Your channel is criminally underrated. Most of the time I really have the faintest idea of what you're doing, but you explain everything so well that I come away knowing something new. I cannot wait to see what you have in store for your new vac chamber! Maybe some more Caesium????
Thank you! Currently the experiments planned with the chamber do not involve cesium. However I thought about testing if cesium can be cooled to a point where it won't react with oxygen anymore. Would be interesting to see if it still reacts at 20 K. But the setup for that experiment would be pretty complicated so I will need to think about how to do it.
Something very important is a heater for the chamber to achieve a good base pressure. You want to get rid of the moisture that sticks to the chamber walls. I used some resistance wire inside a fiberglass hose wrapped around the chamber. Works really well and its quite cheap. Heating the chamber for about and hour should be enough.
@@AdvancedTinkering that will also heat all the electronics attached to the chamber. You want 60°C on the chamber wall. If you wrap for example 15m of that resistance wire around your chamber it will heat it evenly. You can also regulate the temperature if you attach a pid controller.
I am drooling over that vacuum chamber 🥵 Thanks for the shoutout! I'm sorry the charcoal trap didn't work for you, I'm surprised at how much oil mist was passing right through! I think it's possible that my long exhaust line and the knock-off "Alcatel-style" oil mist eliminator that I have installed on my pump are doing a better job of removing the larger droplets before they reach my filter. It may also be that my pump is much larger and slower, and so it produces a less dense mist.
Yes, it's probably the high flow rate so the oil doesn't have a chance to be absorbed. And since your chamber volume is a lot smaller it might work in your case. But even in my case it's better than nothing :) I once had a vacuubrand oil mist filter which worked great. It was some sort of sintered polymer. Hard to describe.
If you ever get the chance, switch to a dry diaphragm roughing pump. My lab uses those in our Pfeiffer HiCube 80s and 300s where oil contamination must be eliminated. Those particular pumps are probably out of your budget, but the roughing pumps are sold individually. The HiCube 300 can get our 5 ft long, 14 in inner diameter chamber down to ~1E-5 torr after 4 hours. When doing your leak test around those plastic o-ring seals, be careful how much helium you spray around them. Helium can absorb into and penetrate the plastic and give you false leaks and continue to offgas helium for quite a while. The helium can also back-diffuse into your helium leak detector, so keep your detector upwind of the area you're spraying. My boss once spent 5 hours trying to find a false leak because of this. I set my helium spray gun to release just the barest amount of helium so that I can only feel the helium on my tongue. Or use a glass of water if that's weird.
All of my colleagues that switched to dry pumps eventually switched back to rotary vane pumps. The oil in rotary vane pumps will catch any dust or powder. Changing the oil will effectively clean the pump. Dry pumps get dirty and are hard to clean. Also, rotary vane pumps generally have higher extraction rates.
You want the cold trap, we want the cold trap, everyone wants the cold trap. You could probably even use it for both oil filters, one large dewar thing. Both oil filters in there, chuck in a few liters of liquid nitrogen or isoprop and dry ice. From what I can see the "trockeneis-shop" sells 10kg of pellets for around 50€. Wouldn't that theoretically also help with the vacuum if stuff starts condensing around there?
Haha, I would really love to have a compressor cold trap. Sure I could use a liquid nitrogen or dry ice cold trap but it would be very expensive over time and I would have to wait for the dry ice to arrive every time I want to use the chamber.
Why not a hot trap? The volume of air coming out the exhaust is so small, just pipe it into the air vent of a Bunsen burner and burn it off with a small flame. The current molecular sieve filter will act as a flame arrestor.
Signal Ditch brought me here and I'm subscribed now :) glad to see more vacuum nerds, really kind of you giving the manufacturer extra credits - would be really interesting how the process of a getting a custom made chamber looked liked.
Now that is a nice looking chamber! So much room for activities 🙂 Really like those vac gauges too, super convenient. I've never been happy with mist eliminators, some oil always seems to leak through. 😢 I snagged a scroll pump off ebay a while back and it's been such a nice upgrade (a lot quieter too!)
Thanks! I can't wait for my first projects with this chamber. Probably PVD (thermal evaporation). Yes, a scroll pump or a diaphragm pump (only works with a new turbomolecular pump which doesn't mind a higher foreline pressure) seems to be the best option. Big fan of your channel by the way!
This is a very sexy vacuum setup and one of the most impressive home science setups. Clearly you've thought things through a lot. Just an idea but ultimately it's your setup - would it make sense to print out some 'hold down' clamps so that the chamber cannot accidentally lift out or get raised out when torquing down bolts? Another idea would be some basic brackets for the side, to hold the lid out of the way and protect the CF flange knife edge, rather than having to set it down on flat surfaces and eat up bench space?
