Multi-reply: "cross-tighten with a torque wrench" - As of this post, I have 40 comments suggesting a torque wrench and approximately 50 recommending a star- or cross-tightening pattern. I hear you. In undergrad (maybe sophomore year?) I was helping to service a piece of vacuum equipment and was actually corrected FROM using the star pattern. I was taught that on these copper conflat flanges, "cutting in" the knife edge linearly by tightening in a circle was the way to go unless you had an absurdly large flange (which this is not). Trying to get a UHV seal with these gaskets is nontrivial, and if you have a system that works, you stick with it. I also know people who cross-tighten their conflat and get great results. In applications like this that are super-finicky, any change to procedure comes with risk. if it aint broke, don't fix it. - During this opening there were maybe 6 or 8 flanges open, and only one of them leaked, and the flange that DID leak had already been problematically leaky during previous openings, so I don't attribute the leak to a problem in tightening but more than likely, a scratch on the flange knife-edge that requires you to torque the gasket down past spec to fill the extra gap with copper. - Regarding the torque wrench, in many many locations on this system, it's almost impossible to fit a regular wrench around these flanges, let alone a necessarily bulkier torque wrench. This flange probably could have been handled by an open-end torque wrench, which I guess I assume exists but have never seen, but in general to work on these academic systems you need to develop a feel for it. Notice how I'm barely moving the wrench with each tighten. "you are wasting helium" - This is actually an extremely common and extraordinarily critical industrial use for low-purity helium. THIS is what the party balloons should be saved for. This method was developed to detect leaks so small they don't generate bubbles, noises, or other observable airflow. we're talking about a handful of atoms weaving their way through cracks in apparently solid pieces of metal. I've found a leak straight through a stainless tig weld on this system using helium. Amazingly enough, a LOT of the semiconductor industry depends on being able to create high quality vacuums for processing.
interesting that there's still stuff that's just based on what you were passed on and there's not one best way known (like the tightening pattern you use). makes it feel like a learned hard skill rather than just a purely academic pursuit
Ultimately conflats are not as finicky as you think. I've reused them on training vessels with a Extremely high vacuum to some success. Landing the groove in the old groove had to play a crucial part in this and the bite was pretty solid. What mattered the most is that we always followed torque specs and when those failed we upped the specs and still followed the cross pattern starting from marked bolt 1. I don't recommend this as a operating guideline, but I'm telling you that those rings are tougher than you think. Following good engineering practice and a little bit of elbow grease you can get a used copper conflat to hold for at least 40 hours in extreme vac. I've seen it done. It did fail, but we only had the one in the training lab so we waited for a batch of new ones to come in, but it will do it.
As someone also in the UHV side of semi, the law of the land where I am is star tighten after it's hand-snug, and always to designed torque. By not doing that you risk dis-leveling the mating surfaces and could end up overcutting a seal, deforming it more than necessary and then actually creating a weaker seal. This leak might have been avoided by proper torquing patterns, but like you said, typically that type of failure can be hidden by deforming the seal even more with additional torque. Just wait until you try baking out the system though, that additional deformation is likely to come back to haunt you.
@@sejongthegreat3044 Does the material expand at all due to tempature changing the torch specs or allowing enough of a gap to make a difference? I don't know why but It instantly popped in my head while reading this all.
Yo Destin - I love your channel! Glad you like the video! I’m not sure why leak-checking an MBE has been deemed viral material but my channel has had a wild week and a half and I’m glad you made it here!
I figured we were all getting recommended this around the same time; I know when I see a banger video topic like this pop up in the recommendations out of no where that I am one of many fine lads the algorithm chose to bless today
@@AlphaPhoenixChannel because the Titile is really intriguing i want to know what the different b/w air tight and vaccume tight. also that's how mysterious Algorithm works
“I could talk about lab equipment for hours and hours” I for one would love to listen about lab equipment for hours and hours. Even if you posted barely edited ramble style videos about niche and specific lab equipment, i’d definitely watch them
"the world is running out of helium!" "But what could you need it other than for making squeaky voices?" "Scientific purposes!" "Like what?" "Releasing it into the air so it makes everyone in the lab's voices all squeaky"
But like for real I’ve had a lot of comments saying this wastes precious helium but THIS is one of those critical industrial uses everybody always talks about…
@@AlphaPhoenixChannel No worries, just use hydrogen, the molecules are even smaller. What could possibly go wrong? I'm not sure, but we had a helium detector as part of the ion pump AFAIR. I'm also pretty sure the helium nozzle was a calibrated leak, not sure why. But the process was exactly the same back then, 1968. What was impressive was the way the helium alarm would trigger even while you were just setting things up the other side of the lab.
Ah Gordon! Ho grab that wrench over there! Good, now gently tighten the bolts... Gently.... Geeeeently... *Beeping starts* AHHHHH! Gordon what have you done!?
Im really glad Tom Scott encouraged us to watch this in his most recent newsletter, because its fascinating stuff; I never considered how much such a insignificant leak could multiply so much under a vacuum.
@@NoNameAtAll2 Yes, in case you haven't made it there already, it has a few suggested youtube videos and a few other internet articles/ findings and is weekly. Really interesting! (I'm going through a backlog of them currently which is why I'm watching it now, 3 months after the newsletter date)
@@NoNameAtAll2 if you google Tom Scott, his website was the first result for me and right on the front page is a link to sign up to it :) (Not keen to put link in comments in case spam filters catch me)
Previous semiconductor worker who delt with vacuum and pressure tight vessels and pipes. -Pressure tight is way harder, especially when the differential is 100psi or more, and silane, hydrofluoric acid, anhydrous ammonia...you know the things that dissolve humans or reliably spontaneous combust in air. With a vacuum you have chances to mess up, with pressurized death gases/liquids, you have 0 chances. Also helium and hydrogen pressure tight is the worst, but we usually used helium to test connections of pies to contain worse chemicals before putting them into production. That being said, randomness from the machines vibrating, and human error caused leaks and the fun beganafterwards.
I'm surprised you tightened the head bolts sequentially around the head rather than in a star pattern to avoid uneven distortion of the gasket. Any reason for this?
That is correct for initial tightening down to torque spec to insure even forces... but, not so much after proper specs have been reached... the "star pattern" (most commonly used) is to keep the pressures close to the same around the circumference and while increasing the torqued pressures to spec to maintain "flatness" or parallel surfaces. Micro-adjustments afterwards are a completely different story. The torque forces applied at this point are increasing in tiny fractions and are deforming the metal gasket like a "putty" at those pressures, Essentially, instead of applying the forces in the flat plane for maintaining parallelism, you actually want to theoretically "chase around" a minuscule blob of metal to fill in those gaps that are allowing molecules through.
@@edwardpaulsen1074 but can't you inadvertently still maintain the leak? By chasing/pinching the gasket seal around to a single point (between your start and end locations) can't you effectively create a material imbalance in that area maintaining the leak and making it harder to seal?
@@miamisasquatch You can inadvertently do pretty much anything as long as its not what you were planning on but sealing vacuum isn't that much of a challenge to be honest. Your car tires are probably like 40psi or something like that which is a 40 psi difference to atmospheric pressure and they seal just fine for the most part. Absolute vacuum is just -1atmosphere so from outside to inside of the vessel there is just a pressure difference of 14.7 psi. So don't worry about it just huck some copper in there and you're good.
Thanks for a very informative video! I spent a couple of years working as a technician at a scientific research facility and often enjoyed wandering through and marvelling at the beam pipes, examining some of the sensor and control equipment and was struck by the constant hum of the various vacuum pumps which had to run continuously to keep the pipes at a near-vacuum as a particle beam was injected.
I finally managed to get a He leak detector for the lab I work at. It is a Varian that was made in about 1985 and needed some work. I managed to get it up and running, and boy am I glad I got it. There is a fancy room that we put together and it took 2 guys 3 weeks to get it mostly leak free (I used that to help justify the leak detector). I later needed to do some modifications to the room which meant finding all of the leaks that were sprung, and it took me about 4hrs to be confident that the room no longer leaks (this included a few leaks that were missed by the previous guys). I put that in there in my status report email as a sort of "I told you so".
