Giant glass diffusion pump and cathode ray tube demo

today on applied science i'm going to show you a demo of this giant glass diffusion pump this is a specialized piece of vacuum equipment that's usually made out of metal but this one is made out of glass so we can see how it works on the inside and so in today's video i'm going to show the details of that why you might need a diffusion pump and then also connect it up to a homemade cathode ray tube that i made out of some lab glassware the diffusion pump works by boiling some fluid in the bottom here and the hot vapor from this boiling travels up the middle of the pump and gets shot out these jets so it's forcing the fluid vapor downwards toward the outside of the pump here and the way this works is it pushes the gas molecules out so the input is actually at the top here where we want the best vacuum to be and the output is over here where the ends of all these vapor jets are so this is very similar to like blowing away dry leaves with the spray from a garden hose almost exactly the same thing you basically want to push the leaves away or the gas molecules away and we're using a stream of something to do it um they're deceptively simple there's no moving parts on the inside all the work is done entirely by these moving gas molecules or by these moving vapor molecules of this special fluid so in the early days they actually used mercury because that actually happens to have a really convenient boiling point and mercury is heavy so you want something that's going to push those dry leaves out of the way so you want something fairly massive obviously boiling mercury is not so great so nowadays we use either silicone oil or even just hydrocarbon oil that has a boiling point that's just about right so let's back up for a minute and think why we would actually want to build a vacuum pump this way like why wouldn't we just use a better and better mechanical vacuum pump when you think of vacuum pumps this is probably what you think of a mechanical pump and these all work in the basic same way and they work very well but only down to a point where we hit a fundamental limit it's not actually a technological limit of building better seals or anything like that it's actually fundamental and i'll show you why so pretend that the plastic beads in here are the air molecules and we want to evacuate this jar of them to ignore gravity i'm going to hold this thing horizontal and shake it around to sort of simulate the motion in there and the way these mechanical pumps work is they're all the same we basically want to connect something up to this volume make the volume bigger withdrawing some of the gas and then expelling it and just doing this over and over to pump it out so let's give it a shot here i'm just going to put this on with some tape and to ignore gravity i'm just going to hold this horizontal obviously you can't tip it like this because that's cheating we're using gravity so i'm going to hold it horizontal and try to shake it randomly and withdraw some gas [Music] so as you can see not surprisingly we got some in so we can you know disconnect from that and those are the gas molecules that we got out now let's try this again but only have one gas molecule in here okay so we've got our one molecule in there now and we'll connect up in the same way shake this thing around and if we withdraw the plunger nothing's happening so it's not pumping out so let's we could even disconnect this and it eventually comes out but it took a while and that's the key thing you could basically have the best vacuum pump in the world and connect it up to here and if there's only one or two molecules in here that aren't colliding with other gas molecules all you can do is wait for the random chance collisions to basically collide with the walls and the gas molecule will eventually work its way out let's try that again we'll put in one i am trying to shake it randomly and it takes a while but it eventually comes out so when you're dealing with super low pressures all you can do is actually wait for the gas molecules to bounce their way out and then your vacuum pump actually becomes like a collector like you basically wait for the molecule to come in and then the only thing you can do is trap it or push it out there's no point in sort of pulling anymore because the molecules aren't forming this continuous sort of fluid so as you can see from the demo this pump does have a design problem it's got a small mouth so you can see the body of the pump is good it's a large diameter but the restriction up here is a pretty big problem i mean this the body of the pump could be connected to deep space and we still wouldn't be able to pump everything out of here in a good way because of this restriction so that this pump really is made much more for demonstration than actual use and if we look at real diffusion pumps that are made for you know serious work they're made out of metal and the mouth is in fact as large as the pump diameter in fact it's as big as it can possibly be and the cool thing is this scales up very well if you have a huge vacuum chamber like the size of a room for you know huge production jobs you can build a diffusion pump that's also as big of a room as big as a room and it still works the same so then the question arises why do we need two pumps why can't we just build a diffusion pump that's big enough to handle the whole job by itself and i think that fundamentally there's no reason it can't work but think again of the dry leaves analogy blowing away dry leaves works great if you just have a few of them on the ground but imagine if you had ten