MAKING PRECISION GAGE BALLS

i'm making a carbide custom size test ball and i'm going to use the d bit grinder to rough in the spherical shape with a very small neck on it the nature of this particular ball gauge is going to get ground with a cylindrical band around the whole ball and that will remove the little post that's about 0 90 inch of diameter that holds the ball for lapping purposes i'm going to do one ball on each end this is the setup for aligning the lateral position of the of the head in this direction so that this axis of rotation of the part coincides with the axis of rotation of the radius or angle adjustment so i do that by getting the indicator probe lined up pretty close to the center of rotation of this so that i can come in and gently come in here and sweep and find my low point we'll get in a little closer here come in and sweep and find the low point get off full go full 180 the other direction and come to my stop again my in-feed stop again and make sure that i'm in the ballpark here remember this is going to get lapped round in the end so within a half thousands or so is good and this is adjusted by the lateral slide here which has the adjuster screw here and the lock is under here on the back so this graduated dial here with the zero obviously is not accurate enough to just line it up by scale then we're going to come in this position to do our axial positioning to make sure that we have enough material and that is adjusted by the axial positioning which is right here this knob and not not that that's the lateral this is the adjust here and the lock for that give lock for that is over on this side so now we have this position such that there will be material to do a full ball uh on here here we're plunging in with the 1a1 diamond wheel and just removing a significant amount of material to establish the neck but also using the side of the wheel as you can see there to wipe the diameter here we're coming out on the outside of the wheel we have to be careful because they're not pointing at each other directly and rounding off the major material on the tip there now i'm using the side of the wheel again because that's nice and perpendicular uh on the contact patch and running around the back side here peeling off the excess now we're just taking successive passes to get closer and closer here checking to make sure where we are got plenty of material here now we're reducing the neck a little further and getting our angle there and getting to where we're going to be able to get the lap to come into the neck checking again make sure we're okay on size spun a little chamfer on there make it a little bit nicer to handle here's our carbides dumbbell and now we're going to rough lap these in to get them a little more round and closer to size before we start lapping with diamond slurry or diamond compound we're going to use a diamond core drill which will do a nice job of rounding this up quickly a little piece of hardwood here with a just a rough countersink in it and what that allows me to do is put this on one side core drill on the other so that the pressure i'm putting on with the core drill i'm putting between my hands i'm not pushing on that skinny little neck because you could just push it off and snap it off and then you're then you're dead in the water [Music] okay we're down to 370. looks like it's going to clean up everywhere i've got one little divot here that i nicked it with a grinding wheel but i think that's going to come out and that's why i've got two sides so do the other side the same procedure while i'm here [Music] [Applause] [Music] so i'm going to use a little aluminum oxide slurry in this purpose of the aluminum oxide is to abrade away the nickel matrix of the core drill which if you don't is going to just rub on nickel nickel on carbide and we're not going to get any cutting action because we're just going to ride on the nickel that's where you saw the smoke coming from was uh see that's much happier now it's got something that's able to remove the uh [Music] let's see what we got here oh yeah get you in close here you see the difference there on the finish just from the the aluminum oxide is exposing the diamond in the core drill that's in a nickel matrix and um without that the nickel just rubs on the carbide and does nothing and keeps the diamond from cutting we took three thousands off of that with that last pass so that's uh that's good everyday the same on this side get a little slurry in the core drill here and get uh of here [Music] [Applause] let's see where we are here that's good we're going all the way down to 358 something so we got a ways to go yet so we pound some more off here there's the inside surface of that core drill these core drills are basically a completely diamond nickel this whole section this is not plated just on the outside so that's why this cut on the inside so nicely now i need to make a iron lap to rough these in to get them close and then it will switch to aluminum for the final finish this is a piece of ductile iron that i rounded up probably with the boring bar and then in the backwards in the bridge port but this allows me to stick this in a collet and i'll turn the other side and we are putting uh since we're making the laps in the lathe it's easier to switch gears and just put this in the drill chuck spin this and let this be the the moving lap and this being the moving