Make vacuum pre-loaded spherical air bearings without special tools

Today, on applied science, we are going to talk about air

bearings

. These are like an engineering cheat code because you get so much for so little the parts cost only a few dollars and in today's video I'm going to show you how. build them without any

special

tools

so besides the super low friction you can also do fun tricks like this it's actually a captive bearing and the way it works is it uses

vacuum

to hold the sphere there so you can take it out and I can see that there is a hole where it was

vacuum

ed.

Put it on and it's captive and very low friction, so let me turn off the vacuum pump so you can hear me a little better and then we'll take a look at what we have. I've been playing around with different air bearing designs off and on for a few months now, but actually, coincidentally, someone called David Priest recently produced a video on air

bearings

and it appeared on Hack a Day, so that sparked my interest. came back and asked me to do this video and I'll link to his in the description so the key here is we're going to talk about graphite bug bearings the magic stuff here is big chunks of graphite and the reason why we use.

This is because it is porous, although it looks like a solid piece, it is actually only about 85 percent i.e. 15 percent open vacuum and another reason to use this is that it is very soft, so if you suffer shocks instead of having steel. in contact with steel you have steel on graphite and it's quite a slippery material anyway it comes off even with your own fingerprints even take it off so the next question is what kind of graphite can you use all the graphite that I'm going to use. The one shown in this video came from Amazon and I tried buying it from different sellers, different types, different shapes and sizes, some of these were just strange.

The cuts probably the best deal is to get a 10 millimeter thick plate like this. Convenient because it has smooth surfaces to begin with, so it's easy to

make

, easy to cut, and I've noticed about the same performance for all these different bits to

make

the

spherical

bearing. I used one more piece like this to make the cylindrical bearing. a chunk like this and I haven't noticed any performance difference between all of these things, so I'm inclined to think that all the standard grade graphite type works well enough. These bearings probably won't be the highest performing bearings ever, but you don't need to get any

special

graphite, that's what I say at least from my experience, so you know, people say that machining graphite is really complicated and they're right, in It's actually pretty terrible, the dust gets everywhere on the floor and it's slippery because it's a dry lubricant and then it gets on your shoes and then you track it somewhere else and I thought, well, you know it's not that bad, but then, one or two hours after one of these machining operations, I lost my nose and I'm not going to show you, but it gets everywhere, let's say yes, dust collection is a pretty big problem.

A really basic way to get started is to just make a plain air bearing and I'll show you. what the structure looks like, uh, you don't even need to use custom fabricated parts, this is actually a PVC pipe with a cap like that and all you have to do is cut a piece of very rough graphite from your 10. A slab millimeter right on the band saw is fine and then stick it on the lid and of course you will have a rough surface on this side so all you have to do is put some sandpaper on a flat surface even the glass works and sand it. flat and the graphite is so incredibly smooth that it won't take any time at all, it's literally just a minute of sanding is enough and you don't need any Crazy Fine sandpaper.

I think I only used 150 grit and you could definitely take your time and improve this a little bit, but the point is that even with 150 grit and a minute of sanding, you'll be good to go and have fully functional air glass. Thin pieces of glass will actually bend like one-eighth inch window glass. Pretty flexible, so if you're going to use that, you need to place it on a very flat surface first. I found that even my nice table here isn't flat enough to make a good air cushion, so if you really have thick glass lying around that actually works a lot better.

One of the biggest challenges I had with this design is that air leaks through the piece of graphite. The trick is that if you glue the plug this way, the glue must have a pretty good shear force with the disc to prevent the entire piece of graphite from coming off the PVC fitting and in David Price's video he had a great idea of cutting a conical shape on this and then gluing it so the glue would form like a wedge so you wouldn't have to do a full shear load to hold it you would actually have a little bit of grip there and I tried it and had some luck, but it leaked past the The chunk of graphite was still a pretty big problem.

Actually, I shouldn't say it's a big deal because if air leaks happen, you know what's still an air bearing, so it's okay if you actually have hole leaks like I say, all this air bearing. The concept is very forgiving in a lot of different ways, so scratches, even nicks and leaks running through your Puck don't actually decrease performance that much, they might make it a little less air efficient, but that's also okay so then I started making designs that were a little more captive to try to fix this air leak problem and the trick here is you want to have two parts to your chamber so instead of just one part like this cap of PVC where you have to glue the piece, what will really be nice is to have two pieces like a top and a bottom, so cut the graphite so that there is a ledge there and then glue the ledge on the top half so that the surface The seal is basically trying to be forced outward by air pressure, what you really want is for the air pressure to force your seal to be better or no worse.