thousand tons of dry leaves all stacked up in a pile the size water jet you need to push them around is you know way outside of practicality so i don't think there's a fundamental reason it's just technically not something that's really done because it's just not efficient let me show you the internal structures of this vacuum pump and then we'll fire it up at the top here we have a cooling coil with water running through it and the purpose of this is to catch vapor molecules that aren't well controlled by the bottom of the pump so we're boiling vapor in there or boiling this oil and we expect the jets to keep it very well controlled and condensed down below but if any of that oil makes it up into our experiment that'd be very bad it would actually poison or contaminate our vacuum experiment so this cooling coil up here is kind of a backstop of a safety of last resort if some of the oil gets up here the coil actually makes an optically opaque barrier between the input and the pumping area of this pump and so if any of these stray oil molecules go up there remember at these super low pressures the molecules hit the walls before they hit another air molecule so if we make something that's optically opaque then they have to contact the cold coils and get condensed before they make it out of the pump next there's a cold cap it's basically an upside down cup that has cooling water also running through it and the purpose of this is to keep the top of that vapor column cold so that the again the vapor doesn't become uncontrolled remember we've got a ton of heat coming up here thousands of watts potentially of hot vapor coming up here and if we just had a piece of glass at the top it would quickly become as hot as the vapor of course and then we might have a molecule that sort of wafts its way up so this cold cap is another invention to make sure that the top of the column stays cold and all the jets stay focused downward next we can see this pump actually has a total of four pumping stages there's one right below the cold cap here and it has the largest area so basically everything between the wall of the pump and this cap is sort of inlet to the pump and then the next stage is here and it's a little bit wider and then the next stage is here and it's also a little bit wider and the reason for that is each one of these stages has a compression ratio just kind of like in a jet engine where the the compressor blades get larger and larger in diameter as the compression gets higher the density gets higher and higher and eventually down at the bottom of the pump we have a relatively high pressure and the fourth stage is actually right here there's another jet that's going to be shooting out tons of hot vapor this way so once the pump has moved most of the gas molecules down to here the last stage blows them this way and that pushes them out this way and out the pump and then it has these two things which are also oil traps i guess the idea is if there's a little bit of oil vapor that makes it this way and hasn't cooled off by the time it gets out to here again this is optically opaque a gas molecule traveling this way has to contact the glass at these low pressures i think i'm not exactly sure how these work but it does manage to catch a little bit more oil before it makes it out of the pump i should also point out that this whole section of the pump is gla is water jacketed so there's actually two layers of glass here with water constantly flowing through here again the amount of heat that's going to be going into this thing is massive and so the amount of cooling also has to be massive to keep this thing from melting down we're going to power the whole pump with this table top burner these things are great it's a real gas guzzler and so it puts out a lot of heat but be sure and buy extra fuel at the store it's just a butane fuel and the mechanical pump that we're going to use is that large orange pump that's on the ground so i'm going to turn that on first and let it run for about 5 or 10 minutes to get most of the air out and then we'll fire up the burner and get the rest of the air out so one thing that's interesting when we first pump down it looks like it's boiling but actually what that is is dissolved gases coming out of the silicone oil so i haven't even turned the burner on yet just lowering the pressure in there means there's a lot of gas coming out of that fluid i think it's actually a unique property of silicone oils that they're just able to dissolve high amounts of gas this is why you have to degas silicone casting resins so we'll let this thing run for another few minutes the bubbling will have stopped and then we'll turn the heat on while this thing is warming up i have another funny story about this pump i actually bought it a year ago for my mass spectrometry video and had a super difficult time getting it working i had it hooked up to the chamber with a vacuum gauge and no matter what i did no matter how much heat i put into this pump i couldn't get the vacuum gauge to read low enough it wasn't pumping turns out totally operator error i had a plastic tube connecting the vacuum pump this diffusion pump to my mass spectrometry chamber and i thought for some reason that the plastic wasn't going to be a problem but it really is in these high vacuum systems it was outgassing fast enough and the restriction was high enough where even though the pump was working it couldn't actually lower the the pressure in the chamber to decent levels a combination of restriction and outcast but anyway what i was going to do is take this thing outside and put it over my foundry burner it