A poor man’s leak detector is using a very sensitive pressure sensor (which you will already have in an ultra high vacuum system) and just taking a squirt bottle and spraying IPA around the machine seals and bolts. If there is a leak the IPA will go inside the leak and when the alcohol hits vacuum it “explodes” and you see the pulse on the pressure sensor. I once watched a Russian physicist do this while his He detector was down and he figured out all the leaks without it and was able to achieve better than 10^-9 (it’s been awhile I forget the exact number but it was our personal record) lol
@@latenttweet that's what the first crew was doing. This is also how we leak check our systems with a convection gage. The problem with this system though is that it has a 200gal reactor on it to act as a huge buffer, but it also uses a very crappy air-referenced vacuum gauge. This system works from atmosphere down to 100Torr or so. The He leak detector worked a treat though, but it would take a while for all of the He to get sucked out of the system to allow the leak detector to calm down again.
@einstein9073I assume Roger meant something more along the lines of "Scientists like to do experiments with different shit in vacuum because its cool and cool shit happens" because it would be really stupid if he meant something more like "Scientists like to do experiments on a vacuum... because they are curious as to how nothing keeps doing that weird thing it's doing".
@@DrDeuteron While that's true, hydrogen can chemically diffuse through some metals, making it basically unstoppable. This is especially true at elevated temperatures.
@@bladdnun3016 Hydrogen permeates through all materials, especially steel. PTFE aka Teflon has the lowest permeation rate of any material available and hydrogen still permeates through it.
@@peterwill9660 I have seen helium permeate through .250" carbon steel plate. Drove me crazy trying to find the "leak" with a "sniffer" in my welds that I was certain were good.
I had the privilege of working on the upgrade of a fusion reactor in the UK. During my time on the project the reactor was being checked for leaks exactly like this. Totally fascinating engineering. Brilliant video :-)
Absolutely fascinating !!! I worked in a company that sold and serviced vacuum equipment.......but nothing quite on the order of what you work with. Regardless, although not part of my responsibilities, I had to learn about vacuum equipment to be able to discuss leakage (and other types of) problems with the engineers who were tasked with producing solutions for our repair people. What I DID learn was that although everyone thoroughly understands pressures above atmospheric pressure, very few people (including many engineers) have a real problem understanding/coping with problems concerning negative pressures. It seems that we're hard-wired to work with positive pressures. I had to sit with some engineers and use my semi-pro knowledge of negative pressures to get those engineers to "re-engineer" their thought processes. They all understood the principles theoretically but couldn't make the mental "leap" to practicality. It was certainly an ego boost for me to be able to see the "light bulb" come on over an engineer's head when my "teaching" finally sank in. Ah, the glory days........before I retired. What a great vid !!!
I would literally pay for content of this quality released on a once a month/nile-red-type basis also, touching on superlative topics, like whatever would be considered a catastrophic failure in this system, would be super interesting.
I used to work on an ultra high vacuum system. We regularly had to vent and open the system to replace test samples etc. I'd love to see a video on your method for pumping down the entire system.
@@zachmarshall6059 a fast entry chamber. A very small chamber, with a small volume, that can be easily isolated and opened. Takes from minutes to hours to pump perfectly a fast entry. Takes days to a week to pump and bake out perfectly a whole big ass UHV setup.
@@DocSineBell Maybe it wasn't that big a setup. Still, even for small stuff like detail electronics or pacemaker bombing etc. I've seen setups with cutoff valves. The trouble tends to be that even labyrinthing with 1- or 2-mp valves sunk into the build, you're still taking your bloody chances that the seal tightness is going to be sufficient.
I started research in a lab this semester using a gravimetric microbalance and mass spectrometers as well, the microbalance uses the same copper gaskets as you are, albeit at a much less perfect vacuum. It is so cool to see the stuff I use in different labs as well
I work with a few guys like you, who can just talk and talk and talk about the equipment they work with. Always fun to ask them questions! Really interesting video!
It was one of the best parts of looking for a new job when I last did it. I know enough about a lot of things in order to ask good questions that elicit a ramble about someone's work. Learn new things, get excited about the human endeavor, what's not to love!
Totally know the feels. Had a vacuum drop in my ToF-SIMS from 10^-11 to 10^-8, which was screwing with our cesium gun. We were up for maintenance anyway, so we went with the invasive option and just changed all the copper gaskets. Fortunately after putting it all back, there were no leaks!
How does ToF-SIMS compare to positron annihilation induced Auger spectroscopy? My PhD work is in space physics so I know little about this but my friend works with positron beams measuring what gets knocked off a target and sent down a ToF tube. Both are high vacuum work and seem super cool.
I like how real you are, you approach your videos as a person living in real life, rather than masquerading as something you're not like most people do
Well there is a huge difference between labs and manufacturing plants. An contamination causing an issue in less than 1% of all experiments might be neglected by many scientists, but at high quantities and low margins, 1% might be the difference between earning a lot of and losing a lot of money.
Thanks for bringing up my memories on working with UHV equipment on university. Memories of many, many bolts, some broken feedthroughs and unreal toughness of platinum wires. And waiting for right pressure to actually start the second and third stage pumps. Good memories :)
I used to work with high vacuum equipment myself--ironically, making parts for quadrupole mass spectrometers! Our vacuum requirements were a little less extreme than yours ("only" in the 10e-7 to 10e-8 torr range, so we could use rubber gaskets) but all the principles were the same. I may point people to this video in the future when I have to talk about my old job!
This is so cool! I’m currently doing my PhD in physics and my friend here works with positron annihilation Auger spectroscopy, so they use a UHV system that measures what comes off a target with a ToF tube. He was talking about the flanges and cleaning the system the other day but I never really truly understood the work that goes into maintaining it until now. I do simulations all day so this is all alien to me! Love all the replies from the scientists in this thread. We’re all just geeking out together and it feels great.
I worked on an STM during undergrad, and it's interesting seeing so many of the principles I learned in that lab being reflected here. Same copper gaskets, same residual gas analysis, same pumps. I mean, that's not surprising for obvious reasons, but it's still interesting to see.
@@AlphaPhoenixChannel We were studying the surface of Ag(111). To be honest, I only got a few hours actually running the STM and never got to do any analysis myself. My efforts were almost entirely focused on building the STM with my professor (and then taking it back apart because the university made him move it).
When I was working on my chemistry PhD, one of the profs had a ultra high vac system he used for is surface chemistry research. I had a class with him where we covered his to find a leak with He. 10^-12 Torr is tough to obtain. Breaking vacuum, cleaning and pumping down is certainly a lot more involved than with my GC-mass spec.
Spent the first year of my career in machining at a shop that made components like this for the semiconductor industry, it's nice to finally see someone showing how a knife edge seal works, the thing that was always so mind blowing to me was the hours we spent hand laying the finish in a certain direction on oring seal grooves so that we didn't create a leak path. Very cool video
I’d love to see how these are manufactured - I can attest that edge is fragile. This same spot had leaked more than once so I think it must have gotten scratched during assembly or a previous service
Unfortunately I no longer work at that shop, but the way we cut knife edges was to bolt the plate to a rotary axis and then to tilt that axis by the angle of the knife edge then rotate the whole plate around the center of the knife edge circle , it became a long day when we were doing flanges with 6+ gunports all splayed out at angles from centerline so they'd have clearance for the larger end of the gun
It's not "obvious" that just because someone works on something, that they find it fascinating. I've run across co-workers in my field of technology who found it an absolutely excruciating bore, but kept at it for the salary. I always felt really sad for those kinds of people. But all of that to say, consider yourself lucky that you're able to work on complex and intricate technology and ALSO find it endlessly fascinating. Cheers!
I worked in the mineral assay industry for many years, and mass spectrometers were pretty much the pinnacle of available instrumentation for the purpose of trace element analysis... for your machine, it's a leak detection accessory. Mind blown.
That’s the best part of scientific/medical terms, everything is categorized and defined the same way. So all the terms sound oversimplified specifically so you know exactly what it means even if you haven’t heard of the concept before. If only English could learn a thing from that.
I´ve actually walked by a clothes stand outside a clothing store where an ungodly amount of X´s were before the "L". 5 or maybe even 7 X´s, i believe. While i walked by it, i quietly said the size with all the "extra"s to myself under my breath and came to the conclusion that i think a new word should be invented for sizes beyond the second or third "X". Extra large, extra, extra large, extra, extra extra large, impressive extra impressive, extra, extra impressive momentous... gargantuan... royal... serious...