After I got comfortable with just making flat air bearings, I found this vacuum preloading technique to be really cool. Basically we have two connections to the bearing, now the smaller one simply adds pressure to the back of the graphite, forcing air through the porous part of the graphite and the other port is connected to the vacuum and the vacuum is simply connected directly to this open chamber here couldn't be simpler it's literally just an open chamber and what's happening is the vacuum is actually trying to pull the flat piece in this case down and the air pressure is trying to force it up and those two forces.

They act together and hold the bearing captive. Pretty powerful concept, they call it preload because if you add too much load, you could actually ground the bearing, as they say, and you can calculate how much area you have. up here and how much pressure you're putting, then you know how much weight the bearing can handle and usually for air bearings it's not that high, it's about 50 or 60 PSI or three to four atmospheres or something like that, and If you calculate the area, you can calculate how much weight you are going to support, you can also change the amount of vacuum area you have in relation to the amount of pressure, a kind of support area, sorry for the compressor noise, like I was saying.

I haven't experimented much with different vacuum pressure ratios, but from the few I've built I've noticed that it's not very critical again as you can get away with a lot of different things, so let's talk about making these. non-flat ones, flat ones are easy because you can get a piece of thick glass and use it as your thing to make it into a perfectly flat surface and you end up with a nice flat air bearing, but if you want to make a Cylinder or sphere, the fit between the cylinder that goes in here and the sphere that goes in here has to be really good, incredibly good, so usually people use precision

tools

, you know, a boring head on a mill or something like that, but it's hard to set up. and it's expensive, so what I'm going to show you is a shortcut of how to do this without any special tools and you'll probably laugh when you see how crude this method is, but it works surprisingly well, the idea is that we make a cut from what we are eventually going to put in the bearing, so to make this

spherical

air bearing you first need to take the material out to get it close and I used a large die.

Basically grinder to hog some material and then made an abrasive cutter by putting super glue on this two inch spherical ball bearing and coating it with 80 grit garnet, basically making sandpaper on the surface of this bearing and then just grinding it to hand. Stick it in there to make a chamber that's about the same diameter as the sphere and I have no patience with hand tools, so when I tell you that even I was able to do this without getting too frustrated, believe me, the process goes pretty quickly. The graphite is so soft that you feel like you're making good progress after every pass you make and that keeps you going, so after using the sandpaper method the fit isn't good enough.

I mean, it's close, but it's not even close. perfect for an air bearing so to make it really fine I took another one or cleaned the sand basically out of the tool bearing and then I got a disc grinder and cut some grooves in the ball bearing and turned it into a file so with these grooves cut here there is a very slight amount of cutting action, but the dial is still absolutely perfect. I mean, there's a lot of area here that the thing can move around and then again, just with your hand, just polish it there like you're. squeeze an orange and it works.

I can't believe it, but again it's fast enough that you won't get frustrated even if you hate hand tools. It's not like I spent all day squeezing this orange so that it fits nicely in the end. What follows is almost zero clearance because the exact same tool makes the exact cut that you know is basically the same part or I mean with these ball bearings I actually used a separate one obviously this is the tool and this is the one that is actually being used but they come out of a factory that has incredibly tight grinding tolerances and similarly for these linear rails I will put links to all of this in the description, these came from McMaster, the ball bearings came from Amazon , the grinding tolerances on these are really tight, so you can be sure that if you buy one of these, even from a different batch, it will have a very consistent diameter.

In retrospect, the file's cutting action was so good that I'm not sure we really needed it. The sandpaper approach maybe depends on how frustrated you are, how good you are with the grinder, or maybe you have another technique to make this first cut a little more precise and then you don't need the 80 grit sanding. so much, but I used a very similar process to make this open sided sphere or open sided cylindrical air bearing, but in this case we didn't have to use the super glue with the sand sprinkled in there, I basically just put some sandpaper with double stick tape on the sphere or on the rod and then cut it again on a file and push the block into this filed area so that it fits well, let me remove this.

