A good habit is to always apply a band-pass to recorded audio. Lower end 100 Hz, upper end maybe 16 kHz. Sometimes you get rumble in the recording that is only audible with a subwoofer. Or you might not even be able to hear some sounds when they're too high.
My tinnitus mask anything below a certain tone, so to all those annoyed by the hum, just imagine living with a concert in your ears all the time. I guess this is my way to say: just chill.
35 years ago, I did an internship at a department literally translated to 'leakage test laboratory', that validated parts found in nuclear power plants, or in my case - particle accelerators. We tested components the size of large car down to bench-top size equipments. I remember how i blown away when I first introduced how 1 fingerprint destroys a high vacuum on car size vessel... just due to its outgassing.
I really hoped he'd explain more stuff like this, because it's mindblowing when you find all this stuff about outgassing, diffusion through glass, gasses sticking to metal walls, etc etc.
@@Alexander_Sannikov It is actually really astonishing, that a void, a vacuum is so complicated. Like on the flange fittings, the inner metal part, is to prevent the o-ring from collapsing when pulling a vacuum. Or the metal bellows 'hoses' are the same, but they do contract laterally - which has to be considered when hooking components up, or they just rip apart.
@@Alexander_Sannikov same reason you can not lubricate a bearing - lets say on the ISS Canada arm. The lubricant would A: outgas and evaporate, but also B: freeze in the -200K++ of space. Hence we got Teflon coatings, lubrication free surface finishes for bearing and guides.
@@SarahKchannel Through which medium would the lubricant lose its heat? Is there enough hydrogen, or some other gas, that can absorb the vibrations/heat?
@@Joorin4711 ok, lets me ask you, why would a human freeze to death if exposed to the vacuum of space if there is nothing to absorb the humans heat ? Its thermodynamics.
I would also suggest to get a male DB-9 plug and connect a switch. If it was my chamber I would by now have shortened the wrong pins at least once, possibly ruining something inside the turbo controller...
There is always this pinch of schizophrenia in his apparatuses, where he spends thousands of dollars, only to cheap out on inexpensive, yet vital, components...
I would suggest making up a little dongle out of an old connector for shorting those pins together. A jumper is a great simple solution but it would be a real shame to accidentally short the wrong pins because you got bumped or couldn't see or counted wrong. Great project.
Finally, something I completely understand. :) Brings back my time many years ago working with Mass Spec. Lots of experience working with rough pumps, turbo pumps, diffusion pumps, etc. I had forgotten the one and done copper rings. In a pinch, we could reuse them once, but we had to torque them extra to get a seal.
I visited a scientific research lab some years back that used a very big and long ring tube to make synchrotron light so that need to hold a very good vacuum in the 10E-10 or better. Once they hade one part of it totally shut down (pumps and everything) for a month for a major upgrade of the very big system, but after that was done it still had 10E-9 in it.
@@MoritzvonSchweinitz water vapor adsorbs onto the surfaces and doesn't like to let go. the water molecules can have very strong binding energies with the surface and it takes quite a lot to get them to leave. Heating helps because it boosts the molecule's kinetic energy to the point where it can free itself. Generally 300°C is required to bake out adsorbed water from stainless steel.
@@MoritzvonSchweinitz Its not water like a pool of water, its water like a layer of water on the surface. Imagine it like suggar on a spoon that got wet. Most of the suggar falls right off, but the rest is stuck to it.
Glass of some sort! I assume tempered or borosilicate, but not sure. Can't imagine they are quartz, would cost more than the chamber! 😱They appear to be sealed with a gasket or oring arrangement. Lots of polymer on my chamber heh
I just got a job that has 3 chambers and one is a high vacuum chamber using two turbo pumps and one roughing pump and prior to this job I didn’t know anything about how they work. This video saved my life
Regarding the KF system: It has its origin in Germany which sometimes leads to confusion due to German abbreveations. KF stands for Kleinflansch (= small flange) and NW is just short for Nennweite (= nominal width/size). The latter has nothing to do with the system itself but is universally used with fittings, piping and tubing in general. A neat feature of the KF flanges is the compatibility of two adjacent sizes. KF10 and 16 have identical outside dimensions and differ only by their inner diameter. The same is true for KF20 and 25 and also KF32 and 40. To connect different size flanges you just need intermediate centering rings. Another possibility are outside centering rings which can be used in any configuration, but are not standardized and maybe a little harder to find. Also in case of stainless steel, glas and ceramic fittings you can use special aluminum centering rings with an integral knife edge seal (available from EVAC and others) which then allow KF systems to be baked out and used even in the EUHV range. Unfortunately the colors of o-rings are not standardized and differ greatly betweed manufacturers.
Oh neat, thanks for all the info! That's cool, I didn't know the origins of the naming scheme, or the compatibility of adjacent sizes. That's a super useful tip to know! That's pretty neat trick with the aluminum knife edge centering rings. Pretty clever!
Have you used such metal seals? I'm a bit suspicious about reliability of this connection, especially after multiple bakeouts. KF is not designed to apply such a high pursue needed for metal seal. Is special clamp also needed?
@@swgar No I haven't used them myself. However these seals are offered by many renowned vendors of vacuum equipment which is why I see no reason to question their specifications. Keep in mind that you don't need much force to compress a knife edge made from soft aluminium. Special screw-type clamps are available but as far as I know not strictly neccessary for this type of seal.
I used to work at Laco (the original manufacturer of this chamber). I did not design this chamber but i know the engineer that did. He also has moved on from laco. FYI most centering rings are viton as you said, but buna is the next most common. Glad to know this thing still works. Also this was a fantastic tutorial about vacuum systems.
Ahhh that's so cool! If you ever bump into him say thanks for me! It's a really awesome chamber and the welding/fabrication is fantastic. I was actually looking at Laco chambers in general before this popped up, the price-to-feature ratio on both the custom and off the shelf stuff seems great.
@@BreakingTaps to add to the drama, Redline chambers is a new competitor to Laco with much of the crew from Laco. They're both qualified companies and both can help you if you need weird adapters, etc.
(soper dooper late I know) A good way to think of the Molecular Turbo is like a tennis ball catcher 🎾 5:30 At a low enough vacuum you're not really "sucking" out gasses like sucking the air out of a capri-sun, it's more like you're trying to capture spaced out molecules bouncing around the chamber. 5:58 (clarification: the gas compresses down out of the exhaust at the bottom in the orientation pictured) his pump mounted upside down 5:49 This is why the turbine blades are shaped so flat and have to spin so fast, imagine super fast molecules that reach your vacuum pump inlet trying to bounce everywhere. The angle of the blades at the beginning move the molecule through and out of the pump as it's trying to move back in pressure (concentration) towards the previous stage. 7:10 This need to "catch" random bouncing molecules at the inlet is why he recommends having an inlet of maximum size without any change or narrowed throat
I have a big turbo pump at home (has an 8" flange). The way I got around my lack of a Conflat seal was to take plumber solder that has no core and solder it end to end to create a o-ring, then I just crushed that in the CF fittings. Makes for a good quick and dirty gasket.
I've worked a lot with HV and UHV systems and that looks like a very nice chamber. Absolutely huge windows, I'm also wondering why anyone needs windows that large. Just a note about CF flanges, the copper gasket ideally are used only once as per the manufacturing guidelines. But in reality we could use one of our largest gasket maybe 2 or 3 times. The technique being that you only tighten them enough plus a little extra such that no leak is detectable, we had a mass spec on the system so a He leak check was super sensitive and quick. Of course in a research setting those little gaskets (
Ahh, interesting! Makes sense, I guess you need _just_ enough seal against the copper and can crank it down a few times before you bottom it out (or the copper hardens). Will keep that in mind! I'd love to add an RGA to my system to help with leak checking, but those seem pretty rare and still expensive on ebay. We'll see!
@@BreakingTaps I repaired a He leak detector for work about a year ago. That was a really fun project. The detector came from ebay for cheap and in non-working condition, and I got to take it apart and get it working again. Couldn't find the original manual though, so I had to get the manual for a different model while hoping for the best.
We sometimes re-anneal seals with a torch when we can't find any on the shelf :P copper is fully annealed by the time you get to cherry red, and quenching afterwards won't make any difference.
Yeah, it is definitely reusable several times, I believe I've even seen it in the manual from some manufacturer, but I'll not mess with it unless I really have to and leak detector is highly recommended for that.