@@GarretRB Yep, this naming happened with radio bands as well. HF -> VHF -> UHF are high -> very high -> ultra high frequencies. It keeps going too. SHF - super high frequency, EHF - Extremely high frequency, THF - tremendously high frequency I'm a little mad it doesn't go Super, Ultra, Mega though.
Its even worse in Biology. They never discover the enzymes that perform sequential steps in a reaction in order, so you get used to struggling with "Enzyme IV coverts A to B, then enzyme II converts B to C, and finally, Enzyme VI coverts C to D; however, sometimes Enzyme I also converts some B to D".
I did vacuum testing for about 3 years. I was only testing at high vac but we also did thermal testing in our chamber. That brings in a whole different issues with out-gassing. We’d always having issues maintaining our required vacuum level when going to hot temperatures. We also did our leak checks with IPA wipes. It never really worked too well but tightening or replacing a gasket was fairly easy and worked every time. It also takes about forever to heat things in a vacuum when it’s only bolted to a plate.
I worked with a large scale multilayer optical coating machine. One of three deposition sources was a large (for meting 100 kg MgF) carbon crucible for a thermal source of the MgF. It was hotter than the sun, at 1x10-6 torr. A pinhole water leak in water cooled conductors (copper pipe) to the crucible heaters would degrade our vacuum in a periodic way, as it froze off to stop leaking as it expanded into the vacuum chamber, but would melt off as it carried current. It was only 6" from the thermal source at 2700C. Really showed me how good of an insulator vacuum can be, or really, how little energy is transferred with radiative heat transfer.
Thankyou for this discussion and fantastic video. I worked building a free electron laser that had a long waveguide from one room with klystrons to another room with the laser and many conflat joints in between with high vacuum (1x10 ^-9 torr) using ion pumps. Built in sections for testing initially we had a high rate of failure using the helium test, requiring disassembly and gasket replacement The waveguide joints are rectangular and the recommended torque pattern is criscross I changed this to begin with a criscross for halfway thru and finishing with sequential torquing. My theory was that a side of the rectangle was moving laterally across the knife edge when a knee threshhold of pressure was approached, thereby moving the knife edge on the copper gasket laterally. By sequentially tightening in contrast a pressure wave was induced linearly along the knife edge. After I switched methods we had no more failures and the P.I. was amazed
Awesome! I'd love to watch more cool info about your equipments and research. You could explain how it works and the principles behind it. Keep it up please!!
I really glad to see this video i am a mechanical design engineer, year 2016 i had been working in a company who made ultra high Vacuum equipments at Bangalore.
I know from my experience in the semi conductor industry that He leak checking can take a long time. Patience is golden. The most problematic leak I encountered was a virtual leak. Corrosion that exists of tiny holes that filles itself when the system is vented and takes forever to pump out.
This is why it is highly recommended to vent systems under vacuum with nitrogen, which does a good job of filling in all those micro cracks and preventing water (which takes orders of magnitude longer to pump out) from filling those cracks instead. Never vent a high vacuum system directly to air if it can be avoided.
This is so interesting as I work under intel at a intel site in Arizona as a Advanced mechanical analytical technician, my job is to go around the site and test gaslines and tie them into tools aka Big machines that make wafers for chips and I do this using 2 very expensive pieces of equipment. A (Inficon UL1000or5000) aka a leak detector which is a mobile pump that pulls to a negative pressure/vacuum until it’s reached its maximum vacuum spec. The other equipment is particle reader and a O2 reader they both speak for themselves anyway what I do with this equipment is exactly what I’d described in this video pulling a vacuum on gas lines to make sure there isn’t any leaks on any of the welds or vcr fittings that are present also using a helium tank to detect a leak which the leak detecter is best at picking up (HPHE-High purity helium) this is a very good informative video good job
Fascinating video. My friend ran a precision optics shop and had to apply various coatings to glass under vacuum. He said that working with high vacuum was an arcane art akin to black magic!
Nice video. When I was an undergrad I worked night shift at a particle accelerator. One night, I couldnt keep the beam aligned. I figured out the problem was a vacuum leak somewhere in the system. I was alone and proceeded to carefully take the system apart and back together section by section. When the chief engineer came in at 7am he asked: what the hell was i doing? I explained and he told me to reassemble it and then get him. I did and he said here is how you fix it. He grabs a wrench and starts banging on each of the vacuum pipes. He explains that occasionally an air (or helium, etc.) bubble clogs the system and banging knocks it loose. Beam was fine afterwards. Now my first attempt to fix anything is to either bang on it or reboot it.
Depends on the component. Silver plated screws are for anti-seize, but silver plated copper gasket are for oxidation control. Copper gasket on the air side will oxidize and flake off after high temp bake. Which will cause dust contamination inside the chamber during disassembly of the flange.
@@wadehsu2347 My 30 years as an ultra-high vacuum technician will force me to respectively disagree with your assesment. I have literaly compressed/replace thousands upon thousands of conflat seals, and have worked closely with the manufactures of them during my Time building the LIGO detectors. I have never once seen a coper gasket flake into a system, but I have seen plenty of the plain non silver coating stick to the knoife edge after baking, however silver plated never stick. copper migrates and will cold weld to stainless at high temp, silver will not. silver is used also as anti seize on vacuum hardware due to stainless liking to stick to itself, so it is used there as well.
Soo cool to finally see someone that I can relate to regarding the daily grind. I deal with this on the daily as well. I only work with 1x10-5 in an 18"x18"x18" chamber though. We use a little Varian helium leak detector to check for leaks. Great idea with the bag to concentrate the helium. We're using the ours to perform elevated temperature (2400°F) tensile tests on high temp alloys and more recently single crystal Niobium, so we're in and out of it about 4x/day. Keeping it clean is a chore!
Reason for your leak: You don’t seem to tighten in a cross pattern. You just go around tightening the bolts as you please. Makes a huge difference in mechanical engineering and so I bet it makes an even bigger difference in a “high vacuum” chamber
@@dupa333jelenia not really, but the rule of thumb is just to tighten slowly the opposite bolts, and do atleast 3-4 cycles or until there isn't a gap between the two flanges
I can’t even imagine! What’s it like working at LIGO? I remember watching the announcement when the first chirp was detected and marveling at the precision of it all
Same thing we do in my company we work with vacuum upto 10^-8. QMS is used for Residual Gas Analysis to find out what gases are in the vacuum chamber and use Helium to monitor pressure change to determine the position of posible leak.
Supper cool. I’m an electrician that just wires up labs like this one. It’s so hard to work in these labs after the equipment has been installed. I walk in set up my ladder and then the tech says don’t touch that it costs $100k. Ok I definitely don’t want to damage anything and leave. No wonder these labs cost millions of dollars to build. Might as well be a clean room. Usually clean rooms are specially built buildings not just a room in a building. Big differences. MBE machines inside RF chambers are the new lab standards. Its all very cool.
Very interesting, reminds me of how you can pressurize a sealed system (let’s say a refrigerant system) with nitrogen. Then you take an Oxygen sensor and wrap it near a join in Saran Wrap. You then pressurize the system with Nitrogen, because nitrogen will displace the Oxygen and if there is a leak, the O2 sensor will alarm because too much Nitrogen is displacing the O2 being read. All in all really cool video!
Dude your voice is so distinct. I don’t remember what video I saw but I remember your voice from a video I saw a couple years ago. Your videos make some of the most difficult concepts to understand so easy to conceptualize! My friend you have a gift and I had to subscribe. Today’s a slow day at the hospital so there’s no telling how many of your vids I’m gonna watch haha
I work in semiconductor and we use a helium leak checked just about every day. Super neat hearing other places it gets used, cause I’d never seen one outside of the building I work in
Hi friend. You brought back memories from f the MBE work I used to do at UCF. AlGaAs laser diodes grown in a Veeco Gen III. One time we spent two days looking for a leak in our shutter assemblies. Lots of jard work in general. Thanks for all your content.
