CNC Finale: Automatic Toolchanger Spindle Install & Testing

this video has been supported by Megatron lean back relax and open wide for the new cnc dentistry robot while it's working I'd like to show you how it all started this cute spindle motor was the first of many CNC parts that I accumulated in 2018 but somehow it didn't grow like the other things I guess I have to water it some more for that I didn't want to bother with a proper water cooling loop so instead I just used a kilometer of silicon tubing [Music] what could possibly go wrong it's neither a rusty shower for my exposed steel surfaces nor melting of the 3d printed holder I'm pretty sure I've configured this VFD correctly last year but maybe it forgot I don't know what happened to there precisely but it cost an internal shot in the windings of one phase I've made a half-hearted attempt to save the thing but no luck no drama either another spindle candidate is already on the way and now we can take this one apart violently out of the water cooling sleeve came the AC induction motor rotor and most importantly its bearings the first one on the connector side of the spindle is just a single normal radial ball bearing and that's okay on a CNC spindle however axial forces can easily exceed radial ones so just like with our ball screws a fixed bearing block is needed at a cursory glance it looks like it's right here equipped with a pair of angular contact ball bearings that would be beautiful but in reality they have just disguised to more regular not even deep groove ball bearings disappointing and good riddance I took this brief spindle story and presented it to the experts at make a Tron manufacturer of higher quality spindles and in response I got a mocking a reproachful box of coolant yeah I guess I deserve it for destroying that innocent China spindle but wait there is more a customized water chiller to circulate that coolant through something with their company logo engraved on what must be the most elegant cnc'd tank cap ever thanks Megatron that is literally very cool oh there's another little box from them but that's probably just accessories and hoses and stuff and the usual paperwork there's also this aluminum holder some huge hsk e25 tool holders and a set of ER 16 collet chucks in a suitcase what a strangely specific set of accessories it's almost as if they knew what kind of cnc spindle i were to try next no idea how they found out but here's my new megatron ATC 80-year that's a water-cooled 2.2 kilowatt automatic tool changer spindle and they also provided a lovingly handmade mixed domain control cabinet for it by that I mean it acts both in the electronic and in the compressed air domain it's all plug and play ready to go but that has never kept me from taking a closer look at everything let's start here v1 is just a mains filter that prevents high-frequency interference from leaving the box the VFD is a good imran MX door with field oriented control meaning that it doesn't just playback pre-recorded waveforms but actually takes a rotor position into account when delivering face current it won't deliver anything though unless the red safety relay gives it the all-clear it's meant to be configured in an all-encompassing safety circuit that only allows machine operation when neither water chiller and nor servo drives are in an error state when no emergency stop buttons are pressed and when the operator has passed a breathalyzer test the pneumatics on the right are one input filter regulator and three or four switchable outputs the blue lines in the complete disregard for good layout and mounting practices show where I have contributed something to this system I wanted another switchable error output for mist cooling that will be by far the biggest compressed air consumer with dozens of liters per minute for that the air doesn't need to be particularly clean so I'm just going to take it from the input directly without err unnecessarily polluting the nice Festo filter regulator to find out where exactly the three original air outputs are going I'm afraid we have to perform another violent spindle teardown this beauty is going to be a challenge though being all full of pre-loaded and precisely torqued precision parts and sensors these connectors alone are absolutely stunning well let's not delay time for the first high end spindle taryn on YouTube I've already prepared the usual high end spindle teardown tools for the occasion but in the very last moment I got an email from the manufacturer with this incredible three the sectional drawing that way we can do it all virtually and safely without damaging the goods at the top there is a series of pneumatic ejecting cylinders they are reset and held in position by these spiral Springs when the cylinders are pressurized the springs are overpowered and the central rod is pushed downwards down there there's an hsk e25 clamping system that looks similar to this one its teeth grab a tool holders from within so when the thing starts spinning a growing centrifugal force is added to the normal spring-loaded clamping force that makes hsk clamping the hottest thing right now not only but especially for high rpm spindles hopefully not hot real angular contact ball bearings with a maximum 42,000 rpm for which this spindle is rated these guys have to be special special and clean there's a labyrinth seal between rotor and stator through which a small amount of compressed air is constantly flowing that way