Amazing video, recently took apart a CVD chamber with these two pumps, I scavenged the roughing and had a vague idea that there would be a second one in series. It is now way more clear how they setup the turbopump and how all these weird fitting I found are supposed to fit together! Keep it up!!
I worked at a company called "Ionex" for about 6 years. I worked on building new, and rebuilding used vacuum furnaces of any brand and type. Everyone was different and made for special purposes. I worked on some really cool and one of a kind systems. Thanks for making this video.
Love the video, always cool to see people setting up high vacuum systems. Nice to see those old Laco Tech chambers around getting a new life, they did and still do a lot of custom work so a lot of the time the old chambers just get scrapped if the company who bought them goes out of business or replaces them. You got your o-ring materials mixed up. The green ones are either silicone or viton, the brown ones are viton, and the black are Buna-N (probably standard 80 durometer). Flouro and Nitrile materials are generally white. Until you get to very high purity vacuum an elastomer seal like Buna or Viton is perfectly fine. Silicone isn't generally good for high vacuum as it has much higher permeation especially for smaller gasses like helium and hydrogen. An issue you may run into if you want to keep that chamber free of contaminants is the teflon tape you used for some of the threaded connections. Teflon has a tendency to migrate in high-vacuum environments and also permeates like a sieve with smaller elements. Locktite 567 is the sealant we've always used and it's meant for high-heat as well so no worries about baking the chamber. Also the biggest choke in your system is going to be those VAT valves. They look like KF16 connectors which are about the smallest you ever want to use for high vacuum, but the actual valves inside those are even smaller and can add a ton of impedance. That old trivac is really quite small for the system you're pumping and even with it backing a turbo it will definitely struggle, that's probably why you aren't getting even higher vacuum. Ideally if you wanted to get good purity in your chamber you should look at a roots-rotary pump though they get pretty expensive, however a newer rotary-vane pump like a DS series pump from Agilent isn't terribly expensive (relative to how expensive vacuum equipment is generally). Looking forward to seeing what you do with the chamber. -Your friendly neighborhood field service tech
I'm a medical linear accelerator engineer, we had a catastrophic failure of vacuum on one of our treatment linacs last month and OH BOY WAS THAT A FUN MONTH OF CLINICAL DOWNTIME We had to bake the acceleration waveguide for a week to get residual water contamination out of the waveguide. (we lost vacuum on the most humid week of the year so far in the UK, and it was quite contaminatsd! It also killed both ion vacuum pumps, which is a bit shite as it feels like it weighs as much as a car when you're trying to hold it during replacement because they're on really awkward parts to access...
😭That sounds awful, can I anti-heart this comment? I assume those ion pumps are so heavy due to giant magnets? Cool you get to work on equipment like that though! I bet accelerators are some pretty awesome instruments to play around with :)
@@BreakingTaps yeah, I can't remember offhand how much the bastards weigh, but it's a right pain in the backside, especially if it is the gun end of the linac (one at the electron gun/magnetron end of the reletavistic section, and the other near the target puck area (near the output, its what converts the high energy electron beam into x-ray photons. Our machines do 6 to 15 MV, and up to 18 MeV in electron mode. I love that I get to play with this stuff, it's a really rewarding career, as we're making it possible for the clinical staff to treat cancer, and do some other cool stuff (like using the linac in hyper accurate mode, with fiducials on the patient, so we have accurate compensation for chest rise, and then do RF ablation of portions of the heart to fix chronic ventricular tachycardia)) It is also a complete left turn from my previous career and undergrad degree in Sound, Light, and Live Event Technology (applied EEE bachelors for roadies, IET accredited and everything), and the stuff I learnt at uni and in my career as a live sound engineer, is just as valid for my new field of work! And if you think about it, the medical linac is naught but a really really dangerous moving light 😁
@@BreakingTaps imgur.com/a/Q5Xa8vv That's from when we stripped the collimator head and all the gantry shielding (about a metric tonne of weight removed to get to this point) The blue circled bit is one of the two ion pumps, the yellow boot is over the target because neutron activation of the target makes it a bit spicy (well, really only higher than 10MV photons and lots of it, this machine is only clinical at 6MV cos the bunker is smol and its shielding isn't good enough for higher energies)
Couple items for you to help get even better vacuum depths. 1: see if you can acquire un denatured ethanol and use it to wipe down the internal surfaces of the chamber, it will easily vaporize and be removed by the the pump, 2: use a high vacuum grease in a thin layer on each of the KF rings, attached to the chamber and on the connections between hoses, also use on the door seal, do not use on iso flange portion of the conical reducer. 3: reduce hose bends as much as possible, keep any bends to a minimum, to improve pumping down time. These have regularly made it possible to get 60-50nTorr range on some of the cameras at work using the TM pump. We use a cryo pump + TM + roughing when it is time to do optical coatings on the mirrors, with pressure readings of 10^-10 range.
Rad, thanks! Have some anhydrous ethanol on hand since I use it for SEM sample prep, will give everything a good wipe down. Hadn't thought about applying vacuum grease to the KF rings, that's a great tip! I'd like to do some re-arranging to get the roughing pump hose shorter, but the current turbo layout makes that a little difficult. 😢
What's the point to use grease at KF O-rings? It will only introduce extra grease into the system. Proper KF connection provides leak rate of less than 1e-11 mbar·l/s. Though it make sense to wipe it with alcohol since KF clamps produce a lot of metal debris. And it definitely make no sense for forevacuum lines. Below bends also does not looks like an issue, it is not that important for forevacuum. I'll rather check what is the ultimate pressure of the forepump, since this turbo needs sub mbar backing pressure to reach highest compression rate.
a interlock between the butterfly valve on the mouth of the turbo and the valve to repressurize the chamber might be a good addition. Turbos real dont like getting a mouthful of "high" pressure air. Or maybe a indicator light to show the position of the butterfly valve to make sure you don't forget to close it.
@Breaking Taps - *You can maintain Ultra High Vacuum even in a system with soft polymer gaskets by means of **_differential pumping_* of the gap between 2 concentric gaskets/o-rings. The rationale is that the rate of gas leakage depends on the availability of gas, so by pumping out most of the gas between the 2 seals, you slow the leak rate of the seal closest to UHV, usually the inner seal. You usually do not even need a great differential vacuum. If the pressure in between seals is 1 millionth of 1 atmosphere, then your leak rate is reduced by a factor of 1 million. But you have to pump that differential volume continuously or semi-continuously in order to get the best results. Fortunately, a _vacuum accumulator_ for medium/low vacuum is cheap and is adequate to keep the leakage into UHV extremely low even with quite long intervals of _no active pumping._
Brilliant, can't wait to see some interesting experiments come out of this chamber, you might be the first youtuber with such a film-friendly chamber, I hope to see more awesome stuff on your channel that I've never seen before!
That vessel is full of photons and self annihilating particles not to mention the countless neutrinos all the same great job on keeping the atmosphere out!
This is fucking awesome. We used a lab setup like this to test ultra low pressure sensors going on satellites. So cool you can try a bunch of experiments out with it on your own. We had to coordinate with a lab tech to get our test done. So cool man!
AFAIR from some nearly forgotten lecture, the molecular/turbo pump comes into play when the mean free path length becomes so big that air molecules more likely bounce from wall to wall in the tubes than hit each other, so air no longer behaves like a fluid and the molecules have to be helped on their way individually. The chamber looks like it was designed as showcase for some kind of exhibition.
we operate 2x identical HV chambers with 14 m³ volume each and we can reach 5x10e-7 mbar with the turbos attached. We love them to test our ion-thrusters!!