This is amazing technology! I don't know if my experience might help you in the future, but about 40 years ago, while working as a new hire industrial engineer for a specialty computer chip manufacturer in NJ, did some engineering work on a leaky vacuum chamber used to make epoxy resin molds for a proprietary do-dad used on the manufacturing side that I didn't even have a clearance to see i was so new.. The machine had to hold a fairly steady vacuum for 8 hours during the curing process. It was a compact unit of no more than four liters, drawn down to a nominal 29 and change @ 59*F with a high-quality AC service style vacuum pump. While pondering the idea of sliding the whole thing inside clear trash bag and using a bit of freon and a Honeywell halogen detector to read the contents, I had an epiphany of sorts - to check each of the five or six parallel ports one at a time. I took a side trip to a pharmacy - bought a dozen "magnum" sized condoms ( on my less than a week-old expense account....) stretched one over each port, and sealed the "semi flaccid" arrangement tightly around the pipe nearest to where it exited the chamber. As the vacuum pump drew down, it also acted on the condom that surrounded the leaking port and its valves - All but one remained unmoved - the condom over the leaker visibly shrink then flattened out as tightly as a conformal coating until the latex finally sealed off the leak and the pump reached its draw-down limit. . As soon as the vacuum was relieved, the location of the leak was made evident at the spot where the condom first released its grip on the surface. Who knows - given what I can see in the video, perhaps something along the lines of a small weather balloon might work to more quickly narrow down your leak.
One of the generic task I do during my daily job is designing of o ring seals for high vacuum chambers for semiconductor tools. And this video is very precise, informative and entertaining at the same time. Keep up the good work 🙌🏻
In my job we use stainless steel gaskets for the gas delivery system into the vacuum chamber, the vacuum chamber itself uses rubber gaskets. The leak check we use is helium as well, really cool system.
One of my consulting clients was Flexitallic, they made spiral wound metal gaskets for vacuum applications like this, but they mostly made high positive pressure seals for steam and petrochem.
I'm a vacuum tube engineer and routinely deal with these systems in industry. Great video, but remember, that at 1e-12mBar, you still have about 1 billion molecules per cubic centimeter and I would certainly challenge your statement on 'not a single atom of air can leak through this gasket.' I will agree that below the -9 through -12 decades the vapor pressure of the metals inside the chamber become more significant than the rate of air leaking in, but it is in fact non zero, and even the best vacuum systems still have billions of molecules inside any useful volume. The leak rates that also approach your -12mBar/L/s range also indicate that your leak rate is effectively on the orders of millennia and that even a tiny ion pump can keep up with this nearly unmeasurably small leak rate.
Interesting, is there a reason you do not pressurize to find a leak? It seems the seals should work in either direction and it would be easier to find a leak from a pressurized system. You don't call those seals crush washers like we do in other industries?
I was in a lab that had both Ebola and Anthrax... So I thought that was pretty fascinating. What was probably more fascinating was how casually I was invited in there..... The professor who worked there told me that once he retired, he planned on making another lab in his old garden shed. I told him "That sounds absolutely wonderful!"
Great video. Reminds me of when I used to apply "clear bra" plastic to cars, and a single eyelash or piece of dust could mean you were redoing hours of work. Even full Tyvek suits could drop a tiny thread and ruin my entire day. I'll bet the cleanup and keep-clean processes at this lab are unfathomably meticulous. Respect.
I used to work at a pharma site looking after their freeze drying plant. We used the exact same method to find leaks on the vacuum chamber, although we attached an external helium leak detector to the equipment.
My workplace has a fairly decent low mass mass spectrometer, and we've found that Swagelok steel compression fittings can have a leak rate about 1E-15 torr L / s. CF fittings are at most 1E-18 torr L / s. We would have to let a CF fitting 'leak' for over a month to get any measurable results, which we didn't do. For comparison, Swagelok rates their fittings to 1E-9 torr L / s and our high end leak detector (ASM340) goes down to 3.8E-12 torr L / s.
I absolutely love your videos, and I've only seen a few of them. You're not only good at what you do, but you're really passionate about it and very excited about it, and I love it.
Super cool informative video. I've recently been welding stainless fittings for an ultra high vacuum application for a company near my shop that makes xray equipment. Just for their r&d department. Neat to see the process of how they are checking my parts for leakage. All copper washer, knife edge flanges. I have to be really careful about my grounds while welding to make sure I don't exit the flange on the knife edge and cause a blemish on that edge.
Multi-reply:
"cross-tighten with a torque wrench"
- As of this post, I have 40 comments suggesting a torque wrench and approximately 50 recommending a star- or cross-tightening pattern. I hear you. In undergrad (maybe sophomore year?) I was helping to service a piece of vacuum equipment and was actually corrected FROM using the star pattern. I was taught that on these copper conflat flanges, "cutting in" the knife edge linearly by tightening in a circle was the way to go unless you had an absurdly large flange (which this is not). Trying to get a UHV seal with these gaskets is nontrivial, and if you have a system that works, you stick with it. I also know people who cross-tighten their conflat and get great results. In applications like this that are super-finicky, any change to procedure comes with risk. if it aint broke, don't fix it.
- During this opening there were maybe 6 or 8 flanges open, and only one of them leaked, and the flange that DID leak had already been problematically leaky during previous openings, so I don't attribute the leak to a problem in tightening but more than likely, a scratch on the flange knife-edge that requires you to torque the gasket down past spec to fill the extra gap with copper.
- Regarding the torque wrench, in many many locations on this system, it's almost impossible to fit a regular wrench around these flanges, let alone a necessarily bulkier torque wrench. This flange probably could have been handled by an open-end torque wrench, which I guess I assume exists but have never seen, but in general to work on these academic systems you need to develop a feel for it. Notice how I'm barely moving the wrench with each tighten.
"you are wasting helium"
- This is actually an extremely common and extraordinarily critical industrial use for low-purity helium. THIS is what the party balloons should be saved for. This method was developed to detect leaks so small they don't generate bubbles, noises, or other observable airflow. we're talking about a handful of atoms weaving their way through cracks in apparently solid pieces of metal. I've found a leak straight through a stainless tig weld on this system using helium. Amazingly enough, a LOT of the semiconductor industry depends on being able to create high quality vacuums for processing.
interesting that there's still stuff that's just based on what you were passed on and there's not one best way known (like the tightening pattern you use). makes it feel like a learned hard skill rather than just a purely academic pursuit
You did 2 ugga duggas when you should have done 3.
Ultimately conflats are not as finicky as you think. I've reused them on training vessels with a Extremely high vacuum to some success. Landing the groove in the old groove had to play a crucial part in this and the bite was pretty solid. What mattered the most is that we always followed torque specs and when those failed we upped the specs and still followed the cross pattern starting from marked bolt 1. I don't recommend this as a operating guideline, but I'm telling you that those rings are tougher than you think. Following good engineering practice and a little bit of elbow grease you can get a used copper conflat to hold for at least 40 hours in extreme vac. I've seen it done. It did fail, but we only had the one in the training lab so we waited for a batch of new ones to come in, but it will do it.
As someone also in the UHV side of semi, the law of the land where I am is star tighten after it's hand-snug, and always to designed torque. By not doing that you risk dis-leveling the mating surfaces and could end up overcutting a seal, deforming it more than necessary and then actually creating a weaker seal. This leak might have been avoided by proper torquing patterns, but like you said, typically that type of failure can be hidden by deforming the seal even more with additional torque. Just wait until you try baking out the system though, that additional deformation is likely to come back to haunt you.
@@sejongthegreat3044 Does the material expand at all due to tempature changing the torch specs or allowing enough of a gap to make a difference? I don't know why but It instantly popped in my head while reading this all.
Fantastic video. Clear, methodical, and you walked me along the entire troubleshooting path. Well done!
Yo Destin - I love your channel!
Glad you like the video! I’m not sure why leak-checking an MBE has been deemed viral material but my channel has had a wild week and a half and I’m glad you made it here!
19 min Ago on a one year old video. Is this a clue for something upcoming :D. Or just weird youtube algorithm
I figured we were all getting recommended this around the same time; I know when I see a banger video topic like this pop up in the recommendations out of no where that I am one of many fine lads the algorithm chose to bless today
Nice to see u here
@@AlphaPhoenixChannel because the Titile is really intriguing i want to know what the different b/w air tight and vaccume tight. also that's how mysterious Algorithm works
"If tightening it doesn't work, tighten it more" -Engineer
unless it is meant to move, in which case lube it
Wd 40 / duct tape
Whack it with a wrench to make it work
How to properly torque a bolt: tighten until the threads strip, then back off a quarter turn.
@@JinKee everything shall be lubed everything
"you may assume a perfect vacuum"
*engineer cries in relief*
lol
Are we allowed to assume spherical cows with their mass concentrated in a single point as well?