part and move the drill that way you can do your machining on here or whatever you need to do instead of having to flip back and forth put a shank on it that fits the drill and all that just makes it's a lot easier just to make the lap spin in the lathe [Music] and that's close enough we don't need to have it completely uh tuned up plus i'm better off leaving this big in case i need to have something else a different size i only have to put a hole in this deep enough for the half of the ball i'm going to use an eight millimeter ball mill because that's the small diameter of the lap uh the 3.358 is the diameter i'm going to i also will be able to touch off here and zero my digital on this now i have the tools zeroed out both on x and z so i can go in more than a little more than half of the half of the radius get room for some wear so i'm going to go about 0.156 would be half so i'm going to go about [Applause] yeah i'm just going to go 250. [Applause] and then i'm going to come out diametric wise to [Music] the difference between the point 358 and the 0.315 which is .043 so i moved off diametric that much and i'm going 0.156 deep because instead of the ball and there's our lap i'm going to chamfer the outside significantly to be able to get in close to the [Applause] the lip check with the part and make sure yes okay no problem i can get all the way without any any interference there that's what i was after you can see i can put this in the drill and just oscillate this like this to get the shape so i just took a dental burr and just manually put these grooves in there so that the diamond slurry can get in between and fill up those channels and smear and keep charging itself instead of getting dried out when it seals up tight so that'll definitely help that only took about literally 60 seconds to put those in there i buy eyeball got some slurry in there let me just take a test little test run here and get this thing tuned up and take a check here and see how this is uh how quickly this is cutting before it's wet it's not finished that nice yet okay there you can see the ball you can see the finish isn't quite improved on this but because the shape of the lap actually changes as you go i'm going to keep flipping back and forth between these because if i do this all the way down to size and then flip it i will actually be changing starting over on the lap because it will it'll be cutting uh in a different contact pattern so by flipping back and forth between these ends i'm going to keep the contact pattern relatively the same as i bring the size down on both ends at the same time starting to round up pretty nice these electroplated tools in general are done by placing diamond in contact with the metal surface electroplating nickel typically on it to grab the diamond particle enough that you attack a single layer on and then putting the cover nickel on enough to leave some significant portion of the diamond exposed problem with that is that the a piece like this where these sharp edges are the field that forms on these will tend to have higher current density on these sharp edges and then that makes the nickel get thicker on those edges and as soon as the nickel touches and the diamond is not exposed you have a basically non-functional tool so a trick is to use a lapping compound some just find some fine grit aluminum oxide and putting it on there and what that does is it allows the silicon carbide to chew at the nickel that is that's exposed uh and expose the diamonds so that they cut and that really helps in a lot of these situations and you can also use a stone that you don't care about and abrade the stone to wear away the nickel matrix to expose the diamonds to a reasonable level look at it with a microscope and see what you're doing but very very handy so here i'm just using the corner of this file to come in here and reduce this reduce this section and i'll do this a little bit and i'll go checking the comparative so i don't know we'll do it but yeah that's that's how i'm getting that sharp corner in there and that's not going to uh your lap can't get in there and remove that sharp corner so if i don't do that it's just going to stay around and uh round it over and keep misshapen [Music] here's the ball in the comparator and i've got it lined up pretty good and what i'm looking for here is you'll notice if we follow the line that we're on where we are here you'll notice right here at the uh trans there it is right here at the transition this is bulging out a little bit here's the other side where i've taken the diamond file and i've started to knock this down and sharpen that corner up because there's only a very small area here where it's going to get ground off cylindrically when i grind a cylindrical band on this here i'm lapping with six microns making sure i get all the way over to the corner the wood block is what i'm putting the flapping pressure on with [Music] it's looking pretty good but i think there's still some scratches on the end [Applause] [Applause] very important to leave the drill rotating with the part when you release if you stop that pause is going to create a ring where that laps still rubbing so very important to come off rotating for the final polish just cut a piece of hardwood square felt sanded the corners found a collet that fits in made sure my two corners hit here on the one and obviously straddle's on there i'll