Cool video and I am really looking forward to seeing the projects you are gonna come up with using this huge vacuum chamber! A minor comment on the operational principle of turbomolecular pumps: they do not really compress air, but rather prevent the molecules from ever going back into the chamber once they reach the turbopump. In a very simple way, you can imagine a turbopump as a "one-way membrane", that only passes molecules in one direction (out of the chamber). The reason is the difference in the kind of gas flow you get at different pressures. Up until around 1e-2 mbar you get a so-called laminar flow of gas: the gas behaves as a liquid and molecules move along the pressure gradient (from high pressure areas to low pressure areas). In this kind of gas flow the mean free path of a molecule (an average distance a molecule travels before it interacts with another molecule or a chamber wall) is much smaller than a characteristic dimension of the vacuum chamber (i. e. diameter for a spherical chamber). For most vacuum chambers a pressure in the range of 1e-2 mbar is when the gas flow type changes to the molecular flow: the gas no longer behaves as a liquid but rather a bunch of relatively "free" molecules that are just zooming through the chamber without interacting much. And since they barely interact with one another and the chamber walls, the "pressure" no longer has the same meaning it has in liquid-like gases -- it is simply a measure of volumetric density of molecules (i. e. how many molecules are there in a certain volume). In molecular flow the mean free path is close to or larger than characteristic chamber dimensions. Since the molecules are barely interacting in the molecular flow, roughing pumps such as rotary vane pumps become very inefficient, since they no longer can produce a pressure gradient in such a low pressure (it kind of makes no sense to talk about pressure gradients because, again, molecules are not interacting). Further reducing the pressure becomes a game of chance: we can only hope that molecules in their random motion will reach the inlet flange of the pump and go through it. The problem with rotary vane pumps though is that they are very slow when compared to the speeds with which molecules are zooming around. You can imagine that a molecule that actually gets inside the rotary vane pump can travel tens or hundreds of meters inside it before it even makes one revolution to throw the molecule out of the exhaust port. This means that even though the molecules might reach the inlet port of a rotary vane pump, they have a pretty high chance of escaping it before the pump has a chance to ever "pump out" the molecule. This is why the efficiency of rotary vane pumps at such pressures drops to basically zero. And that's where the turbomolecular pumps get into play! You've said correctly that the rotor blades are rotating very fast and that speed depends on the pump -- especially on it's size! Because the idea is to have a part of the blades to rotate at the linear speeds that are comparable to the average speed of the molecules. Larger pump diameter --> lower RPM. And if you look at the shape of the blades and the way they are positioned, the sole idea is to push the "trapped" molecules down the pump and not letting them escape back into the chamber (hence the high rotational speed -- the molecule basically always has it's "retreat" path blocked by stator and constantly rotation rotor blades). And finally, that is the reason you want to connect the turbopump directly to the chamber at the larges flange possible -- because it means that statistically there is a higher chance that the molecule will hit the inlet flange and therefore get trapped and pumped out of the chamber. I hope at least someone finds this explanation useful. :) Source: I deal with high- and ultrahigh vacuum chambers in my PhD so I've had to learn about them a little bit. In no way am I an expert in vacuum technology though, so take my explanation with a grain of salt.
Very surprising to see windows that large. Windows pose a major challenge to maintaining vacuum/atmosphere quality. Diffusion of oxygen and other gasses through the windows is directly proportional to the surface area of the window(s).
It surprises me how much normally "air tight" materials allow air through...glass! Wow! Thanks for sharing that - I had no idea. So even if all of the seals were perfect in the chamber, air would get through the glass and presumably to a lesser extent through the stainless walls?
Yeah agreed, really odd! Makes me think the prior owners only used this for mid range vacuum experiments or dirty stuff, or like thermal bakeouts of microthrusters or something. Dunno, wish I knew!
The Real Engineering channel just posted a long video about SpinLaunch, where they discussed vacuum diffusion properties for a chamber made of mild steel. Check it out.
Holy smokes, nice rig! I hope you got a lot of this at auction, I'm looking just at the fittings and seeing dollar signs. Cool system and great explanation, thanks!
Thanks! Almost all of it is scrounged from ebay over like the last four years 😁 A few specific adapters were purchased new because I couldn't find them otherwise, but the bulk of it is all surplus gear. Hoses, fittings, turbo, chamber, rotary vane pump, all the electronics and gauges. Took a while to get all the pieces I needed at a reasonable price but if you have enough saved-searches eventually the right thing pops up :)
@@BreakingTaps That's great. Patience always pays off for stuff like this. Have you explored the world of industrial auctions yet? Not sure where you are but in the northeast there are a few auction houses that specialize in auctioning off items from machine shops and labs that are either going out of business or selling off assets. I've found incredible deals on metrology equipment in the past - my biggest hurtle is usually the size of my car and shop.
@@parkermusselman9824 A little, but haven't looked into them too closely yet. Partially for the same reason (car size, and until recently, shop size) and partially because I'd be worried about my own impulse control :) I'm up in northern VT which isn't a great area for industrial stuff, but the southern NH/MA/CT area seems pretty great and not too far away. Might have to start poking around...
@@parkermusselman9824 And shipping! I won a pallet of electronics gear for $250 and it cost me just over $400 for the freight! OUCH! (and that was 5 years ago! no telling what it would cost now!)
That turbo pump is a molecule slinger, relying on chance encounters and not letting the strays back into the chamber. I had an oil diffusion pump in my setup. Some turbopumps won't survive spinup above a certain pressure.
If you trust your welding skills, you could cut a 100mm hole in your box to replace the 50mm one and replace the flange fitting. Then your T-M pump would fit properly and work a bit faster. Of course you'd need an appropriately sized butterfly valve as well.
Hmmm... not sure I trust my welding skills enough for that. But might be good motivation to practice! It definitely pumps a lot slower (2-3x ?) than when I had it mounted to the correctly-sized port on the door.
I'll rather consider to machine an ISO flange without threads for nuts and build a fixture to provide some pressure to the flange and hold pump in place to prevent accidental displacement. Usually once chamber is evacuated an atmospheric pressure is more than enough to provide a proper seal for O-ring flanges. Stainless steel is a bit trickier to weld and even though virtual leaks shoul not be an issue for 1e-6 mbar it's likely you'll create some by improper welding technique.
I needed a vacuum chamber to de gass resin so I made a chamber out of an old metal shopvac with a plexi window in the detachable lid. For a vacuume I hooked up a hose with a 1 way check valve and ran it to the intake manifold of the car. The idling vacuum was pretty good but for more you can just rev the motor and then close the throttle, the 1 way valve lets the vacuum in the chamber drop further and keeps it there. CHEAP but effective.
Impressive system! It was useful for me to know the names of those flange systems, as well as being a generally interesting introduction to high vacuum generation.
Are your large windows protected? I have seen a large diameter glass lid on a vacuum chamber break from a small scratch on its surface and would have quite some respect towards these windows when the chamber is under vacuum. Yes, at first they would implode, but once the high velocity glass shards hit anything they will also fly outwards in all directions...
Not yet, but it's on the short list. Going to put up some kind of polycarbonate window/shield or something in front of it to help catch any potential-grenading 😬
12:57 Green is HNBR, brown is viton, and black is NBR (sometimes viton). Typically. Some compound manufacturers will color compounds however they want though.
This is awesome! Puts my little chambers and diff pumps to shame! I love how your turbo pump is mounted upside-down, is this deliberate to stop junk being able to fall in to it?
I worked at a job for 2 months doing electron beam welding. those chambers were roughly 32"x32"x40" for the small machines, but the big one was 54"x52"x60" the rough vacuume pump was a roots blower with a 10" inlet. the gas diffusion pump on the high vac side had like a gallon of oil or something (I do remember it was $640 a gallon!) Our windows sucked though. it was an 18x12 leaded glass window 4" thick and it was just coated in metal vapor. the lightbulbs inside were basically blacked out from the vapor deposition, but it happened in like 3 days to a new bulb so it was just scraped off with a razor when we needed light the welding was neat too. we could do like 3"-5" of penetration in inconel with a heat affected zone of only 0.025" wish that job had worked out but the management wasn't right for me
Vacuum equipment may be different, but in automotive and i believe most industries, black is nitrile (bnr), green is hydrogenated nitrile (hbnr) and brown is Viton. Nitrile is actually pretty good, but hbnr is quite superior than that for a little more cost, and Viton is even better but much more expensive.
I bet that little port had a ' Pirrani gauge' in it at one time for roughing pressures but what brought back memories for me is the 'one use only' copper gaskets as I can remember University's being so broke they would nip them up enough to seal and that way they could get multiple uses out of them ! as they got older the more you had to 'persuade' the thing to seal by hanging your body weight of it ! . yeah massive windows very cool....cheers.
Haha "persuasion". To be fair, I could 100% see myself doing that too. 😁 Ahh that's a good point about the bottom flange, I bet you're right. Doesn't really make sense for anything else, being so small and isolated down there!
@@BreakingTaps It might be good for electrical feedthroughs if you're operating/monitoring something in the middle and don't want obstructions getting in the way.