@@akrinornoname2769 Also don't forget pi = 3
@@alexcompagnucci8508 gravity = 10 m/s^2
@@alexcompagnucci8508 pi = 3 = e
“I could talk about lab equipment for hours and hours” I for one would love to listen about lab equipment for hours and hours. Even if you posted barely edited ramble style videos about niche and specific lab equipment, i’d definitely watch them
"the world is running out of helium!"
"But what could you need it other than for making squeaky voices?"
"Scientific purposes!"
"Like what?"
"Releasing it into the air so it makes everyone in the lab's voices all squeaky"
You made me laugh while I was eating chocolate and now I have to clean up my desk.
But like for real I’ve had a lot of comments saying this wastes precious helium but THIS is one of those critical industrial uses everybody always talks about…
@@AlphaPhoenixChannel No worries, just use hydrogen, the molecules are even smaller. What could possibly go wrong?
I'm not sure, but we had a helium detector as part of the ion pump AFAIR. I'm also pretty sure the helium nozzle was a calibrated leak, not sure why. But the process was exactly the same back then, 1968. What was impressive was the way the helium alarm would trigger even while you were just setting things up the other side of the lab.
Lmao 🤣
@@daze8410 I've never understood why they don't just collect the helium instead of just flushing it
"hey whats inside this thing?"
"nothing, truly nothing"
Virtual particles: *allow us to introduce ourselves*
Its filled with quantum fluctuation+neutrinos of course
So what's inside this chamber?
SPAAACE
@@Auhrii Ask them to sign your guest card and they just disappear.
@@Z-zl yeah, this engineer guy kinda looks like he did a bump before recording. Or maybe he's high on science
Gordon doesn't need to hear all this He's a highly trained professional
@@Dezmont01 Ah yes you're right, Gordon we have complete confidence in you
@@jimbow7787 Well, go ahead. Let's let him in now.
Why do we all have to wear these ridiculous ties?
Ah Gordon! Ho grab that wrench over there!
Good, now gently tighten the bolts... Gently.... Geeeeently... *Beeping starts* AHHHHH! Gordon what have you done!?
We have to hope. Otherwise there could be some "Unforeseen Consequences".
Im really glad Tom Scott encouraged us to watch this in his most recent newsletter, because its fascinating stuff; I never considered how much such a insignificant leak could multiply so much under a vacuum.
Tom has a newsletter?
@@NoNameAtAll2 Yes, in case you haven't made it there already, it has a few suggested youtube videos and a few other internet articles/ findings and is weekly. Really interesting! (I'm going through a backlog of them currently which is why I'm watching it now, 3 months after the newsletter date)
@@macaronisheep where does one find it?
@@NoNameAtAll2 if you google Tom Scott, his website was the first result for me and right on the front page is a link to sign up to it :)
(Not keen to put link in comments in case spam filters catch me)
this is also the reason Elon Musks "Hyperloop" won't be built.
Previous semiconductor worker who delt with vacuum and pressure tight vessels and pipes.
-Pressure tight is way harder, especially when the differential is 100psi or more, and silane, hydrofluoric acid, anhydrous ammonia...you know the things that dissolve humans or reliably spontaneous combust in air. With a vacuum you have chances to mess up, with pressurized death gases/liquids, you have 0 chances. Also helium and hydrogen pressure tight is the worst, but we usually used helium to test connections of pies to contain worse chemicals before putting them into production. That being said, randomness from the machines vibrating, and human error caused leaks and the fun beganafterwards.
I'm surprised you tightened the head bolts sequentially around the head rather than in a star pattern to avoid uneven distortion of the gasket. Any reason for this?
That is correct for initial tightening down to torque spec to insure even forces... but, not so much after proper specs have been reached... the "star pattern" (most commonly used) is to keep the pressures close to the same around the circumference and while increasing the torqued pressures to spec to maintain "flatness" or parallel surfaces. Micro-adjustments afterwards are a completely different story. The torque forces applied at this point are increasing in tiny fractions and are deforming the metal gasket like a "putty" at those pressures, Essentially, instead of applying the forces in the flat plane for maintaining parallelism, you actually want to theoretically "chase around" a minuscule blob of metal to fill in those gaps that are allowing molecules through.
@@edwardpaulsen1074 A fascinating response to a question that had me thinking already. Thank you both!
@@edwardpaulsen1074 but can't you inadvertently still maintain the leak? By chasing/pinching the gasket seal around to a single point (between your start and end locations) can't you effectively create a material imbalance in that area maintaining the leak and making it harder to seal?
@@miamisasquatch well if that happens, don't you just tighten it again?
@@miamisasquatch You can inadvertently do pretty much anything as long as its not what you were planning on but sealing vacuum isn't that much of a challenge to be honest. Your car tires are probably like 40psi or something like that which is a 40 psi difference to atmospheric pressure and they seal just fine for the most part. Absolute vacuum is just -1atmosphere so from outside to inside of the vessel there is just a pressure difference of 14.7 psi. So don't worry about it just huck some copper in there and you're good.
Glad that this channel is now appearing on my Recommended! Fascinating videos!
⠀
No likes? Impossible
Woah
Hello Eret!
I didn't know you were into engineering
Cool
hey eret!
I've just been continuously watching vids on this channel throughout my entire day, I can totally see you hitting 2^20 subscribers very, very soon
@@revan552 it's a thing with this channel that every doubling, a new play button is made by him... Not oddly specific at all
Of course a channel I love would watch channel I love.
@@nutmeggaming11261 it is oddly specific. You're wrong. Take the embarrassment
i watch your videos lol
more like 1^ ... wait...
I'm in no way, an engineer. But I loved this well explained video. Thanks for bringing us into your lab, and simplifying the process for us!
Thanks for a very informative video! I spent a couple of years working as a technician at a scientific research facility and often enjoyed wandering through and marvelling at the beam pipes, examining some of the sensor and control equipment and was struck by the constant hum of the various vacuum pumps which had to run continuously to keep the pipes at a near-vacuum as a particle beam was injected.
I finally managed to get a He leak detector for the lab I work at. It is a Varian that was made in about 1985 and needed some work. I managed to get it up and running, and boy am I glad I got it. There is a fancy room that we put together and it took 2 guys 3 weeks to get it mostly leak free (I used that to help justify the leak detector). I later needed to do some modifications to the room which meant finding all of the leaks that were sprung, and it took me about 4hrs to be confident that the room no longer leaks (this included a few leaks that were missed by the previous guys). I put that in there in my status report email as a sort of "I told you so".
GOT EM
A poor man’s leak detector is using a very sensitive pressure sensor (which you will already have in an ultra high vacuum system) and just taking a squirt bottle and spraying IPA around the machine seals and bolts. If there is a leak the IPA will go inside the leak and when the alcohol hits vacuum it “explodes” and you see the pulse on the pressure sensor. I once watched a Russian physicist do this while his He detector was down and he figured out all the leaks without it and was able to achieve better than 10^-9 (it’s been awhile I forget the exact number but it was our personal record) lol
@@latenttweet that's what the first crew was doing. This is also how we leak check our systems with a convection gage.
The problem with this system though is that it has a 200gal reactor on it to act as a huge buffer, but it also uses a very crappy air-referenced vacuum gauge.
This system works from atmosphere down to 100Torr or so.
The He leak detector worked a treat though, but it would take a while for all of the He to get sucked out of the system to allow the leak detector to calm down again.
Great video! Would love to see more of this. I’m especially curious how such a high vacuum is achieved besides having perfect seals.
I’ve got an ever-changing plan for a video on pumping in the works. There are 7-8 pumps on the system I think and we add two more during servicing.
Video about vac pumps! Looking forward to it!
I had a 160 hour course on the subject while at university.
And that's only the start.
Vaccuum technology is a whole branch of study and engineering.
Pumps only go so far. A molecular sieve system is usually used to remove remaining molecules to get closer to a perfect vacuum.
@einstein9073I assume Roger meant something more along the lines of "Scientists like to do experiments with different shit in vacuum because its cool and cool shit happens" because it would be really stupid if he meant something more like "Scientists like to do experiments on a vacuum... because they are curious as to how nothing keeps doing that weird thing it's doing".
The gasket no atom can pass.
Hydrogen: *is that a challenge?*
Click-bait.
@@DrDeuteron While that's true, hydrogen can chemically diffuse through some metals, making it basically unstoppable. This is especially true at elevated temperatures.
@@DrDeuteron helium is smaller than hydrogen?