was going to be a do or die moment either this thing was going to start pumping or it was going to melt or implode or something but we didn't get that far the point is that i tried this with all kinds of different electric burners i had hot plates i had cook tops i had induction heaters everything and i couldn't make it pump and i assumed it was a problem with the heat but anyway i'll show you the instructions for this pump too there you need to translate them and luckily i used google lens with pretty good effect to get most of this out and then a kind person on twitter helped me with the rest they claim that this pump is 800 watts which i i don't believe i think it requires substantially more than 800 watts to pump properly maybe even like two or three times that much in fact what i'm going to do is crank this thing up all the way now that we're kind of warmed up and i've noticed that to get this thing to pump properly it pretty much needs to be full tilt this this particular gas burner needs to look like about like that to get decent pumping out of it so it's almost ready as you can see the vapor line is rising higher and higher it's already dripping out of here and then we'll do a time lapse so you can see the rest of the pump warming up so you can see the last jet is almost working there's still quite a bit of condensation happening in the column and in another minute or so that inner piece will be so hot that there's no more condensation happening inside the column and that top jet will start working and at that point the pump is fully operational and pumping as fast as it can you can see the lower jets are already working and there's a huge amount of condensing oil here so you can't see the jets of course because there aren't droplets it's actually just vapor but as soon as the vapor hits this cold wall which is kept cold by the flowing water it turns back into liquid oil and drips back down to the bottom which is what this this fluid is here and then this last jet is pretty active too there's a huge amount of vapor coming this way condensing out here and then flowing back into the pump this way the water chiller is a refrigerated unit that i got off ebay and used for lots of different things it basically consists of a refrigerator a cooling coil and a pump to move the water around okay so now that we're pumping it at proper vacuum levels let me tell you about this cathode ray tube that i built i used the erlenmeyer flask and i coated the bottom on the inside with some phosphor and i got this phosphor from cleaning that electroluminescent paint that i've showed in previous videos and i cleaned it by washing it in solvent repeatedly so that i got just the powder out the rest of the paint has a bunch of polymer binders and things but i didn't want that i wanted just the powder the phosphor itself so i let that dry on the inside and made this nice nice phosphor coating and then for the electron gun inside there i used a tiny light bulb where i crushed the glass envelope and i'm just using the tungsten filament itself so we're going to try this first with just plain tungsten and with the focus in the state it's in it's kind of hard to see what's going on oh there we go i'm driving those deflection coils with the signal generator and my audio amplifier to power those coils and i wound the coils to be about you know six or eight ohms or something pretty close to a speaker as you can see the deflection is not very big it's only about you know a couple of centimeters or something like that and the reason is that those deflection coils have to be pretty far away because of the shape of the flask normally a crt neck would be fairly narrow at the deflection area and it makes it easier for the magnetic fields to get a good deflection one thing i've always wanted to try is to make a barium emitter so i mentioned that we're using pure tungsten here and i basically just took a light bulb and crushed the glass envelope and we're using the tungsten filament directly to boil electrons off but as it turns out tungsten is a pretty lousy emitter of electrons barium is much better and so what we're going to do is put some of this which looks like bacon on there but it's actually barium carbonate and heat it up in vacuum and what will happen is some of that barium carbonate becomes barium oxide and then that's a much better emitter of electrons like a thousand times better than tungsten so we should get a much brighter spot for a given amount of thermal energy that we're putting into that emitter the downside is that once you've done this barium treatment you can't ever expose the filament to atmospheric conditions again i think because the water in the air reacts with the barium and makes barium hydroxide which is not a good emitter so this is not a problem for commercial cathode ray tubes where they're sealed up at the factory it's a perfectly sealed glass envelope so you can have something that's sensitive in there but something like an electron microscope it's not going to work because you have to keep opening the microscope to atmospheric so those use pure tungsten emitters and apparently give up this thousand-fold increase that you could have with a barium emitter so let's see if this works the filament burned out so we're going to try it again well as you can see the experiments with the barium carbonate are kind of hit or miss so we're going to have to experiment with that in a future video anyway i hope you found that interesting see you next time bye

Channel: Applied Science

Views: 179,534

Keywords:

Id: SrNVLCHrJtY

Length: 16min 36sec (996 seconds)

Published: Sun Jul 19 2020