I like your setup, I would however swap around your TMP backing valve input/output. My experience has shown that the backside of the plunger should always face the rough pump. it will force the seal to be greater when the foreline is at atmosphere and the chamber is pumped.
The bottom port is probably for some sort of holder to go through and hold things at the center of the chamber. That way it can be in line with the side ports for whatever application.
First, most vacuum pumps are "rotary" vacuum pumps (including the turbomolecular pump). This one is distinguished by the rotating, spring-loaded VANES that sweep gas from the inlet to the exhaust port. Second, turbomolecular pumps DO work at atmospheric pressure, but the rotor faces too much resistance at atmospheric pressure and will quickly burn out the motor. 2 mbar differential pressure makes a great deal more difference when the inlet is at 2 mbar absolute than when the inlet is at 1000 mbar. Black O-rings are Buna-N rubber. Different materials have different heat and chemical properties as well. You should be careful not to scratch those flange faces. Scratches and surface roughness are sources of slow leaks. Experienced people can, and do, re-use copper gaskets a few times, but they do work harden more each time making more torque necessary each time.
I've seen a lot of UHV(ultra high vacuum) system on physics labs. Yes they have a ton of con flat flanges, yes they have to change copper gaskets every time disassembling it, yes they comes down to 10^-15 mbar, which is crucial for experiments that one atom on some light beam's way could ruin days of making that vacuum for that specific experiment.
10e-15 mbar is extreme high vacuum and I doubt that such pressure is reachable in stainless chamber without cooling the whole setup to cryo temperatures and with commercially available pumps. Also I never seen gauges able to measure pressure below 1e-12
Hi , I've been working with high vacuum systems for 40+ years. Ditch the oil pump it's messy and hurts ultimately how low you can go. The turbo pump your using is choked down by the way you have it connected. You need a big port for the pump to chamber and a gate valve to isolate the turbo so it can stay operating when you vent the chamber . Are you using dry N2 to vent? If not the water in the air keeps the vacuum poor for days. How about heated walls for outgassing the chamber? I could go on but that's for starters to improve the system. I've used all types of pumps and systems so I'm kinda an expert. Art
wish I'd seen this when i was setting up my vacuum system... If you ever need an inert glovebox, it's clear you know how to handle this sort of stuff and you're welcome to borrow mine. game. I am sure to learn something.
Real engineering did a video on spin launch system recently and went through different vaccum pumps including turbo pumps.. Good visual representation of how it works..
Those windows make think the chamber was designed for an optical process or to use with optical diagnostics. The port in the door would be great for looking at scattering or fluorescing. Looks really well setup to be a lab tool if you don't need super high vacuum.
Yep! Big o-ring around the edge of the door. I think it needs replacing too, there's a spot that needs a bit of coaxing (and some vacuum grease) to seal up during the initial pump down. I haven't disassembled the windows but they appear to be a gasket or oring as well.
I notic the last time you said this is not expensive was about the rotary vane pump, then onto the fittings..... I actually bought a turbo pump and controller for < £100 and one day I'll get some fittings. Would be interesting to see a homemade thermocouple vacuum gauge vs a store bought one. Beautiful chamber.
I'm really curious whether one could get working turbo for that price. As a first test you could briefly start it in the air. If it rotates at all it is a good sign
I believe it's definitely possible to make a Pirani gauge, likely it'll not work till 1e-4 but something like 0,1-0,01 mbar looks doable. But to calibrate it you'll need a calibrated gauge.
With regard to the turbo mounted to the door, you'd not want to be rotating it relative to the gyro force from the rotors anyhow and so you'd lose your ability to keep it spinning
Make a superalloy using vapor deposition of 5-10 random metals in equal quantities and use hypersonic vibration to randomize the deposition of the depositron reconstition on a freezing condensation accumulator, then use the superalloy to build something cool!
its super weird that i just watched a video about spinlaunch? the project to fire payload into orbit on the cheap and the various stages of pumps and pump types for different levels of vacuum was covered in that video, hat tip to the irish nerd, real engineering, for that video. ;)
Cody’s Lab made a mercury-based vacuum pump that can get all of the way down to, well, the vapor pressure of mercury. It's quite slow though, probably best suited as a third pump. A final step.
The windows had me the most curious, I work with 10^-11 vacuum so you don’t see that I’m curious how much vacuum they can handle I’d assume -6 -7 is close to the max.
Probably! The door and the windows are all sealed with large orings as well, so that's definitely going to be the limiting factor. And glass itself is somewhat permeable if I remember correctly?
full of nothing while being lit? is that possible? not a glib comment,but serious question from an extremely curious BUT ill memoried,uneducated guy on the webs. love your channel period
"hello I have empty metal box!" everyone: Holly damn this is awesome! Oh man... How not to love science channels when you get happy seeing metal box :D
I was taught that, when turning a turbo-molecular pump off, to carefully flood the chamber with air in 2-5min, in order to slow the turbo down quickly and shorten the time it can potentially spend in resonance regions.
How flat are the big windows? Easy to get a rough idea by expanding the beam of a narrow frequency laser and looking at the diffraction patterns where the front and back surface reflections overlap. (He-Ne, or green laser pointer should do it)
Apologies for the electrical hum from 1:10 to 3:50, didn't notice it until after publishing 😬
A good habit is to always apply a band-pass to recorded audio. Lower end 100 Hz, upper end maybe 16 kHz. Sometimes you get rumble in the recording that is only audible with a subwoofer. Or you might not even be able to hear some sounds when they're too high.
It would have been nice to hear the sound of the pumps. Audio is a very important part of any youtube videos, docus, or movies.
I didn't notice it after publishing ;-)
Beat at least all TikTok musics
My tinnitus mask anything below a certain tone, so to all those annoyed by the hum, just imagine living with a concert in your ears all the time. I guess this is my way to say: just chill.
35 years ago, I did an internship at a department literally translated to 'leakage test laboratory', that validated parts found in nuclear power plants, or in my case - particle accelerators.
We tested components the size of large car down to bench-top size equipments.
I remember how i blown away when I first introduced how 1 fingerprint destroys a high vacuum on car size vessel... just due to its outgassing.
I really hoped he'd explain more stuff like this, because it's mindblowing when you find all this stuff about outgassing, diffusion through glass, gasses sticking to metal walls, etc etc.
@@Alexander_Sannikov It is actually really astonishing, that a void, a vacuum is so complicated.
Like on the flange fittings, the inner metal part, is to prevent the o-ring from collapsing when pulling a vacuum.
Or the metal bellows 'hoses' are the same, but they do contract laterally - which has to be considered when hooking
components up, or they just rip apart.
@@Alexander_Sannikov same reason you can not lubricate a bearing - lets say on the ISS Canada arm. The lubricant would A: outgas and evaporate, but also B: freeze in the -200K++ of space. Hence we got Teflon coatings, lubrication free surface finishes for bearing and guides.
@@SarahKchannel Through which medium would the lubricant lose its heat? Is there enough hydrogen, or some other gas, that can absorb the vibrations/heat?
@@Joorin4711 ok, lets me ask you, why would a human freeze to death if exposed to the vacuum of space if there is nothing to absorb the humans heat ? Its thermodynamics.
That little piece of jumper wire in DB-9 connector had me laughing. It looks pretty funny around this whole expensive system.
I would also suggest to get a male DB-9 plug and connect a switch. If it was my chamber I would by now have shortened the wrong pins at least once, possibly ruining something inside the turbo controller...
Yeah that was bit of a confusing part.
yea might need replacing with a plug that has a switch on it lol
There is always this pinch of schizophrenia in his apparatuses, where he spends thousands of dollars, only to cheap out on inexpensive, yet vital, components...
@Olaf Willocx You can actually connect a Palm handheld running management software to the RS-232 port.
I would suggest making up a little dongle out of an old connector for shorting those pins together. A jumper is a great simple solution but it would be a real shame to accidentally short the wrong pins because you got bumped or couldn't see or counted wrong. Great project.
Finally, something I completely understand. :) Brings back my time many years ago working with Mass Spec. Lots of experience working with rough pumps, turbo pumps, diffusion pumps, etc. I had forgotten the one and done copper rings. In a pinch, we could reuse them once, but we had to torque them extra to get a seal.