@@bladdnun3016 Hydrogen permeates through all materials, especially steel. PTFE aka Teflon has the lowest permeation rate of any material available and hydrogen still permeates through it.
@@peterwill9660 I have seen helium permeate through .250" carbon steel plate. Drove me crazy trying to find the "leak" with a "sniffer" in my welds that I was certain were good.
I had the privilege of working on the upgrade of a fusion reactor in the UK. During my time on the project the reactor was being checked for leaks exactly like this. Totally fascinating engineering. Brilliant video :-)
How many working fusion reactors does the UK have?
@@tedmoss I’ve only heard of JET, the joint European torus
I'm jealous. I want my own fusion reactor.
@@tedmoss There's the JET Reactor and MAST Reactor which are in operation. I was part of the upgrade of the MAST Reactor.
how cool is it to just casually say "oh yeah i was working on the upgrade of this nuclear fusion reactor last summer" in conversation
Absolutely fascinating !!! I worked in a company that sold and serviced vacuum equipment.......but nothing quite on the order of what you work with. Regardless, although not part of my responsibilities, I had to learn about vacuum equipment to be able to discuss leakage (and other types of) problems with the engineers who were tasked with producing solutions for our repair people. What I DID learn was that although everyone thoroughly understands pressures above atmospheric pressure, very few people (including many engineers) have a real problem understanding/coping with problems concerning negative pressures. It seems that we're hard-wired to work with positive pressures. I had to sit with some engineers and use my semi-pro knowledge of negative pressures to get those engineers to "re-engineer" their thought processes. They all understood the principles theoretically but couldn't make the mental "leap" to practicality. It was certainly an ego boost for me to be able to see the "light bulb" come on over an engineer's head when my "teaching" finally sank in. Ah, the glory days........before I retired.
What a great vid !!!
I would literally pay for content of this quality released on a once a month/nile-red-type basis
also, touching on superlative topics, like whatever would be considered a catastrophic failure in this system, would be super interesting.
I knew nothing of vacuum chambers and now I feel like I could machine my own that could be half as good as this. You are an amazing communicator.
I used to work on an ultra high vacuum system. We regularly had to vent and open the system to replace test samples etc. I'd love to see a video on your method for pumping down the entire system.
You really had to open the entire system every single time you had to change the sample?
@@DocSineBell how else they gonna get it lmao
@@zachmarshall6059 a fast entry chamber. A very small chamber, with a small volume, that can be easily isolated and opened. Takes from minutes to hours to pump perfectly a fast entry. Takes days to a week to pump and bake out perfectly a whole big ass UHV setup.
@@DocSineBell Maybe it wasn't that big a setup. Still, even for small stuff like detail electronics or pacemaker bombing etc. I've seen setups with cutoff valves. The trouble tends to be that even labyrinthing with 1- or 2-mp valves sunk into the build, you're still taking your bloody chances that the seal tightness is going to be sufficient.
@@zachmarshall6059a load lock chamber.
I started research in a lab this semester using a gravimetric microbalance and mass spectrometers as well, the microbalance uses the same copper gaskets as you are, albeit at a much less perfect vacuum. It is so cool to see the stuff I use in different labs as well
I work with a few guys like you, who can just talk and talk and talk about the equipment they work with. Always fun to ask them questions! Really interesting video!
It was one of the best parts of looking for a new job when I last did it. I know enough about a lot of things in order to ask good questions that elicit a ramble about someone's work. Learn new things, get excited about the human endeavor, what's not to love!
Totally know the feels. Had a vacuum drop in my ToF-SIMS from 10^-11 to 10^-8, which was screwing with our cesium gun.
We were up for maintenance anyway, so we went with the invasive option and just changed all the copper gaskets.
Fortunately after putting it all back, there were no leaks!
name a better combo than anime profile pics & time-of-flight secondary ion mass spectrometry
@@von... ppl come in all shapes, colors and types of hobbies!
Also, it's a game character 😉
That’s quite the vacuum.
How does ToF-SIMS compare to positron annihilation induced Auger spectroscopy? My PhD work is in space physics so I know little about this but my friend works with positron beams measuring what gets knocked off a target and sent down a ToF tube. Both are high vacuum work and seem super cool.
"We use it to grow crystals"
*breaking bad music plays*
*Breaking Benches*
-These vacuums are pure up to 10^-10 torr. You must be him.
-Say my name.
-Ultra High Vacuum
-you’re goddamn right.
I wonder. Maybe
Gasket: "NONE SHALL PASS"
Hydrogen: "hold my beer"
Helium enters the chat.
@@HappyHarryHardon - brilliant mate, good on ya
@Andrew Crews lol
Neutrinos: "beesh, please."
lets just bypass it with a armor piercing bullet full of ozone (idk why ozone but who cares)
I like how real you are, you approach your videos as a person living in real life, rather than masquerading as something you're not like most people do
We make quadrupole mass specs at my workplace. I love hearing an outside view on how they work and the leak test process as well!
As someone who works in semiconductor equipment manufacturing I cannot imagine trying to keep vacuum flanges clean in a non-clean room environment
Well there is a huge difference between labs and manufacturing plants. An contamination causing an issue in less than 1% of all experiments might be neglected by many scientists, but at high quantities and low margins, 1% might be the difference between earning a lot of and losing a lot of money.
Thanks for bringing up my memories on working with UHV equipment on university. Memories of many, many bolts, some broken feedthroughs and unreal toughness of platinum wires. And waiting for right pressure to actually start the second and third stage pumps. Good memories :)
I used to work with high vacuum equipment myself--ironically, making parts for quadrupole mass spectrometers! Our vacuum requirements were a little less extreme than yours ("only" in the 10e-7 to 10e-8 torr range, so we could use rubber gaskets) but all the principles were the same. I may point people to this video in the future when I have to talk about my old job!
This is so cool! I’m currently doing my PhD in physics and my friend here works with positron annihilation Auger spectroscopy, so they use a UHV system that measures what comes off a target with a ToF tube. He was talking about the flanges and cleaning the system the other day but I never really truly understood the work that goes into maintaining it until now. I do simulations all day so this is all alien to me!
Love all the replies from the scientists in this thread. We’re all just geeking out together and it feels great.
I am an ion implant engineer and leaks under high vacuum is one of the most common fail modes. This is a killer video
Amazing that molecules can squeeze their way through seals in a near perfect vacuum, considering it's only ~14.7 psi on the chamber.
Nature abhors a vacuum
its still full of neutrinos
@@shadesmarerik4112 - I heard they mutated and are heating up the planet.
@@RFC-3514 u heard? where?
@@shadesmarerik4112 - From a famous Indian scientist: th-cam.com/video/bXdBzpRDR5I/w-d-xo.htmlm33s
Mass spectrometer? Making crystals?
*Flashbacks to that one incident in the anomalous materials lab in Sector 7G*
at least it isn't anti-mass!
We assured the administration that *nothing will go wrong*
_Why do we have to use these ridiculous ties?_
@@ghoulbuster1 As in the nothingness (the vacuum) inside the chamber will fail :P
They're waiting for you Dr Freeman, in the test chamber.
This was very interesting. You are a charming and effective host. Thank you for sharing!
I lovr watching happy people talking about things they are passionate about. Always charming
I worked on an STM during undergrad, and it's interesting seeing so many of the principles I learned in that lab being reflected here. Same copper gaskets, same residual gas analysis, same pumps. I mean, that's not surprising for obvious reasons, but it's still interesting to see.
Cool! What were you looking at?
@@AlphaPhoenixChannel We were studying the surface of Ag(111). To be honest, I only got a few hours actually running the STM and never got to do any analysis myself. My efforts were almost entirely focused on building the STM with my professor (and then taking it back apart because the university made him move it).
I can never understand why people “thumbs down” this type of content?
This guy gives me major "Jake from Avatar" vibes, with his appearance and the camera angle and content etc.
When I was working on my chemistry PhD, one of the profs had a ultra high vac system he used for is surface chemistry research. I had a class with him where we covered his to find a leak with He. 10^-12 Torr is tough to obtain. Breaking vacuum, cleaning and pumping down is certainly a lot more involved than with my GC-mass spec.
I refer to Vac-Seal as "Liquid Shame".