Thank you! That's funny, I had the exact same idea and have already designed clamps that would have pressed the chamber to the aluminum extrusion from above. But it turns out that the chamber is so heavy that it's very hard to turn it. When I tightened the screws, the whole stand moved because I attached feet that have casters. In addition, I can lower feet on which the stand then stands securely. I may eventually print the brackets anyway. Just to be safe. They are not CF flanges but ISO K flanges but you are definitely right. My plan was to design a mechanism to make it easier to lift and move the flange aside. Possibly using gas springs to make the process smoother. There is definitely still a lot to do :)
Wow that is beautiful! I first expected the bolts to be for electrical experiments but they don't look insulated from the chamber itself 🤔 Super excited to see what you have in store - the breadth of your engineering talents are very impressive!
I'm using pure mineral oil (Weissöl) for my vacuum pump to deal with the oil mist. If a bit of oil gets into the air it will not cause any issues. Mineral oil seems to be a valid vacuum oil option, it's used in the food industry.
I own a pfeiffer oil mist separator that works flawlessly. Have tried to measure any residual mist coming out of the exhaust port but it really does seem clean. Theyre extremely expensive new but sometimes appear online on the second hand market. Perhaps something to keep an eye out for.
10:51 Accidental Hyperspectral Imaging (On a serious note, a “Filter Wheel” and/or that Raspberry Pi Camera based Open Source Spectrometer could be interesting additions! )
Don't know if you've solved the oil mist problem yet, but oil adsorbs best when cold. So perhaps somehow chill your oil mist eliminator. You could also try percolating the output air through oil.
Looking great! My condolences for the whole oil issue. Switching over to oil-free scrolls or membrane pumps is just such a quality of life improvement. If you have the valves and hoses for it, being able to bypass the TMP to pump down the chamber with the roughing pump is a very useful feature to have. Did you already check your setup for leaks with your new leak detector? If you can see plasma from the ionization gauge, putting a spectrometer there might allow you to do a little bit of composition analysis. It will be exciting to see what experiments you will do with this.
Thanks! Yes, a bypass for the turbomolecular pump is definitely planned. But the valves for the large flanges are very expensive. The chamber was tested with a leak tester by Wissel Vakuum after construction. I myself have not tested it yet but will probably do so. Recording the spectrum is a very interesting idea. I have to test which wavelength range I can measure through the borosilicate sight glass. If it absorbs too much I will have to see if I can build a vacuum feed through for a light guide.
@@AdvancedTinkering Good luck with the testing, and don't let permeation through the O-rings lead you astray. Regarding the windows, borosilicate should be fine for most applications with good transmission in VIS and near IR. For UV light I'd recommend UV grade fused silica, for really far IR zinc selenide, and for widest transmission all around calcium fluoride.
@@AdvancedTinkering Roughing through the turbomolecular pump is very convenient and a good way to keep oil from the roughing pump out of the chamber. I only ever did that, and didn't ever have trouble with dust ingress but you may know for sure that you require the ability to bypass the TMP for roughing. A small TMP like that should take 10 min at most to spin down gently and 15 min to start back up. Isolating the roughing pump is a more useful capability and well worthwhile, especially if you have a ballast behind TMP. Putting the leak detector behind the TMP is very useful, especially since it allows you to mount and demount it without breaking vacuum. Feel free to ignore me as long as you are sure you understand your requirements :)
7:25 Any idea what temperature they need to reach? It would add more Complexity, and probably Energy Use and Cost as Well, but depending on the temperature needed, a DIY Version may not be too far out of reach. Granted try finer filters/powders, and maybe even something akin to what is used for compressed air first of course!
Why not use a oil bath with a bubbler tube submersed in it. Like you do a vapor trap in chemistry. Maybe you need to add a one way valve to make sure it can never suck back the oil form the trap. Don't know if the architecture of the pump would allow such a case.
That might be possible. Just a bubbler would probably still let a lot of mist through. But using some sort of diffuser (like a sintered metal sponge) might increase the surface area enough. But then the flow resistance might be too high. I think I will leave it like it is for now. Once most of the air has been pumped out the oil mist stops. At some point I might get a proper professional oil mist filter.
Absolutely AMAZING Build! Those 3D Printed Parts Were Really Novel, and that Camera Feature will be neat, especially for long timelapses waiting for a layer to grow. Do you have the Files for these published anywhere like Printables, Thingiverse, or even a GitHub etc? Granted I don’t have the budget for my own yet, but I’d imagine other channels/enthusiasts, or even to a degree “proper” laboratories could benefit.
Thank you! I haven't uploaded the files yet. Don't even have a Thingiverse/Printables account yet. I wanted to make a parametric drawing before that, so you can easily adapt it to different flange sizes. But probably better if I upload it now. I can see it coming that I won't have time for the parametric drawing in the end and won't upload anything at all.