Hah, I forgot you worked on that stuff! Will have to pick your brain about that :)
I once did a training for ultra high vacuum (
With steel the problem is mostly hydrogen, not the air?
Why is humidity such a problem that you have to heat the system up? Shouldn't any moisture just boil f an vaporize because of the vacuum?
I visited a scientific research lab some years back that used a very big and long ring tube to make synchrotron light so that need to hold a very good vacuum in the 10E-10 or better. Once they hade one part of it totally shut down (pumps and everything) for a month for a major upgrade of the very big system, but after that was done it still had 10E-9 in it.
@@MoritzvonSchweinitz water vapor adsorbs onto the surfaces and doesn't like to let go. the water molecules can have very strong binding energies with the surface and it takes quite a lot to get them to leave. Heating helps because it boosts the molecule's kinetic energy to the point where it can free itself. Generally 300°C is required to bake out adsorbed water from stainless steel.
@@MoritzvonSchweinitz Its not water like a pool of water, its water like a layer of water on the surface. Imagine it like suggar on a spoon that got wet. Most of the suggar falls right off, but the rest is stuck to it.
Those giant windows are amazing! I’ve never seen anything remotely like it in a high vac system. What are they made of?
Amen!
What are they sealed with??
Glass of some sort! I assume tempered or borosilicate, but not sure. Can't imagine they are quartz, would cost more than the chamber! 😱They appear to be sealed with a gasket or oring arrangement. Lots of polymer on my chamber heh
I have also seen polycarbonate windows, not good for UHV
Are they rated for any temperature? I guess limited by the elastomer gasket?
I never realized how big the difference between vacuum and high vacuum is. Thanks for making that clear.
I just got a job that has 3 chambers and one is a high vacuum chamber using two turbo pumps and one roughing pump and prior to this job I didn’t know anything about how they work. This video saved my life
What is the rate of TMPs?
Regarding the KF system: It has its origin in Germany which sometimes leads to confusion due to German abbreveations. KF stands for Kleinflansch (= small flange) and NW is just short for Nennweite (= nominal width/size). The latter has nothing to do with the system itself but is universally used with fittings, piping and tubing in general. A neat feature of the KF flanges is the compatibility of two adjacent sizes. KF10 and 16 have identical outside dimensions and differ only by their inner diameter. The same is true for KF20 and 25 and also KF32 and 40. To connect different size flanges you just need intermediate centering rings. Another possibility are outside centering rings which can be used in any configuration, but are not standardized and maybe a little harder to find. Also in case of stainless steel, glas and ceramic fittings you can use special aluminum centering rings with an integral knife edge seal (available from EVAC and others) which then allow KF systems to be baked out and used even in the EUHV range. Unfortunately the colors of o-rings are not standardized and differ greatly betweed manufacturers.
Oh neat, thanks for all the info! That's cool, I didn't know the origins of the naming scheme, or the compatibility of adjacent sizes. That's a super useful tip to know! That's pretty neat trick with the aluminum knife edge centering rings. Pretty clever!
Have you used such metal seals? I'm a bit suspicious about reliability of this connection, especially after multiple bakeouts. KF is not designed to apply such a high pursue needed for metal seal.
Is special clamp also needed?
@@swgar No I haven't used them myself. However these seals are offered by many renowned vendors of vacuum equipment which is why I see no reason to question their specifications. Keep in mind that you don't need much force to compress a knife edge made from soft aluminium. Special screw-type clamps are available but as far as I know not strictly neccessary for this type of seal.
The Action Lab is shaking in his boots right now. Great video as always :)
I used to work at Laco (the original manufacturer of this chamber).
I did not design this chamber but i know the engineer that did. He also has moved on from laco.
FYI most centering rings are viton as you said, but buna is the next most common.
Glad to know this thing still works.
Also this was a fantastic tutorial about vacuum systems.
Ahhh that's so cool! If you ever bump into him say thanks for me! It's a really awesome chamber and the welding/fabrication is fantastic. I was actually looking at Laco chambers in general before this popped up, the price-to-feature ratio on both the custom and off the shelf stuff seems great.
@@BreakingTaps to add to the drama, Redline chambers is a new competitor to Laco with much of the crew from Laco.
They're both qualified companies and both can help you if you need weird adapters, etc.
@@sungear Good to know! Bookmarking if/when I need to get something custom made. Cheers!
@@BreakingTaps $$$$$$$$$$$$ :)
I swear this guy is doing literally everything I wanna do, he's living my dream lol well at least 35% of my dream
You see all the gray hair though right?
Same for me I really want to do his job
(soper dooper late I know)
A good way to think of the Molecular Turbo is like a tennis ball catcher 🎾
5:30 At a low enough vacuum you're not really "sucking" out gasses like sucking the air out of a capri-sun, it's more like you're trying to capture spaced out molecules bouncing around the chamber.
5:58 (clarification: the gas compresses down out of the exhaust at the bottom in the orientation pictured) his pump mounted upside down
5:49 This is why the turbine blades are shaped so flat and have to spin so fast, imagine super fast molecules that reach your vacuum pump inlet trying to bounce everywhere. The angle of the blades at the beginning move the molecule through and out of the pump as it's trying to move back in pressure (concentration) towards the previous stage.
7:10 This need to "catch" random bouncing molecules at the inlet is why he recommends having an inlet of maximum size without any change or narrowed throat
I have a big turbo pump at home (has an 8" flange). The way I got around my lack of a Conflat seal was to take plumber solder that has no core and solder it end to end to create a o-ring, then I just crushed that in the CF fittings. Makes for a good quick and dirty gasket.
I've worked a lot with HV and UHV systems and that looks like a very nice chamber. Absolutely huge windows, I'm also wondering why anyone needs windows that large.
Just a note about CF flanges, the copper gasket ideally are used only once as per the manufacturing guidelines. But in reality we could use one of our largest gasket maybe 2 or 3 times. The technique being that you only tighten them enough plus a little extra such that no leak is detectable, we had a mass spec on the system so a He leak check was super sensitive and quick. Of course in a research setting those little gaskets (
Ahh, interesting! Makes sense, I guess you need _just_ enough seal against the copper and can crank it down a few times before you bottom it out (or the copper hardens). Will keep that in mind!
I'd love to add an RGA to my system to help with leak checking, but those seem pretty rare and still expensive on ebay. We'll see!
@@BreakingTaps I repaired a He leak detector for work about a year ago. That was a really fun project. The detector came from ebay for cheap and in non-working condition, and I got to take it apart and get it working again. Couldn't find the original manual though, so I had to get the manual for a different model while hoping for the best.
We sometimes re-anneal seals with a torch when we can't find any on the shelf :P copper is fully annealed by the time you get to cherry red, and quenching afterwards won't make any difference.
Yeah, it is definitely reusable several times, I believe I've even seen it in the manual from some manufacturer, but I'll not mess with it unless I really have to and leak detector is highly recommended for that.
"Full of Nothing" is how my cat describes herself when it's five minutes past her feeding time
Hahahah
Time to test vacuum chamber with the cat
Gura nyaa
Super relatable (i don't own a cat tho)
these god damned animals are the most insane, intense borderline having pieces of shit and i would die for them
Amazing video, recently took apart a CVD chamber with these two pumps, I scavenged the roughing and had a vague idea that there would be a second one in series. It is now way more clear how they setup the turbopump and how all these weird fitting I found are supposed to fit together!
Keep it up!!
What a way to start a Monday!
I worked at a company called "Ionex" for about 6 years. I worked on building new, and rebuilding used vacuum furnaces of any brand and type. Everyone was different and made for special purposes. I worked on some really cool and one of a kind systems. Thanks for making this video.
This is so impressive! Thanks for explaining it all.
Love the video, always cool to see people setting up high vacuum systems. Nice to see those old Laco Tech chambers around getting a new life, they did and still do a lot of custom work so a lot of the time the old chambers just get scrapped if the company who bought them goes out of business or replaces them.
You got your o-ring materials mixed up. The green ones are either silicone or viton, the brown ones are viton, and the black are Buna-N (probably standard 80 durometer). Flouro and Nitrile materials are generally white.
Until you get to very high purity vacuum an elastomer seal like Buna or Viton is perfectly fine. Silicone isn't generally good for high vacuum as it has much higher permeation especially for smaller gasses like helium and hydrogen.