Spent the first year of my career in machining at a shop that made components like this for the semiconductor industry, it's nice to finally see someone showing how a knife edge seal works, the thing that was always so mind blowing to me was the hours we spent hand laying the finish in a certain direction on oring seal grooves so that we didn't create a leak path. Very cool video
I’d love to see how these are manufactured - I can attest that edge is fragile. This same spot had leaked more than once so I think it must have gotten scratched during assembly or a previous service
Unfortunately I no longer work at that shop, but the way we cut knife edges was to bolt the plate to a rotary axis and then to tilt that axis by the angle of the knife edge then rotate the whole plate around the center of the knife edge circle , it became a long day when we were doing flanges with 6+ gunports all splayed out at angles from centerline so they'd have clearance for the larger end of the gun
It's not "obvious" that just because someone works on something, that they find it fascinating. I've run across co-workers in my field of technology who found it an absolutely excruciating bore, but kept at it for the salary. I always felt really sad for those kinds of people. But all of that to say, consider yourself lucky that you're able to work on complex and intricate technology and ALSO find it endlessly fascinating. Cheers!
I worked in the mineral assay industry for many years, and mass spectrometers were pretty much the pinnacle of available instrumentation for the purpose of trace element analysis... for your machine, it's a leak detection accessory. Mind blown.
I found this extremely fascinating! The way the gaskets are deformed like that to create that much of a seal is so cool! (I’m such a nerd)
“Ultra High vacuum”
Brought to you by the guy who created “extra extra extra large” the size 😂
That’s the best part of scientific/medical terms, everything is categorized and defined the same way. So all the terms sound oversimplified specifically so you know exactly what it means even if you haven’t heard of the concept before. If only English could learn a thing from that.
I´ve actually walked by a clothes stand outside a clothing store where an ungodly amount of X´s were before the "L". 5 or maybe even 7 X´s, i believe.
While i walked by it, i quietly said the size with all the "extra"s to myself under my breath and came to the conclusion that i think a new word should be invented for sizes beyond the second or third "X".
Extra large,
extra, extra large,
extra, extra extra large,
impressive
extra impressive,
extra, extra impressive
momentous...
gargantuan...
royal...
serious...
@@GarretRB Yep, this naming happened with radio bands as well. HF -> VHF -> UHF are high -> very high -> ultra high frequencies.
It keeps going too. SHF - super high frequency, EHF - Extremely high frequency, THF - tremendously high frequency
I'm a little mad it doesn't go Super, Ultra, Mega though.
Its even worse in Biology. They never discover the enzymes that perform sequential steps in a reaction in order, so you get used to struggling with "Enzyme IV coverts A to B, then enzyme II converts B to C, and finally, Enzyme VI coverts C to D; however, sometimes Enzyme I also converts some B to D".
I did vacuum testing for about 3 years. I was only testing at high vac but we also did thermal testing in our chamber. That brings in a whole different issues with out-gassing. We’d always having issues maintaining our required vacuum level when going to hot temperatures. We also did our leak checks with IPA wipes. It never really worked too well but tightening or replacing a gasket was fairly easy and worked every time. It also takes about forever to heat things in a vacuum when it’s only bolted to a plate.
I worked with a large scale multilayer optical coating machine. One of three deposition sources was a large (for meting 100 kg MgF) carbon crucible for a thermal source of the MgF. It was hotter than the sun, at 1x10-6 torr. A pinhole water leak in water cooled conductors (copper pipe) to the crucible heaters would degrade our vacuum in a periodic way, as it froze off to stop leaking as it expanded into the vacuum chamber, but would melt off as it carried current. It was only 6" from the thermal source at 2700C. Really showed me how good of an insulator vacuum can be, or really, how little energy is transferred with radiative heat transfer.
Yesssssssssssss! Talk about the work you do that you are able to!
Keep up the wonderful work, love seeing someone passionate about what they do. And as always, great presentation
I do indeed find this stuff very interesting. Thanks for sharing it!
Thankyou for this discussion and fantastic video. I worked building a free electron laser that had a long waveguide from one room with klystrons to another room with the laser and many conflat joints in between with high vacuum (1x10 ^-9 torr) using ion pumps. Built in sections for testing initially we had a high rate of failure using the helium test, requiring disassembly and gasket replacement The waveguide joints are rectangular and the recommended torque pattern is criscross I changed this to begin with a criscross for halfway thru and finishing with sequential torquing. My theory was that a side of the rectangle was moving laterally across the knife edge when a knee threshhold of pressure was approached, thereby moving the knife edge on the copper gasket laterally. By sequentially tightening in contrast a pressure wave was induced linearly along the knife edge. After I switched methods we had no more failures and the P.I. was amazed
Awesome stuff! What were you pointing the FEL at?
I never thought I would be so interested in whatever this is
Very interesting video. I would love to see more.
Awesome! I'd love to watch more cool info about your equipments and research. You could explain how it works and the principles behind it. Keep it up please!!
This is supper cool! I would actually love to know what kind of pumps and for how long they need to work to remove “all” atoms inside the system
It’s some fascinating technology - I really want to film a video about achieving ultra high vacuum but don’t know now when it will be possible...
Molecular pumps such as diffusion or turbomolecular pumps, and also cryogenic pumps
imeaktsu7 turbo-molecular pump, rough pump, tight vacuum chamber, gases, heat.
I really glad to see this video
i am a mechanical design engineer,
year 2016 i had been working in a company who made ultra high Vacuum equipments at Bangalore.
I used to work for a company that built large vacuum chambers. leak checking and fixing welds is a whole other kinda fun.
"Where's the leak, Ma'am."
-Patrick Star
Awesome and interesting insider look. Thanks AP!
These videos keep blowing my mind! Love it! (Steve Mould sent me)
I know from my experience in the semi conductor industry that He leak checking can take a long time. Patience is golden.
The most problematic leak I encountered was a virtual leak. Corrosion that exists of tiny holes that filles itself when the system is vented and takes forever to pump out.
This is why it is highly recommended to vent systems under vacuum with nitrogen, which does a good job of filling in all those micro cracks and preventing water (which takes orders of magnitude longer to pump out) from filling those cracks instead. Never vent a high vacuum system directly to air if it can be avoided.
This is so interesting as I work under intel at a intel site in Arizona as a Advanced mechanical analytical technician, my job is to go around the site and test gaslines and tie them into tools aka Big machines that make wafers for chips and I do this using 2 very expensive pieces of equipment. A (Inficon UL1000or5000) aka a leak detector which is a mobile pump that pulls to a negative pressure/vacuum until it’s reached its maximum vacuum spec. The other equipment is particle reader and a O2 reader they both speak for themselves anyway what I do with this equipment is exactly what I’d described in this video pulling a vacuum on gas lines to make sure there isn’t any leaks on any of the welds or vcr fittings that are present also using a helium tank to detect a leak which the leak detecter is best at picking up (HPHE-High purity helium) this is a very good informative video good job
Even if what you're explaining is very simple, it is very interesting! Thanks:)
This is absolutely fascinating stuff! Thanks!
NEXT EPISODE: Fixing the space station with a steel wire
Fascinating video. My friend ran a precision optics shop and had to apply various coatings to glass under vacuum. He said that working with high vacuum was an arcane art akin to black magic!
Nice video. When I was an undergrad I worked night shift at a particle accelerator. One night, I couldnt keep the beam aligned. I figured out the problem was a vacuum leak somewhere in the system. I was alone and proceeded to carefully take the system apart and back together section by section. When the chief engineer came in at 7am he asked: what the hell was i doing? I explained and he told me to reassemble it and then get him. I did and he said here is how you fix it. He grabs a wrench and starts banging on each of the vacuum pipes. He explains that occasionally an air (or helium, etc.) bubble clogs the system and banging knocks it loose. Beam was fine afterwards. Now my first attempt to fix anything is to either bang on it or reboot it.
"Actual engineering" ... "the real science"... you know you're talking to an engineer when they use these terms.
Cuz it’s sorta kinda true...
That uhhh, “potato cannon” looked a little sus lololol
silver plating is not for corrosion resistance, it is used as an anti-seize in high temp aplications
Depends on the component. Silver plated screws are for anti-seize, but silver plated copper gasket are for oxidation control. Copper gasket on the air side will oxidize and flake off after high temp bake. Which will cause dust contamination inside the chamber during disassembly of the flange.