i'll chamfer the od put the hole in there just like i did for the lap there's the same pocket i put in with the eight millimeter ball mill on center down deep enough to leave some clearance and then off center to get to where the ball just sits in almost to the 50 point a quarter micron in the wood so i have the balls wrapped with two layers of on tape to not scratch them up if they fall off i'm going to try to actually hold on to this before it falls off in this particular case i'm in feeding wall apart stationary and then slowly rotating the part kind of creep feed cutting around with whatever depth we had radial depth of cut and that basically one of the reasons is it allows me to only have two hands so one's trying to hold the ball the other is in feeding first and then grabbing the spindle axis to spin it around and it's pretty amazing how small that can get this ultra micro green carbide is very is actually very strong so it can get extremely small diameter there you can see and that ball is still holding on this is going to get ground off anyhow because this is going to get ground cylindrically across this point it's part of the gauge it's not part of me needing it for a manufacturing method it's the gauge actually needs a certain diameter band ground onto the ball okay here's a real time sequence on making the holder to hold the balls with mounting wax so that i can grind the band on them in the grinder making a relief here about the width of it about the width of the grinding wheel diamond grinding wheel i'm gonna put one ball on each side so that uh i can just flip it and it'll be ready to go just facing off to get a clean edge here it's just a lap carbide tool i have a center drill that's been topped off to be the uh still have room for the bottom but not be extra deep so it won't be using a lot of wax in the bottom change my pool number here and uh go to zero on that line up my drill chuck and just go in here until i get a razor edge and i know i'm there same thing on the other end [Music] this is lead lloyd pre-machining steel that's why it's cutting so nicely with no coolant no nothing that's it that's how quick things go with a solid tool post and tool presets turns out that the ball is sitting on the bottom a little bit not being guided by the outside so i'm just going to go in here with the other end the normal center drill and relieve that so that it doesn't uh keep the ball from seating on the outer perimeter to make it run concentric now when i check the ball fit in here it's seating in the on the outer periphery so that it'll run concentric i'm using mounting wax here this melts at about 200 degrees and i'm just going to melt some in here smelt the part in it and we'll be good to go [Applause] you can see here i have that little tit where that was attached horizontal to the ball axis and that will get ground off as we grind this cylindrical the wax dissolves in alcohol what i'm doing here is i'm just getting rid of the wax on the surface of the ball where i want to indicate it in to have the ball running true i don't want to trust the sitting in the seat i just wanted that to be close but i'm rubbing off the any of the dissolving off any of the wax that's on the ball surface so i can indicate right on the ball to tweak in the tir before i grind the cylindrical diameter on here i'm grabbing that 3 8 diameter piece by about 3 16 of an inch axially in the collet so that i have the ability to tap it around to get it to run true so i'm indicating directly on the ball so that when we grind the diameter the cylindrical surface will be concentric with the ball now we're grinding with the resin bond grinding wheel i think it's maybe 150 grit uh got to be very careful here to keep use coolant so that you don't warm the part up and melt the wax and you also have to be very careful with the micrometer that you don't twist the part and break loose the the wax mounting that wax is very brittle has no ductility whatsoever so um just taking light passes here cooling on not really difficult to grind that diamond wheel's dressed very true and leaves a really nice finish so it's cutting this without any problem and there we are with final size [Applause] just sitting in some alcohol let the excess dissolve off of these here's my measuring setup we have a tool maker flat with foam insulators stuck inside so i can move it without touching with my hands and warming it up we have a little xy stage here that's clamped to the lip of the tool maker flat we have carbide double a gauge blocks that's plus or minus two millionths and the bottom one is just a stage it has a wear block rung to it so that the um we're not wearing a hole in it from where we're dragging the ball across it's technically we're not we're not harming it at all but it's good practice to use a wear block when you're going to use this as an as an anvil instead of an actual just measurement device and then we have their two blocks that are are rung together to our nominal 0.358 inch reading then we're just going to use the gauge to tell the difference between the two the xy stage allows us to move the ball uh to find the high point which is very difficult without something to hold it still in each direction where you can just find the high point in one direction and then move and find the other high point the other direction at 90 degrees so this is our setup the little plastic leaf