An issue you may run into if you want to keep that chamber free of contaminants is the teflon tape you used for some of the threaded connections. Teflon has a tendency to migrate in high-vacuum environments and also permeates like a sieve with smaller elements. Locktite 567 is the sealant we've always used and it's meant for high-heat as well so no worries about baking the chamber.
Also the biggest choke in your system is going to be those VAT valves. They look like KF16 connectors which are about the smallest you ever want to use for high vacuum, but the actual valves inside those are even smaller and can add a ton of impedance.
That old trivac is really quite small for the system you're pumping and even with it backing a turbo it will definitely struggle, that's probably why you aren't getting even higher vacuum. Ideally if you wanted to get good purity in your chamber you should look at a roots-rotary pump though they get pretty expensive, however a newer rotary-vane pump like a DS series pump from Agilent isn't terribly expensive (relative to how expensive vacuum equipment is generally).
Looking forward to seeing what you do with the chamber.
-Your friendly neighborhood field service tech
I'm a medical linear accelerator engineer, we had a catastrophic failure of vacuum on one of our treatment linacs last month and OH BOY WAS THAT A FUN MONTH OF CLINICAL DOWNTIME
We had to bake the acceleration waveguide for a week to get residual water contamination out of the waveguide. (we lost vacuum on the most humid week of the year so far in the UK, and it was quite contaminatsd!
It also killed both ion vacuum pumps, which is a bit shite as it feels like it weighs as much as a car when you're trying to hold it during replacement because they're on really awkward parts to access...
We start off with the combination roughing/turbomolecular pump and then switch over to the ion pumps to maintain and achieve ultra high vacuum
😭That sounds awful, can I anti-heart this comment? I assume those ion pumps are so heavy due to giant magnets? Cool you get to work on equipment like that though! I bet accelerators are some pretty awesome instruments to play around with :)
@@BreakingTaps yeah, I can't remember offhand how much the bastards weigh, but it's a right pain in the backside, especially if it is the gun end of the linac (one at the electron gun/magnetron end of the reletavistic section, and the other near the target puck area (near the output, its what converts the high energy electron beam into x-ray photons. Our machines do 6 to 15 MV, and up to 18 MeV in electron mode.
I love that I get to play with this stuff, it's a really rewarding career, as we're making it possible for the clinical staff to treat cancer, and do some other cool stuff (like using the linac in hyper accurate mode, with fiducials on the patient, so we have accurate compensation for chest rise, and then do RF ablation of portions of the heart to fix chronic ventricular tachycardia))
It is also a complete left turn from my previous career and undergrad degree in Sound, Light, and Live Event Technology (applied EEE bachelors for roadies, IET accredited and everything), and the stuff I learnt at uni and in my career as a live sound engineer, is just as valid for my new field of work! And if you think about it, the medical linac is naught but a really really dangerous moving light 😁
@@BreakingTaps imgur.com/a/Q5Xa8vv
That's from when we stripped the collimator head and all the gantry shielding (about a metric tonne of weight removed to get to this point)
The blue circled bit is one of the two ion pumps, the yellow boot is over the target because neutron activation of the target makes it a bit spicy (well, really only higher than 10MV photons and lots of it, this machine is only clinical at 6MV cos the bunker is smol and its shielding isn't good enough for higher energies)
@@keleighshepherd345 Interesting that you would get neutrons from interaction of electron beams with the X-ray target.
Couple items for you to help get even better vacuum depths. 1: see if you can acquire un denatured ethanol and use it to wipe down the internal surfaces of the chamber, it will easily vaporize and be removed by the the pump, 2: use a high vacuum grease in a thin layer on each of the KF rings, attached to the chamber and on the connections between hoses, also use on the door seal, do not use on iso flange portion of the conical reducer.
3: reduce hose bends as much as possible, keep any bends to a minimum, to improve pumping down time.
These have regularly made it possible to get 60-50nTorr range on some of the cameras at work using the TM pump. We use a cryo pump + TM + roughing when it is time to do optical coatings on the mirrors, with pressure readings of 10^-10 range.
Rad, thanks! Have some anhydrous ethanol on hand since I use it for SEM sample prep, will give everything a good wipe down. Hadn't thought about applying vacuum grease to the KF rings, that's a great tip! I'd like to do some re-arranging to get the roughing pump hose shorter, but the current turbo layout makes that a little difficult. 😢
What's the point to use grease at KF O-rings? It will only introduce extra grease into the system. Proper KF connection provides leak rate of less than 1e-11 mbar·l/s. Though it make sense to wipe it with alcohol since KF clamps produce a lot of metal debris. And it definitely make no sense for forevacuum lines.
Below bends also does not looks like an issue, it is not that important for forevacuum.
I'll rather check what is the ultimate pressure of the forepump, since this turbo needs sub mbar backing pressure to reach highest compression rate.
a interlock between the butterfly valve on the mouth of the turbo and the valve to repressurize the chamber might be a good addition. Turbos real dont like getting a mouthful of "high" pressure air. Or maybe a indicator light to show the position of the butterfly valve to make sure you don't forget to close it.
You have a very good intro to vacuum chamber and pump variations with turbo and various connections and seals.. Enjoyed it very much.
@Breaking Taps - *You can maintain Ultra High Vacuum even in a system with soft polymer gaskets by means of **_differential pumping_* of the gap between 2 concentric gaskets/o-rings. The rationale is that the rate of gas leakage depends on the availability of gas, so by pumping out most of the gas between the 2 seals, you slow the leak rate of the seal closest to UHV, usually the inner seal. You usually do not even need a great differential vacuum. If the pressure in between seals is 1 millionth of 1 atmosphere, then your leak rate is reduced by a factor of 1 million. But you have to pump that differential volume continuously or semi-continuously in order to get the best results. Fortunately, a _vacuum accumulator_ for medium/low vacuum is cheap and is adequate to keep the leakage into UHV extremely low even with quite long intervals of _no active pumping._
Brilliant, can't wait to see some interesting experiments come out of this chamber, you might be the first youtuber with such a film-friendly chamber, I hope to see more awesome stuff on your channel that I've never seen before!
That vessel is full of photons and self annihilating particles not to mention the countless neutrinos all the same great job on keeping the atmosphere out!
This is fucking awesome. We used a lab setup like this to test ultra low pressure sensors going on satellites. So cool you can try a bunch of experiments out with it on your own. We had to coordinate with a lab tech to get our test done. So cool man!
AFAIR from some nearly forgotten lecture, the molecular/turbo pump comes into play when the mean free path length becomes so big that air molecules more likely bounce from wall to wall in the tubes than hit each other, so air no longer behaves like a fluid and the molecules have to be helped on their way individually.
The chamber looks like it was designed as showcase for some kind of exhibition.
we operate 2x identical HV chambers with 14 m³ volume each and we can reach 5x10e-7 mbar with the turbos attached. We love them to test our ion-thrusters!!
Looks like fun. Deep vacuum is fascinating and more often than not frustrating.
Cool video and I am really looking forward to seeing the projects you are gonna come up with using this huge vacuum chamber! A minor comment on the operational principle of turbomolecular pumps: they do not really compress air, but rather prevent the molecules from ever going back into the chamber once they reach the turbopump. In a very simple way, you can imagine a turbopump as a "one-way membrane", that only passes molecules in one direction (out of the chamber). The reason is the difference in the kind of gas flow you get at different pressures. Up until around 1e-2 mbar you get a so-called laminar flow of gas: the gas behaves as a liquid and molecules move along the pressure gradient (from high pressure areas to low pressure areas). In this kind of gas flow the mean free path of a molecule (an average distance a molecule travels before it interacts with another molecule or a chamber wall) is much smaller than a characteristic dimension of the vacuum chamber (i. e. diameter for a spherical chamber). For most vacuum chambers a pressure in the range of 1e-2 mbar is when the gas flow type changes to the molecular flow: the gas no longer behaves as a liquid but rather a bunch of relatively "free" molecules that are just zooming through the chamber without interacting much. And since they barely interact with one another and the chamber walls, the "pressure" no longer has the same meaning it has in liquid-like gases -- it is simply a measure of volumetric density of molecules (i. e. how many molecules are there in a certain volume). In molecular flow the mean free path is close to or larger than characteristic chamber dimensions.