Hello, you are making amazing content, I have been fascinated by vacuum technology for a long time. Your channel and everything on it really interested me. Could you please share with me some information about your vacuum chamber? I'm interested in as much as possible, pipe size, wall thickness... Regards. @Advanced Tinkering
not used my vacuum pump in a while so cant be exactly sure but i have that same filter as yours but there is a small drain pipe that leads back to a valve on the pump housing to recycle the oil in the trap.
Nice project man! I'm glad that you mentioned the oil mist. Mine also does that, and I wasn't sure if it was normal as I'm new to all of this. I had it sorted by routing the outlet through an oil separator for compressed air. Does yours also heat up to a ridiculous temperature after an hour? I can't hold my hand on the box with the oil. Thanks for the Signal ditch channel recommendation.
The first vacuum chamber I did a lot of work with had a 16.5 in CF flange as the main closure. It weighed about 30 kg. 36 bolts to be torqued to nominally 360 in-lb, though in practice 450 was fine and let you increment 350-400-450 to get multiple closures out of a copper gasket (that practice works 90% of the time, and the other 10% you just torque it to 450 anyway). And of course they had to be tightened in that pattern, 4 twists per bolt with the wrench. After a year of working with it I noticed my arms were way stronger than they had any right to be for someone who never exercised them, but then I understood why. When possible I grabbed some colleague or even innocent bystander to help me lower the flange into place, but eventually I realized that with good posture I could control it perfectly well by myself. It had lifting eyes, I wasn't trying to hold it with fingertips on edge.
ich hab ein sehr günstiges setup aber mein Ölfilter funktioniert ziemlich gut. Ich habe ein Stück Schlauch genommen (ca 1 meter lang) und am Anfang und Ende Zellulosetücher reingestopft. (küchenrolle oder papiertaschentücher) Das ganze ist lang genug, dass das Aerosol abkühlen kann und dicht genug, dass es durch die Tücher durchgequetscht werden muss ohne einen Umweg zu haben (z.b. um filterkugeln herum , was bei höheren luftströmen am anfang des pumpens vemtl. passiert)
1:48 I knew you were living in an submarine. I'd expect at least some kind of pressure difference near the living quarters of the owner of such nice vacuum equipment.
That Swagelok needle value might not be suitable to actually close completely. Please check the series of that value. A "metering value" (at least series MH, ML, MS and MV) is only used for controlling flows, not turning them on/off (hence metering). You will find that you shear the valve stem off when you actually want to close it. Ask me how I know.
My first attempt at catching oil mist was an old sock which naturally did nothing. Eventually I gave up and just plumbed the pump to a garden hose which I hang out the window. Your chamber is hot. Does the manufacturer (Bissel?) have a catalog I could stuff under my mattress with the rest of my collection?
Good video, sexy chamber and clever use of a vacuum filter. 1. what department are you in (Elias is clearly a chemist)? 2. where did get the Triangular mending plate for the t-slots?
Haha, true! But the large pump from my newer videos just isn't practical for me (at least not yet). So I will eventually start using the smaller ports again.
Danke! Ich versuche nicht zu zählen was ich so für meine Projekte insgesamt ausgebe :D. Falls du nur die Kammer meinst: Die hätte ich mir niemals leisten können, weshalb ich Wissel Vakuum unglaublich dankbar für ihr Geschenk bin.
With so many stuff in the vacuum side of the roughing pump, you severely hinder your vacuum performance there. With dual stage pumps you could reach well into molecular flow regime. But with the obstructionss this will take ages.
Do you mean the foreline trap, hose and valve? I tried keeping it as simple as possible. Eventually I will probably move the valve of the turbo to be able to isolate the roughing pump. But as far as I can tell everything else is necessary. How would you do it?
@@AdvancedTinkering I don't know. When you say it is necessary, I have to believe you. I only worked with either high end automated apparatus, or very simple setups with just the pump attached via a large diameter hose. And the latter processes (welding) were not sensitive to that tiny scale of oil contamination. BUT it was possible to get into the 10-4 mbar range with a good dual stage pump alone within a minute. But maybe your tasks are more demanding. The diameter of the pump connection becomes very important below 10-3 mbar because of free molecular flow. And putting a fine mesh in is about the opposite of large diameter piping. (Molecular flow blew my mind when I first heard about that, years ago!) I guess you know because you seem like a like minded premium nerd ;) , but in short: When the pressure drops low enough so that the gas particles rarely interact any more with each other, but mostly the walls (free path length > vessel diameter), the flow loses its property of directionality. Because particle interactions with the walls mostly don't reflect particles. They get adsorbed to the surface and shortly after get desorbed again... with random direction! So further pump down becomes a matter of statistics of a particle finding its way to the exit and getting kicked out. This is why the piping diameters get very large with high vacuum systems, and further pump down takes ages. Might be an interesting topic for a future video.