here is just gently pressing down just enough to be able to pull to slide the ball around on the on the gauge block and show some other details here we've got the cable on the electronic indicator is taped here at the base because at these low magnifications here where i'm reading five millionths per division um even just touching that cable will influence the reading or even touching the um the uh indicator base will influence the reading so i'm going to actually do the measuring process right now and what i'm going to do is i i'm sitting on the gauge block right now and i'll come up here and zero out my electronic indicator and i don't have to touch anything this is just electronic turning the knob to get it zeroed all right so we're right on zero there we'll come back to here and then we will grabbing the foam pieces we will move until we were gently roughly visually over this peak of the ball and then we're going to use the xy handles here to drive it around until we get our max reading here so i'm going to focus in on this while i do that so you can see what's going on here so i'm going to try to find one direction see if we can get on scale here there we go now we're on scale and i come up find roughly the high point there and then come up and find the high point the other direction and come across here and you gotta check back and forth between the two and you also have to get your fingers off of the micrometer heads because your finger pressure influences it also okay so that's 50 millionths on that scale but now it's always good to back up to the gauge block stack and make sure our zero is still good and you can see our zero has shifted a little bit so we're going to re-zero that we'll come back to the ball again and i'm going to find the high point on the ball there we go and we'll come in here and finding the high point both ways real touchy remember that's only 5 millionths of an inch between those red lines there okay so about maybe 52 and a half let's come back and check our zero again and pretty close so that's the general procedure there on how i'm calibrating these balls i have spun this for roundness to see how consistent it is diameter wise and it's very close it's probably within five millionths of the same size everywhere the other thing that's good here is that the carbide being both the gauge block and the ball means that i don't have to do any thermal uh compensation for what my ambient temperature is if my ambient temperature isn't exactly 68 degrees uh standard gauging temperature then there could be errors if i had either steel gauge blocks in the carbide ball or vice versa several degrees off from there you actually would technically need to compensate for that because the measurements are have to be good at 68 degrees fahrenheit so you have to account for the differing coefficients of expansion what the sizes of the parts are at this elevated temperature typically and then what what or the size change is going to be when it comes back to 68 degrees another advantage of what i'm doing here is uh if you use parallel jaw measuring systems to that have flat surfaces contacting the ball and the gauge blocks there's all kinds of parallelism errors and things that can get involved that make the the measurement much more difficult than what i'm doing here here i'm eliminating all that because i have sphere on plane contact on the bottom of the ball sphere on sphere on top dirt is typically not an issue it pushes out of the way uh meaning my new particles and things will get pushed out of the way instead of creating an erroneous reading so a lot of things in my favor here uh obviously by design to make this work out but i just thought this would be interesting to see a um you know a a rudimentary way of measuring a ball and you know in the home shop yeah but this isn't exactly your typical home shop but uh just using your head about what matters doing the xy slide to be able to do this if you didn't have that controlled motion for x and y to be able to move that ball and find that exact peak never happened you can't do it by hand it's impossible so i knew i had this in mind for a long time and this first time i had an opportunity to actually need to implement it and works real nice okay we're going to uh nerd out here on what we have to calculate when we're talking about extreme accuracy on gauging things down in the micro inch level so we're just going to take the scenario that we have in my example where i have a ruby contact point touching on a carbide sphere which is the part we're measuring sitting on a carbide gauge block which is our reference plane and remember that we zeroed on the step height of the gauge block stack to establish where zero would be for the electronic indicator but we know that there are deformations between any two surfaces that contact depending on their shape depends on what the contact deformation is and for things like sphere on sphere or sphere on plane like this there are hertzian contact calculations that allow us to calculate the amount of stress that's in the materials the contact diameter of what the actual squished contact zone is and then the normal approach which is the distance from points remote from the contact zone how close did let's say the center of the ball move towards each other uh with a given amount of force so um we in this example are going to go through that i'm using the