I think he knows what he's doing
@@wadehsu2347 My 30 years as an ultra-high vacuum technician will force me to respectively disagree with your assesment. I have literaly compressed/replace thousands upon thousands of conflat seals, and have worked closely with the manufactures of them during my Time building the LIGO detectors. I have never once seen a coper gasket flake into a system, but I have seen plenty of the plain non silver coating stick to the knoife edge after baking, however silver plated never stick. copper migrates and will cold weld to stainless at high temp, silver will not. silver is used also as anti seize on vacuum hardware due to stainless liking to stick to itself, so it is used there as well.
Thanks! Now I can finally fix that pesky leak in my MBE chamber I keep in my basement.
9/10 customer tech support.
Soo cool to finally see someone that I can relate to regarding the daily grind. I deal with this on the daily as well. I only work with 1x10-5 in an 18"x18"x18" chamber though. We use a little Varian helium leak detector to check for leaks. Great idea with the bag to concentrate the helium. We're using the ours to perform elevated temperature (2400°F) tensile tests on high temp alloys and more recently single crystal Niobium, so we're in and out of it about 4x/day. Keeping it clean is a chore!
Amazing content.
Keep it up :D.
Whoever drew that potato cannon had something else on their minds...
Reason for your leak: You don’t seem to tighten in a cross pattern. You just go around tightening the bolts as you please. Makes a huge difference in mechanical engineering and so I bet it makes an even bigger difference in a “high vacuum” chamber
That was my first thought when I saw it. Does the device come with a tightening pattern? Many a head gasket on an ICE failed due to this.
Couldn't read the pinned post where he addresses this? You lost that bet.
@@dupa333jelenia not really, but the rule of thumb is just to tighten slowly the opposite bolts, and do atleast 3-4 cycles or until there isn't a gap between the two flanges
As a vacuum engineer for LIGO this was a well done video! Now imagine trying to leak hunt down a 2.5 mile long vacuum arm...
I can’t even imagine! What’s it like working at LIGO? I remember watching the announcement when the first chirp was detected and marveling at the precision of it all
@@AlphaPhoenixChannel Its a blast! Currently feels like one fiasco after the next though. We are in the process of upgrades.
Same thing we do in my company we work with vacuum upto 10^-8. QMS is used for Residual Gas Analysis to find out what gases are in the vacuum chamber and use Helium to monitor pressure change to determine the position of posible leak.
"We use it to grow crystals" why did my brain jump straight to meth
So cool! I'm always amazed at how scientists find simple solutions to problems that seem impossible. Way cool!
Supper cool. I’m an electrician that just wires up labs like this one. It’s so hard to work in these labs after the equipment has been installed. I walk in set up my ladder and then the tech says don’t touch that it costs $100k. Ok I definitely don’t want to damage anything and leave. No wonder these labs cost millions of dollars to build. Might as well be a clean room. Usually clean rooms are specially built buildings not just a room in a building. Big differences. MBE machines inside RF chambers are the new lab standards. Its all very cool.
Very interesting, reminds me of how you can pressurize a sealed system (let’s say a refrigerant system) with nitrogen. Then you take an Oxygen sensor and wrap it near a join in Saran Wrap. You then pressurize the system with Nitrogen, because nitrogen will displace the Oxygen and if there is a leak, the O2 sensor will alarm because too much Nitrogen is displacing the O2 being read. All in all really cool video!
Dude your voice is so distinct. I don’t remember what video I saw but I remember your voice from a video I saw a couple years ago. Your videos make some of the most difficult concepts to understand so easy to conceptualize! My friend you have a gift and I had to subscribe. Today’s a slow day at the hospital so there’s no telling how many of your vids I’m gonna watch haha
I work in semiconductor and we use a helium leak checked just about every day. Super neat hearing other places it gets used, cause I’d never seen one outside of the building I work in
Hi friend. You brought back memories from f the MBE work I used to do at UCF. AlGaAs laser diodes grown in a Veeco Gen III. One time we spent two days looking for a leak in our shutter assemblies. Lots of jard work in general.
Thanks for all your content.
This is amazing technology! I don't know if my experience might help you in the future, but about 40 years ago, while working as a new hire industrial engineer for a specialty computer chip manufacturer in NJ, did some engineering work on a leaky vacuum chamber used to make epoxy resin molds for a proprietary do-dad used on the manufacturing side that I didn't even have a clearance to see i was so new.. The machine had to hold a fairly steady vacuum for 8 hours during the curing process. It was a compact unit of no more than four liters, drawn down to a nominal 29 and change @ 59*F with a high-quality AC service style vacuum pump. While pondering the idea of sliding the whole thing inside clear trash bag and using a bit of freon and a Honeywell halogen detector to read the contents, I had an epiphany of sorts - to check each of the five or six parallel ports one at a time. I took a side trip to a pharmacy - bought a dozen "magnum" sized condoms ( on my less than a week-old expense account....) stretched one over each port, and sealed the "semi flaccid" arrangement tightly around the pipe nearest to where it exited the chamber. As the vacuum pump drew down, it also acted on the condom that surrounded the leaking port and its valves - All but one remained unmoved - the condom over the leaker visibly shrink then flattened out as tightly as a conformal coating until the latex finally sealed off the leak and the pump reached its draw-down limit. . As soon as the vacuum was relieved, the location of the leak was made evident at the spot where the condom first released its grip on the surface. Who knows - given what I can see in the video, perhaps something along the lines of a small weather balloon might work to more quickly narrow down your leak.
When you were talking about the epoxy seal, all I could think was “FLEX SEAL! IT EVEN SEALS TO PERFECT VACUUM”
One of the generic task I do during my daily job is designing of o ring seals for high vacuum chambers for semiconductor tools.
And this video is very precise, informative and entertaining at the same time.
Keep up the good work 🙌🏻
In my job we use stainless steel gaskets for the gas delivery system into the vacuum chamber, the vacuum chamber itself uses rubber gaskets. The leak check we use is helium as well, really cool system.
One of my consulting clients was Flexitallic, they made spiral wound metal gaskets for vacuum applications like this, but they mostly made high positive pressure seals for steam and petrochem.
I'm a vacuum tube engineer and routinely deal with these systems in industry. Great video, but remember, that at 1e-12mBar, you still have about 1 billion molecules per cubic centimeter and I would certainly challenge your statement on 'not a single atom of air can leak through this gasket.' I will agree that below the -9 through -12 decades the vapor pressure of the metals inside the chamber become more significant than the rate of air leaking in, but it is in fact non zero, and even the best vacuum systems still have billions of molecules inside any useful volume. The leak rates that also approach your -12mBar/L/s range also indicate that your leak rate is effectively on the orders of millennia and that even a tiny ion pump can keep up with this nearly unmeasurably small leak rate.
Interesting, is there a reason you do not pressurize to find a leak?
It seems the seals should work in either direction and it would be easier to find a leak from a pressurized system.
You don't call those seals crush washers like we do in other industries?
I was in a lab that had both Ebola and Anthrax... So I thought that was pretty fascinating. What was probably more fascinating was how casually I was invited in there..... The professor who worked there told me that once he retired, he planned on making another lab in his old garden shed. I told him "That sounds absolutely wonderful!"
He might already have anthrax in his garden. It lives in soil just fine.
Great video. Reminds me of when I used to apply "clear bra" plastic to cars, and a single eyelash or piece of dust could mean you were redoing hours of work. Even full Tyvek suits could drop a tiny thread and ruin my entire day. I'll bet the cleanup and keep-clean processes at this lab are unfathomably meticulous. Respect.
I used to work at a pharma site looking after their freeze drying plant. We used the exact same method to find leaks on the vacuum chamber, although we attached an external helium leak detector to the equipment.
My workplace has a fairly decent low mass mass spectrometer, and we've found that Swagelok steel compression fittings can have a leak rate about 1E-15 torr L / s. CF fittings are at most 1E-18 torr L / s. We would have to let a CF fitting 'leak' for over a month to get any measurable results, which we didn't do. For comparison, Swagelok rates their fittings to 1E-9 torr L / s and our high end leak detector (ASM340) goes down to 3.8E-12 torr L / s.
I absolutely love your videos, and I've only seen a few of them. You're not only good at what you do, but you're really passionate about it and very excited about it, and I love it.
Super cool informative video. I've recently been welding stainless fittings for an ultra high vacuum application for a company near my shop that makes xray equipment. Just for their r&d department. Neat to see the process of how they are checking my parts for leakage. All copper washer, knife edge flanges. I have to be really careful about my grounds while welding to make sure I don't exit the flange on the knife edge and cause a blemish on that edge.