.
I remember sitting next to an ASML employee and fellow Dutchman once on a flight from LA to Amsterdam. And even though he was candid about his workplace and I have a mechanical engineering background and an interest in computer hardware in general I didn't get much wiser than that he did something with software for his employer. This was not too long after the IP theft case in San Diego became news, so I could imagine that they were extra careful already back then.
The first rule of ASML club… you don’t talk about ASML club.
If this is your first wafer… you have to fab.
If the foil sticker peels off the welle lens, the fab stops and you have an unscheduled down.
Although drilling an oil well may not sound particularly high tech, it is. A dynamically positioned drill ship may be drilling in water 1500 meters deep with a hole 1500 meters or more below the sea floor. All this relies on many things working properly. When somethings breaks, like the systems that keep the ship on station, all hell breaks loose. It's a million dollars or more per day to operate the ship, and the ship may be 200 miles from shore. The technicians on board are the best of the best. They have to be.
Bruce Willis used to do that, but they needed him in space and he had to save his daughter's boyfriend and he was the first person to die in space.
We dont care about that.
Those dynamic positioning systems are so critical. There was a diving support vessel years ago that got caught in a storm and its DP system failed at the worst moment, began drifting while divers below were still tethered to the vessel. Led to one of them having their gas and hot water lines severed leaving him thousands of feet alone at the bottom of the sea with no air. Crazy story, he survived and was rescued for the record, nobody knows exactly how he survived so long without air. Name is Chris Lemons for anyone who wants to check the story out.
I'm in field service for semiconductor vacuum pumps. When customer calls saying an EUV tool is logged down for an unscheduled down that tool immediately jumps to the top of the priority list.
Edwards or UlVac?
@@MillersMotors Edwards
The field engineers for these machines must be being paid in gold.
The sad thing is that they aren't. They are paid decently for hardware engineers, but nothing close to software.
70k/year state side with a bachelor, endless unpaid overtime
@@dougdimmadimsdale9571 Yikes, I hope not in a high cost-of-living state. And unpaid overtime!? I understand for a salaried office-worker but how is that even a thing for an FSE?
@@dougdimmadimsdale9571 Is this current? Because I know that last year they offered 80-85K for new grad production engineers in the northeast US. It's hard to believe that field service would be less than that.
@@apexaviourhigh cost and high taxes. but of course biden claims 6figs without a bachelor, yeah right maybe with 15years of experience
i work as a field service engineer and this video has a scary amount of information i wouldnt expect people to know
But do you feel that your work is finallyy being recognised and appreciated ?
My cousin works at the ford plant and at $40,000/min for downtime, they have a helicopter ready to fly to another plant to get parts if they need it.
Ford = Für Opa reicht das
That’s German: Good enough for grandpa ;)
money pouring down the drain and you get that customer support guy "welll in 48hours we can escalate it..." 😱🤯🤯🤯
That’s when you get your bosses boss to call their boss 😂
We only talk to line up our ideas and strategies then go to lunch 😂
Dude just pronounced hijink perfectly and still apologized for it lol
Fault prediction reminded me of 2001: A Space Odyssey. An error in fault prediction was the thing that clued the astronauts in that the Hal 9000 had a mental health issue.
what an incredible channel. take care of yourself, it's well-deserved.
Thanks for that closing statement! Keep up the great content! 👍
Imagine being the driver for the spare part. You get pulled over by the police and try to explain to them that every second they waste costs 30$. What a fever dream
It really boggles the mind how computers are so complicated
They're simultaneously the most simple thing, a glorified relay, and the most complicated thing we've ever made because we put billions of them into tiny areas.
@@myne00 The computer is the greatest achievement of human knowledge and the most complicated single thing. Just think of how many different disciplines are required to make one, and to make it do something useful, quickly.
They are actually very simple, just the same thing repeated many many times over and over.
@@andrewallen9993 that's something someone not actually deeply familiar with computers would say. No single person will ever be able to completely comprehend all the hardware and software parts of a modern computer.
@@blink182bfsftw I am deeply familiar with computers having started with icl1900, ibm360 and hp2100.
Steve Wozniak was however the last person in history to ever design the hardware and os of a computer by himself :)
You nailed Hijink, and another great upload. Thanks!
Now just wait till you hear about wafer scrap. Now that's the real kicker.
Broke a boat load of 4" wafers at NMOS 4, Ed Bluestein.... too much coffee! Luckily at the 03 level, just added poly contacts.... not as bad as 07 after metallization. Used to run a Leitz wafer microscope, inspect 1000 wafers a night... I think they put meth in that coffee.