calculations that are in spots engineering handbook and this calculation right here is for the normal approach it tells you how far the center of the ball moves towards the plane in this case sphere on plane uh with a given amount of load p and then this is the other one we all will use is sphere on sphere uh same thing contact diameter that's here the contact stress but all we care about is the normal approach so these are the formulas i'm using from spots this is the book spots design and machine elements very handy book i have these formulas in my 48 gx in directories for the different configurations sphere on sphere sphere on plane and in this one i'm doing the spheron sphere that we did for the gauge ball contact the probe gauge contact and you can see our radius here of the ball at 0.179 radius of the probe on the electronic indicator modulus elasticity of the carbide ball at 90 million modulus elasticity of the ruby ball at 55 million and the force at 0.011 pounds force roughly 5 grams and n is what i've used for the normal and it's out at two three four millionths of an inch so that's how we get there and this is the format that on the um 48 all of the variables come up and once you've entered the equation in in one fashion you can back solve for any of the others so you can see here we've got r1 r2 e1 e2 for the modulus elasticities force and the normal approach so it having in the calculator where you can just bring this up bang punch in numbers and go to town is very helpful takes a little while to put your equations in but it's it's worthwhile in the long run so one of the important things with any kind of technical calculation is properly assessing the situation so we have the contact point we have its force pressing on the ball now in this particular case the normal approach of the contact point to the ball is just the squish zone that we see here how much is this going to depress into the ball and make the reading shorter than it actually is so that's one calculation but the ball is slightly more complicated in that we have the weight of the ball giving us a force on this we have the weight of the the force of the manipulator that little plastic strip that i had pressing down on the ball so that i can drive the ball around in two in two x y coordinates to get the the absolute peak that's putting a force on there and then it's also got the gauge force added to it so it's the sum of the gauging force the manipulator force and the ball weight is the actual contact force here and then we have to make sure we're using the correct uh modulus elasticities of the materials and all that so the gauging force is five grams which is .011 pounds or 0.049 newton and the radius is 0.03 inch or 0.762 millimeter the i actually measured the manipulator pressure at the amount of deflection it had 20 grams of force is 0.044 pounds and then we also had the ball force which is its weight so i took the volume times its density gives us a pounds and in pounds force is 0.014894 pounds which is 0.066 newtons once we have all that we can go down and do the actual calculation so we plug in these values in the calculator we take the force 0.049 newtons the radius 0.03 inch and the radius of the part the ball itself 0.179 the force 0.011 pound or 0.049 newton that approach turns out to be a 1.234 micro inch or a 1.234 millionth of an inch which is 31 nanometers so that is just the that movement we have to now add the with the ball and plane deformation is so we've got the gauge force itself the manipulator force plus the ball force that total force is 0.069 pound or 0.31 newtons the only thing that matters here is radius because that's the the other is the plane approach there is 1.844 micro inch or millionth of an inch 47 nanometers well remember when i said to be sure you actually understand the problem and that you're perceiving things correctly well you've heard me say that we were going to add these normal approach of the of the contact and the normal approach of the ball and that's not correct so ignore the total that i had down below here of the total approach being something that's totally wrong so i'm going to show you where i went to so here's the progression i go through to keep my mind straight the material is the same on all the master and the part we're assuming that the parts actually exactly on size just to keep things simple and easier to understand uh gauging pressure is the same everything's the same so we have this example here sphere on plane sphere on plane same gauging pressure it's going to read the same doesn't matter if you have 50 pounds gauging force or you know five grams gauging force it's going to read the same the next example here is we're showing if we put a radius on the i have a ball contact down here ball on plane now this the only difference that's occurred is we've changed this so where this would have read zero now it's going to read directly read the deformation from this ball on plane from whatever the loading is it's going to read it correctly now we move up to here and we look at this and we have the same setup and we have no change in deformation here the only difference is now we have sphere on sphere so the difference is the difference in the deformation that occurs here under this loading versus the deformation that occurs here under this loading or the delta d formation which we have down here below and we take the