Since the molecules are barely interacting in the molecular flow, roughing pumps such as rotary vane pumps become very inefficient, since they no longer can produce a pressure gradient in such a low pressure (it kind of makes no sense to talk about pressure gradients because, again, molecules are not interacting). Further reducing the pressure becomes a game of chance: we can only hope that molecules in their random motion will reach the inlet flange of the pump and go through it. The problem with rotary vane pumps though is that they are very slow when compared to the speeds with which molecules are zooming around. You can imagine that a molecule that actually gets inside the rotary vane pump can travel tens or hundreds of meters inside it before it even makes one revolution to throw the molecule out of the exhaust port. This means that even though the molecules might reach the inlet port of a rotary vane pump, they have a pretty high chance of escaping it before the pump has a chance to ever "pump out" the molecule. This is why the efficiency of rotary vane pumps at such pressures drops to basically zero.
And that's where the turbomolecular pumps get into play! You've said correctly that the rotor blades are rotating very fast and that speed depends on the pump -- especially on it's size! Because the idea is to have a part of the blades to rotate at the linear speeds that are comparable to the average speed of the molecules. Larger pump diameter --> lower RPM. And if you look at the shape of the blades and the way they are positioned, the sole idea is to push the "trapped" molecules down the pump and not letting them escape back into the chamber (hence the high rotational speed -- the molecule basically always has it's "retreat" path blocked by stator and constantly rotation rotor blades). And finally, that is the reason you want to connect the turbopump directly to the chamber at the larges flange possible -- because it means that statistically there is a higher chance that the molecule will hit the inlet flange and therefore get trapped and pumped out of the chamber.
I hope at least someone finds this explanation useful. :)
Source: I deal with high- and ultrahigh vacuum chambers in my PhD so I've had to learn about them a little bit. In no way am I an expert in vacuum technology though, so take my explanation with a grain of salt.
That Jumper Cable Made My Day !
Very surprising to see windows that large. Windows pose a major challenge to maintaining vacuum/atmosphere quality. Diffusion of oxygen and other gasses through the windows is directly proportional to the surface area of the window(s).
It surprises me how much normally "air tight" materials allow air through...glass! Wow! Thanks for sharing that - I had no idea. So even if all of the seals were perfect in the chamber, air would get through the glass and presumably to a lesser extent through the stainless walls?
Yeah agreed, really odd! Makes me think the prior owners only used this for mid range vacuum experiments or dirty stuff, or like thermal bakeouts of microthrusters or something. Dunno, wish I knew!
The Real Engineering channel just posted a long video about SpinLaunch, where they discussed vacuum diffusion properties for a chamber made of mild steel. Check it out.
This one will make for much better videography though :-D
@@BreakingTaps Are you going to install LED lighting inside the chamber for illumination?
Holy smokes, nice rig! I hope you got a lot of this at auction, I'm looking just at the fittings and seeing dollar signs. Cool system and great explanation, thanks!
Thanks! Almost all of it is scrounged from ebay over like the last four years 😁 A few specific adapters were purchased new because I couldn't find them otherwise, but the bulk of it is all surplus gear. Hoses, fittings, turbo, chamber, rotary vane pump, all the electronics and gauges. Took a while to get all the pieces I needed at a reasonable price but if you have enough saved-searches eventually the right thing pops up :)
@@BreakingTaps That's great. Patience always pays off for stuff like this. Have you explored the world of industrial auctions yet? Not sure where you are but in the northeast there are a few auction houses that specialize in auctioning off items from machine shops and labs that are either going out of business or selling off assets. I've found incredible deals on metrology equipment in the past - my biggest hurtle is usually the size of my car and shop.
@@parkermusselman9824 A little, but haven't looked into them too closely yet. Partially for the same reason (car size, and until recently, shop size) and partially because I'd be worried about my own impulse control :)
I'm up in northern VT which isn't a great area for industrial stuff, but the southern NH/MA/CT area seems pretty great and not too far away. Might have to start poking around...
@@BreakingTaps I shouldn't enable but when I was living in the Boston area I signed up for the Joseph Finn co. and Branford Group mailing lists.
@@parkermusselman9824 And shipping! I won a pallet of electronics gear for $250 and it cost me just over $400 for the freight! OUCH! (and that was 5 years ago! no telling what it would cost now!)
Loved the breakdown, looking forward to whatever projects use this!
i just stumbled upon this, have no knowledge in vacuum systems but the hardware looks like alot of fun
That turbo pump is a molecule slinger, relying on chance encounters and not letting the strays back into the chamber. I had an oil diffusion pump in my setup. Some turbopumps won't survive spinup above a certain pressure.
Yeah, I've never seen a chamber with a pump on the top though.
@@Les_Grossman That would be expensive, to be sure!
@@derkeksinator17 We use them vertically in the lab I work in. Zero issue.
Wish this video came out 5 years ago when I had to start learning a lot of this!
If you trust your welding skills, you could cut a 100mm hole in your box to replace the 50mm one and replace the flange fitting. Then your T-M pump would fit properly and work a bit faster. Of course you'd need an appropriately sized butterfly valve as well.
Hmmm... not sure I trust my welding skills enough for that. But might be good motivation to practice! It definitely pumps a lot slower (2-3x ?) than when I had it mounted to the correctly-sized port on the door.
I'll rather consider to machine an ISO flange without threads for nuts and build a fixture to provide some pressure to the flange and hold pump in place to prevent accidental displacement. Usually once chamber is evacuated an atmospheric pressure is more than enough to provide a proper seal for O-ring flanges.
Stainless steel is a bit trickier to weld and even though virtual leaks shoul not be an issue for 1e-6 mbar it's likely you'll create some by improper welding technique.
I needed a vacuum chamber to de gass resin so I made a chamber out of an old metal shopvac with a plexi window in the detachable lid. For a vacuume I hooked up a hose with a 1 way check valve and ran it to the intake manifold of the car. The idling vacuum was pretty good but for more you can just rev the motor and then close the throttle, the 1 way valve lets the vacuum in the chamber drop further and keeps it there. CHEAP but effective.
Cool. Can't wait to see the projects you have lined up.
Impressive system! It was useful for me to know the names of those flange systems, as well as being a generally interesting introduction to high vacuum generation.
Are your large windows protected? I have seen a large diameter glass lid on a vacuum chamber break from a small scratch on its surface and would have quite some respect towards these windows when the chamber is under vacuum. Yes, at first they would implode, but once the high velocity glass shards hit anything they will also fly outwards in all directions...
Not yet, but it's on the short list. Going to put up some kind of polycarbonate window/shield or something in front of it to help catch any potential-grenading 😬
@@BreakingTaps Try to get AR coated polycarbonate or a thick AR coated window glass. Good to preserve the optical qualities of the windows.
12:57
Green is HNBR, brown is viton, and black is NBR (sometimes viton). Typically. Some compound manufacturers will color compounds however they want though.
For vacuum equipment mainly Viton is used, it could be any color
This is awesome! Puts my little chambers and diff pumps to shame! I love how your turbo pump is mounted upside-down, is this deliberate to stop junk being able to fall in to it?
The thumbnail looks like something This Old Tony would make. Cool video.
I worked at a job for 2 months doing electron beam welding. those chambers were roughly 32"x32"x40" for the small machines, but the big one was 54"x52"x60"
the rough vacuume pump was a roots blower with a 10" inlet. the gas diffusion pump on the high vac side had like a gallon of oil or something (I do remember it was $640 a gallon!)
Our windows sucked though. it was an 18x12 leaded glass window 4" thick and it was just coated in metal vapor. the lightbulbs inside were basically blacked out from the vapor deposition, but it happened in like 3 days to a new bulb so it was just scraped off with a razor when we needed light
the welding was neat too. we could do like 3"-5" of penetration in inconel with a heat affected zone of only 0.025"
wish that job had worked out but the management wasn't right for me
Vacuum equipment may be different, but in automotive and i believe most industries, black is nitrile (bnr), green is hydrogenated nitrile (hbnr) and brown is Viton. Nitrile is actually pretty good, but hbnr is quite superior than that for a little more cost, and Viton is even better but much more expensive.
Aha! That very well could be it, I was being lazy and didn't look it up ahead of time 😅 Cheers!
It's likely to be different. 99 % of rubber-like vacuum seals are claimed to be Viton, it could be of any color
This channel went from Breaking Taps to Braking Physics. Staying subbed. This is awesome.
Those fittings remind me of the tri-clamp fittings I use with my home distillation equipment. The o-rings are made of ptfe.