Ex-ASML employee here, actually the most pressure came from tsmc, once upon a time there is one machine with down time long as 3 days lol, no one knows how
haha is that the record?? only three days? seems like a minor/moderate outage (I mean obviously expensive but three days is pretty good imo)
@@nixic_ unexpected down time for three days and no one knows how(on CS team, in production), the longest down time I have experienced is in installation phase, unexpected down for about 3 weeks, it was pure pain
Y'all got any information on the overhead track systems. What the hell goes into scheduling and moving the foups into different process modules. Especially rerouting if a piece if equipment goes down
AMHS systems are looped into automation like every other tool. Integrated as such, it's fully aware with which tools are running production.
It’s mind blowing that this stuff is even possible. Given how fast things are moving and how accurate things need to be.
12:52 Traffic Jam on southbound 405 at Tustin 😂 Nice find! 😂
Thank you for the insightful summary at the end, it is 101% true that such complex systems need constant care to keep running.
I'm a cnc machinist studying EE and was hoping to get a fab job. This is super interesting to me, you don't see very much on day to day operations in the fabs. Thanks!
Answered MANY questions. My compliments.
Great video. This push for super efficiency and performance to recoup costs and maximise ROI reminds me of London Heathrow airport. A plane full of passengers takes off or lands every 30 seconds... until something happens (too much wind requiring greater safety distances, a single plane aborting landing and having to go round to rejoin the queue amongst many others). This single exception ricochets through all the flight schedules across European airports as the exception causes a slight delay which causes further delays through the day until the airports close at night and some attempt at recovery can be made overnight. As a seasoned flight passenger, the learned lesson of 'stay at an airport hotel overnight and get the first flight out' is a sure-fire success strategy! Perhaps Asianometry could make a video about the tuning of airport and airspace efficiencies?
In my younger days, I worked on CD measurement SEMs. The drag with those is there was usually just one per product line, so when it went down, the whole line was down. It felt like all eyes were upon me as soon as I entered the fab. I knew if the problem was in the electron gun, it would take an overnight pump down to reestablish vacuum. I used to think those were fairly complicated systems, but compared to modern litho, they seem simple.
I never thought about this issue.
Thanks Jon, This gives a whole new set of dimensions to one of my standard quips (when someone wants to jump into a new or untried/untested technology): "Can you fix it when it breaks?" [usually the answer is 'no.']. Thank God for ASML support!
Amazing info!
The "High Tech" machines that I design are both cutting edge and yet 6,000,000 years behind ASML
Fixing a computer that fixes computers that fixes computers that produces computers...
Awsome video. Hope you elaborate one day about exactly the path the waffers run around the fab.
Excellent overview of a little-observed support system and procedures. Great! Are you ever likely to do an analysis of how Philips has essentially disappeared from the forefront of electronics and other technologies, and how it contributed to the establishment and growth of other international companies?
Ha answered a lot of questions that I never had.
All the technicians and engineers keeping our world running deserve maximum respect. Being up to the tsak and bearing all the responsibility plus hard work on their shoulders is mind bugling. You couldn't pay me enough to be in that position.
Just replaced the expired "guai guai"(乖乖)in the machine, no big deal. 😂
Wow, just read up on the practice. Never knew this was a thing. I wonder if there's a clash between the guai guai proponents and boring inspectors that insist that a food item has no place in a clean room environment.
Technology from Uranus ! Don’t put your head in there Big mike likes you !you might get a nek burn🍭
Mostly watch these videos since I got interested, when I was doing research prior to buying the ASML stock. It has turned out to be one of my best investments, and the lithography process is pretty fascinating. I have a history in the medical industry, building machines that sort proteins. The sorting was done using monochromatic UV spectroscopy.
Inside of machine at 11:01 reminds me of Giger's Alien art...
This really is fascinating.
That chart at 11:25 is glorious
*uploaded 4 mins ago*. wtf?
@@coincrazy3563 paying channel membership gives early access.
Going back to 2000 every fab has been able to generate similar charts with a cd sem and overlay measurement system. The ASML onboard tests were similarly impressive for stage or lens testing.
@@coincrazy3563 prolly a channel supporter perk
At least I hope this channel gets the fundies
@@tomarmadiyer2698 Exactly. BTW, I think the value for support of this channel is very good. It's mostly early access, so everyone eventually gets access to the content which is a nice way to do it.
Maybe costly while down, however they earn a lot of money when its working. Before customers had back up computers and only relied on one, IBM used to guarantee to get parts to the customer with in 2 hours if the part is in country and 4 hours if it was not a part that very rarely fails.
zernike polynomials jumpscare, working on those right now
I worked in an automobile final assembly plant. Every minute a vehicle rolls off the line, so when we're down, every minute is one vehicle's worth of profit lost. Depending on the plant, that can be $2,000 or $20,000 or $50,000. So its interesting to me that one machine can capture the same value as one whole assembly plant.