probe on gauge block and the probe on ball and take the difference between them and we're out here to 0.062 micro inch or 1.56 nanometers difference extremely small but keep in mind if these pressures were different and radii were different things these could get significant numbers so we're we're still here just trying to understand the mechanics of what's going on so this is the difference in those now we come back to our last example which is what we actually have and this is going to read less by the delta deformation here plus this deformation at the bottom and that turns out to be down here where we have the deformation at the ball versus plus the delta deformation it takes us to 1.906 millionths or 48 nanometers so that's how you would go through the the calculations shows how easy it is to just make assumptions without thinking it through uh and there's nothing saying i didn't still make a mistake going here that somebody might find but i'm pretty sure this is correct so for high level gauge manufacture and measurement these things all have to be taken into consideration and uh compensated for so it's this isn't like uh you know ridiculous out in space type stuff this is something that happens with um any high precision measurement down in the millionth uh range when when single digit millions matter obviously one of the things that is important to realize it doesn't matter how many decimal places you put out here that doesn't mean that things are that accurate uh things like modulus elasticity aren't exact and that will change what some of these values are there's all kinds of things that are always approximations this is a chrome carbide gauge block and this is a tungsten carbide uh ball i'd use the same modulus for for both but i'm sure there's probably some minor difference and yes if you change that modulus elasticity by you know several thousand pounds um it will change the number when you're out to you know nine decimal places um but uh you always have to remember that you know using a bunch of decimal places doesn't make it that accurate um so these are these are still approximations but at least you understand the principles of what's going on and that you have to consider everything is rubber think of these one good way to think of of contacts like these think of this as a beach ball beach ball golf ball and this is a foam mattress all of these are all deforming this ball the the the um upper balls deforming the beach balls deforming the beach balls deforming on the mattress they're both complying some they get a contact patch uh and that having that um very mushy material mentality about everything even when it's tungsten carbide or ruby is important because it helps you mentally picture what's going on and being aware of it even if you don't calculate it being aware of it is important we always talk about how you have to think of everything as being made of rubber that everything deforms nothing is infinitely rigid and no matter how small the loading things are moving so this is a nice opportunity to show you how true that is this is a very rigid indicator stand uh yes there are bigger ones but in general this is a very solid unit doesn't has no or play anything you see moving here is going to be purely distortion i have a kimwipe here a single kimway little one that i'm just going to hang on the back like that and put it on and take it off all right on the back of that so you i just want you to see what i'm doing because i need to zoom in now on the indicator at one millionth per division you can see down here we're on the one millionth per division scale i'm going to zoom in here to the actual dial here and at this range these you also have thermal issues going on just my body heat and things are going there but i'm going to zero it i just took the kimwipe off kim wipe on oop fell off try this again kim wipe on can wipe off i'll zero it again so it's easy to see can wipe on can wipe off can wipe on can wipe off and wipe on can wipe off so you can see that minuscule little piece of tissue paper hanging on the end of the post back there all the way here at the back move that almost a full millionth of an inch so that's how sensitive things are at this level you have to really be aware of thermal things and mechanical things as far as deformations as soon as you drop down below 10 millionths things get really squirrely so if you didn't fall asleep through all that um you get a congratulations from me because um unless you're really whacked like uh some of us um this this is boring as all get out but uh i hope you enjoyed it here you see this precision ball in use in calibrating a precise conical seat edge establishing this ground surface that i'm grinding to exist on exactly a certain uh gauge line diameter where where the diameter of the cone is a certain uh diameter so that's what i was doing here i'm actually have my gauge block stack sitting right on the surface that i'm grinding the ball is in the tapered seat and then i traverse from the ball peak of the ball to the master and then back again to the peak of the ball and on the grinder you can find the peak of the ball just with the two axes moving back and forth in and out and left and right to get on the peak of the ball you

Channel: ROBRENZ

Views: 332,050

Rating: 4.9065638 out of 5

Keywords:

Id: fEoonCLTCbE

Length: 47min 25sec (2845 seconds)

Published: Sat Mar 06 2021