Oh, I should also point out that without the prior vacuum. -load friction here is super low. I'm sure you've noticed as I was talking about this, it just spins and spins, it's really low friction and as some of the air escapes out the side of the bearing, it actually holds it. I think that once it is disturbed in a certain direction, the escaping air actually keeps it spinning. If I turn on vacuum preload, the thing probably changes the geometry slightly and although the friction is very, very low, it's not as low without vacuum preload. so let me bring this one down and I'll set up the big sphere or the big cylindrical.

I have some magnets here holding it captive and we're running at about 60 PSI of intake air and I was a little worried that this one had a lot of exposed graphite that doesn't really go into the bearing and I was worried that air was going to leak out, but it turns out that the porous structure really once it is

loaded

. with pressurized air, even if you have an exposed face that is not part of the bearing, that is not a problem and one thing I have noticed is that with a setup like this, if this is offset with the shaft it will tend to lean a little and a once you've ruined the air mattress, like once you've altered the consistency of that air mattress, it's no longer going to be a bearing and everything will literally stop, but I have it set up here, so it can't really get too far away and this thing is wide enough to actually hold it, so it's great to have a bearing you can take apartquickly like this and then put it like this and it's, you know, pretty.

Very close to zero friction, another point is that stopping and starting is very good, so even if you don't care too much about overall friction, the fact that it takes a small amount of effort to start is one of the air bearings. type of key features there is no friction and then there cannot be any because there is no contact another interesting feature of error bearings is that the wear is zero because the two parts do not touch each other there is a small cushion of air between them the bearing can It never really wears out, I mean, in theory, because there's actually no contact with dust and dirt and anything that gets in there could scratch it or something, but there's no contact metal to metal or between moving part and stationary part, so what cannot be worn out.

I ended up 3D printing them. an fdm printer and using simplified 3D to make the cut and the idea is that it is relatively cheap to 3D print objects of this size with fdm and the tightness of the 3D printing is fine, it is acceptable and the strength is also almost good. I've had them up to about 70 or almost 80 psi and one exploded at about 80. This has three outer layers and about 40 percent fill, so I'd say they're about okay, you'll probably want to beef it up a bit. Additionally, another technique I used was to coat the interior surface with epoxy to improve the air holding capacity and also a little more strength.

I'll show you this flat captive air bearing in action, it's actually a lot of fun. to play with so now it's locked onto the piece of glass and if I can keep it perfectly level it will stay there and um it's actually really hard to show on video maybe what this feels like but it's very strange if you're looking for a plate of granite surface, you can get it on Amazon anyway and of course the performance is really great because it's super flat and it won't be influenced by the surface you have it on. It will always be very flat and the performance is noticeably better than even thick glass.

You can tell it's not even close to level, in fact that makes it fun to play with because it's like a bouncing ball, but it's an Electric Transformer A couple more final notes. The air accessories I'm using are somewhat standardized. They are 1032 straight threads used with a rubber gasket to seal so it is not a tapered pipe thread and I will put links to all of this for McMaster. for the pressure tube I'm using a 1 16 inch inside diameter hose just because it's so small and flexible, it makes the demonstrations a little nicer. The glue is just two part epoxy.

I experimented with a few other types of glue, but there are really two. The epoxy part is by far the best, so I just used that kind of biggest innovation, if you can call it that, in my design was to use this ledge so that the thing is captive in this way, the air pressure you're just putting more force into that glue joint. and keeping it closed, try to avoid any kind of steep areas where the glue is responsible for holding the thing together in cut. One thing I haven't tried yet but think will work pretty well is to do a closed bearing like, basically, you know, push this through the block and then blow the air out everywhere, as opposed to my single cylindrical bearing. side.

It would be nice to have a fully enclosed cylindrical bearing and the way I would do it is to drill a hole that is under 10 or 20 mil in size and then make a file or broach using the same method I showed for the large spherical cylinder, basically you just cut slots in this thing and push it in by hand very carefully and the graphite is very smooth. it will just get out of the way it's almost like a centerless grinder as if it didn't work this technique wouldn't work for putting your custom made file into a drill or any other electrical machinery because then it would be up to the Running out of that machinery is very low, the benefit of This method is that it is essentially centerless because you are using your hands to find the center and that is what allows the method to be so accurate, so it is something you all can try.

Air bearing enthusiasts, okay, I hope you found it interesting and see you next time, bye.