30 cheese burgers a second! that'a a lot of burgers....
Thanks!
Thanks 👍
Excellent video 😊
Would you consider doing a similar video on related test/test equipment? Sort/package test.
My uncle ran an automobile factory. His job was to keep the lines running. Not sure if the stress or chemicals took him down first.
Your content is excellent
Would be interesting to see each machine put into a digital twin dojo but assuming only ASML could do this
I used to be involved with integrated circuit manufacturing.
It's incredible how the technology has advanced in the last forty years.
It seems like Star-trek to me now.
Yet again, a perfect video for everyone
When will Focus come out on Audible, I wanted to download it right away but its not in the bookstore just yet. I definitely am interested in listening to this while and work and when driving through traffic. Thanks.
For some reason it makes me think of the old vacuum tube computers and the constant replacement of the vacuum tubes.
Yup that EUV takes a lot of energy to generate and is very harsh on optics exposed to it. The lasers tubes that drive it and optics would be consumable at the rate it fires in these machines.😮
With such costs per hour of downtime, I kind of would've expected a team being at the ready at all times to jump into "GO! GO! GO!" mode, working at pit stop speeds to change lens elements or whatever.
When an ASML machine goes down, I get paid well. I word in repair. Motion Control repair is a Great STEM field for those who are able to learn and be effective in troubleshooting.
The process of overlay is referred to as Registration, where one layer is registered to the prior layers.
Ai modeling is worthless for low volume complex scenarios. I don't see that changing anytime soon.
Highly chaotic systems are notoriously difficult to model. A small sample set chaotic system.... well... good luck with that.
Very informative!
Another one of your videos I'll keep bookmarked to show the interns + new hires (also if you say "well-ee" vs "well-uh" the Zeiss guys will bully you lol)
I can only imagine how they have to treat the rigging that they use to move all of this heavy stuff.
Great Video as Always!.. Would it be possible for you to do a deep dive on how the chips are separated from the wafer and then processed into a finished chip? I've seen a lot of your videos and if I have missed that you have already done a video on that I apologize. Love Your Videos!
In the meantime look into the A/T (assembly and test) process. You should find what you're looking for.
@@jurassicpark104 Cool.. I'll definitely look into that.. Thanks!
At the computer chip company I used to work at there were people I regarded as 'million dollar engineers'. They were so exceptional that an entire project/product depended on their ability to debug critical issues. But they were never paid a million dollars.
I just like the fact that on the thumbnail, the dude is using a Lenovo Thinkpad. Not a MacBook or something. 😂
i dig this channel :)
Sure looks lke a lot of things that can go wrong :) Brings back memories of working field 'engineer'.
Its like when the starter is broke on your Nissan i understand the pain.
Downtime… is a horrible thing if you work in a plant that runs 24/7…
Good production management has a method to include scheduled downtime in their plan…
Unscheduled downtime is what they work hard to avoid….
Paying more for good quality machinery… pays for itself!
Nice work Jon, you sound better today.
😃
these machines are so unique that their many modes of failures are unknown and unexplored. these machines also becomes obsolete rather quickly so they don't have time to mature and have all the kinks worked out. This will always be the perpetual problems with cutting edge technology. it takes decades or even centuries for technologies to mature to a point where you can just take it to a garage to fix it.
He sounds better today
Offtopic: I remember how much of a "black magic" it was to produce plastic model kits in the 1980s (that is: the molds). You could watch a printed catalog and buy the kits in the store, but how those were made in the first place remained a complete mystery to the average customer. In some way to me this resembles the chip production of today. Thanks for the video.
"A wiki of sad times." We have one of those already: Wikipedia. The only encyclopedia you are not allowed to use for reference material in school.
Those citations are a usually-quick-enough check for digital media
Hardbound still takes some legwork
just site the sources listed on the Wikipedia page 😁
@@freddy4603
Verify that shiz first
48 hours?? Imagine having to lose $5,000,000 to downtime before you can be escalated to level 2 support
peace be upon you sir
I was field engineer servicing production printers. Never really liked making decision whether it is time to replace problematic part or try extend its life by servicing it. And stakes were far lower. I woukd not be able to handle pressure making decisions on this kind of machine.
This is quite different from when you are servicing a 30+ year-old device, the developers of which no longer exist and if something happens, there are only 2-3 maintenance personnel and those spare parts that are in the next room.
You pronounced’Hijink exactly right! 👌
An interesting thought is that the iron ore processing hub I work at loses 1 million an hour of downtime which is prehistoric compared to this complexity of these machines