Amazing bit of kit, I'm really looking forward to seeing ionized gas in there.
I bet that little port had a ' Pirrani gauge' in it at one time for roughing pressures but what brought back memories for me is the 'one use only' copper gaskets as I can remember University's being so broke they would nip them up enough to seal and that way they could get multiple uses out of them ! as they got older the more you had to 'persuade' the thing to seal by hanging your body weight of it ! . yeah massive windows very cool....cheers.
Haha "persuasion". To be fair, I could 100% see myself doing that too. 😁
Ahh that's a good point about the bottom flange, I bet you're right. Doesn't really make sense for anything else, being so small and isolated down there!
@@BreakingTaps It might be good for electrical feedthroughs if you're operating/monitoring something in the middle and don't want obstructions getting in the way.
I like your setup, I would however swap around your TMP backing valve input/output. My experience has shown that the backside of the plunger should always face the rough pump. it will force the seal to be greater when the foreline is at atmosphere and the chamber is pumped.
Show us more ultra low vaccum stuff! Diggin this ❤️😊
The bottom port is probably for some sort of holder to go through and hold things at the center of the chamber. That way it can be in line with the side ports for whatever application.
Aha, that would make sense!
Alpha Phoenix has a great video on CF flanges and ultra vacuum systems. Great video Mr. Taps.
I’ve rebuilt that model of trivac. It’s not horrible. Basically all the older vacuum pumps are simpler and easier.
That's good to know! I think mine is in dire need of some maintenance so I've been working up the courage to figure it out :)
Cool setup. A vacuum tube developer's dream!
Great basic explanation of the chamber and it's functions. Can't wait to see it in action in the future. Thanks for posting and take care!
that is the coolest vacuum chamber i have ever seen! i look forward to some neat videos!
dude, you're so close to 100k subs. that's awesome!
🥳
First, most vacuum pumps are "rotary" vacuum pumps (including the turbomolecular pump). This one is distinguished by the rotating, spring-loaded VANES that sweep gas from the inlet to the exhaust port. Second, turbomolecular pumps DO work at atmospheric pressure, but the rotor faces too much resistance at atmospheric pressure and will quickly burn out the motor. 2 mbar differential pressure makes a great deal more difference when the inlet is at 2 mbar absolute than when the inlet is at 1000 mbar.
Black O-rings are Buna-N rubber. Different materials have different heat and chemical properties as well. You should be careful not to scratch those flange faces. Scratches and surface roughness are sources of slow leaks. Experienced people can, and do, re-use copper gaskets a few times, but they do work harden more each time making more torque necessary each time.
I've seen a lot of UHV(ultra high vacuum) system on physics labs. Yes they have a ton of con flat flanges, yes they have to change copper gaskets every time disassembling it, yes they comes down to 10^-15 mbar, which is crucial for experiments that one atom on some light beam's way could ruin days of making that vacuum for that specific experiment.
10e-15 mbar is extreme high vacuum and I doubt that such pressure is reachable in stainless chamber without cooling the whole setup to cryo temperatures and with commercially available pumps. Also I never seen gauges able to measure pressure below 1e-12
Hi , I've been working with high vacuum systems for 40+ years.
Ditch the oil pump it's messy and hurts ultimately how low you can go.
The turbo pump your using is choked down by the way you have it connected. You need a big port for the pump to chamber and a gate valve to isolate the turbo so it can stay operating when you vent the chamber . Are you using dry N2 to vent? If not the water in the air keeps the vacuum poor for days. How about heated walls for outgassing the chamber? I could go on but that's for starters to improve the system.
I've used all types of pumps and systems so I'm kinda an expert. Art
Awesome Video, your workshop impresses me every time 😄
wish I'd seen this when i was setting up my vacuum system...
If you ever need an inert glovebox, it's clear you know how to handle this sort of stuff and you're welcome to borrow mine. game. I am sure to learn something.
Fascinating. Utterly outside my field of expertise. Subscribed.
Real engineering did a video on spin launch system recently and went through different vaccum pumps including turbo pumps.. Good visual representation of how it works..
Thank you for your excellent explanations
Spectacular bit of kit!
Those windows make think the chamber was designed for an optical process or to use with optical diagnostics. The port in the door would be great for looking at scattering or fluorescing. Looks really well setup to be a lab tool if you don't need super high vacuum.
Those windows have to be really thick based on their huge size. Really good score!
I assume the windows and the door use some sort of rubber gaskets as well? That door seems like it would be a nightmare to get to seal well.
Yep! Big o-ring around the edge of the door. I think it needs replacing too, there's a spot that needs a bit of coaxing (and some vacuum grease) to seal up during the initial pump down. I haven't disassembled the windows but they appear to be a gasket or oring as well.
That mass flow controller is actually a rebranded Bronkhorst EL-FLOW, although its old, it's a very good instrument.
I love this channel, pumped for what’s next
I notic the last time you said this is not expensive was about the rotary vane pump, then onto the fittings..... I actually bought a turbo pump and controller for < £100 and one day I'll get some fittings. Would be interesting to see a homemade thermocouple vacuum gauge vs a store bought one. Beautiful chamber.
I'm really curious whether one could get working turbo for that price. As a first test you could briefly start it in the air. If it rotates at all it is a good sign
I believe it's definitely possible to make a Pirani gauge, likely it'll not work till 1e-4 but something like 0,1-0,01 mbar looks doable.
But to calibrate it you'll need a calibrated gauge.
Very cool !. Could be cool to look at light scattering from "Smoke" particles with and without electric field.
With regard to the turbo mounted to the door, you'd not want to be rotating it relative to the gyro force from the rotors anyhow and so you'd lose your ability to keep it spinning
Aha, I hadn't even thought of that. Yeah definitely would not have wanted to move it while spinning!
Make a superalloy using vapor deposition of 5-10 random metals in equal quantities and use hypersonic vibration to randomize the deposition of the depositron reconstition on a freezing condensation accumulator, then use the superalloy to build something cool!
Recent video on High Entropy Alloys might interest you :)
thanks, an invaluable touchstone
its super weird that i just watched a video about spinlaunch? the project to fire payload into orbit on the cheap and the various stages of pumps and pump types for different levels of vacuum was covered in that video, hat tip to the irish nerd, real engineering, for that video. ;)
impressive project 😃
thanks for sharing your expirences with All of us 👍😃
If you put heater into the chamber and heat up to 100 degree C, it will help you get 10^-8 level vacuume.
Can't wait to see what you do with it! Cheers...
Thank you for sharing all of these details!
Cody’s Lab made a mercury-based vacuum pump that can get all of the way down to, well, the vapor pressure of mercury. It's quite slow though, probably best suited as a third pump. A final step.
The window in front of your vacuum chamber is relatively large. If the front glass is not very thick, it may break, causing much damage.
The windows had me the most curious, I work with 10^-11 vacuum so you don’t see that I’m curious how much vacuum they can handle I’d assume -6 -7 is close to the max.
Probably! The door and the windows are all sealed with large orings as well, so that's definitely going to be the limiting factor. And glass itself is somewhat permeable if I remember correctly?
16:45 Get a D9 male socket and solder on a switch or short the pins and use inserting the socket as a switch.
full of nothing while being lit? is that possible? not a glib comment,but serious question from an extremely curious BUT ill memoried,uneducated guy on the webs.
love your channel period
"hello I have empty metal box!"
everyone: Holly damn this is awesome!
Oh man... How not to love science channels when you get happy seeing metal box :D
I was taught that, when turning a turbo-molecular pump off, to carefully flood the chamber with air in 2-5min, in order to slow the turbo down quickly and shorten the time it can potentially spend in resonance regions.
Mine makes a horrid howl as it spins down but the manufacturer said it is normal and won't hurt it at all. Still very irritating!
Not a bad idea. Harmonics are not good for any high rotating masses, Regardless if a manufacturer says it’s fine.-I used to work for Varian
Very cool. I look forward to seeing experiments!
i think TC is quite common for the "tapered edge flange" as well. i think it originates from tri-clamp.
Very well explained.
How flat are the big windows? Easy to get a rough idea by expanding the beam of a narrow frequency laser and looking at the diffraction patterns where the front and back surface reflections overlap. (He-Ne, or green laser pointer should do it)
Lovely filmed and well explained! Thanks for sharing all of this :)