no dust particles can enter the motor is an AC induction machine it's state or has turned in ground and pressed into the water cooling sleeve before that it looks similar to this last but not least of sensors which detect rotation and the state of the clamping system that will be very important when we are making an automatic tool changer for the thing for now though I'll just focus on the installation this all-around carefree spindle package has been delivered set up and ready to go so all that's left to do for me is some machine alignment and trimming the x-axis has to be parallel to the clamping plate and the spindle has to be perpendicular to it sounds pretty easy right well on a micrometer level nothing is the spindle shaft temperature can make it grow and shrink sufficiently to make a difference the final quarter turn of an already snug screw can ruin and alignment completely I've modified it a beautiful spindle holder slightly but then screwed it down onto one of its own chips huge impact in the micrometer Department time for the maiden flight this is the rocket launch like starting procedure that makes sure that all subsystems are ready and we don't end up exploding during ascent I thought I had already worked pretty hard on this whole trimming thing and wanted to put it to the test by slowly facing this aluminum piece and looking at the resulting surface finish well here's what came out it's reasonably flat I guess but the spindle is not yet perpendicular not one bit the entire gantry had a slight forward nod not quite as dramatic but you get the idea now the easy thing to do would have been to shim the motor holder and to compensate for the north that way but who would choose the lame easy way if instead you can wrangle back and forth a few hundred kilograms of Steel for a day or two I did that until I measured four to six microns back to front with a fully extended lever gauge with a smaller diameter tool the deviation will be even smaller of course afterwards I did the same thing the side to side I haven't reached perfection yet but I'm better than the parallels and the precision device that I have available at the moment so I don't think it would make a lot of sense to go any further right now although I've got a pretty strong urge to do so I'm not gonna lie throughout this lengthy process I somehow managed to maintain x-axis parallelity as well and I think my machine will be able to maintain all of these adjustments in almost all situations to for example I don't expect to normal cutting forces to ever accede to the 25 kilogram which I'm pulling here so the only sources for higher gantry deflection are machine inertia in abrupt movements resonant vibrations and accidents but I'm not planning to have any of those so we should stay within a single-digit number of microns usually that's about the diameter of a red blood cell so I'm happy with that for now let me just redo the surface finish test real quick after this epic effort it better be good these are also some very conservative feet and speed parameters and the tool a - flute solid carbide end mill is almost new what's not working in surface finishes favored here is the back-and-forth lawnmower mode that'll leave the usual CNC tool marks and make things look a little less perfect than they actually are and you know what they actually are pretty damn good know you can partially see the previous lines of tool marks through the step-over and a brief touch with a flat stone reaches the entire surface I couldn't think of any better method for showing surface roughness than this other than buying a surface roughness meter but let's finally start making some productive chips after the 35 degrees C tomorrow it's getting really cold in here now I'm machining an acrylic light guide for a photodiode the shape is deceptively simple to parallel a concentric squares one of them a bit smaller to make them super flat I'd like to machine the sides only with the side of the milling bit that works beautifully for the first three sides it makes a terrible mess though just like everything else I'm going to do on this machine I'm afraid then near the end the part kind of hangs from its last thread and breaks off in a not very clean way the result is mostly beautiful which is not a big deal with an easy material like acrylic better workholding and a finishing end mill would have improved it even more next similarly bad recording and only marginally more difficult material aluminum I am trimming down a stack of these sheet metal ucroe files which will become telescopic way covers with everything still being new I'm not moving particularly fast yet but I am removing quite a pile of material here not really a spindle related topic but if there's enough interest I can make a separate video about these let's do more and faster this is going to be an adapter to connect a hand wheel to an old welding transformer unfortunately this random piece of stock material with a beautiful surface finish is way too big so we've got a lot of material to pulverize doing that with a precision high-frequency spindle is like going to the dentist for a pedicure it's technically possible but not really meant for that with a nine millimeter - flute solid carbide end mill at 15,000 RPM 10 millimeter depth of cut and 1000 millimeter per minute feed this is how much aluminum we can shred 0 millimeter pretty easy [Laughter] 0.1 millimeter a good value for a clean finishing pass 0.2 millimeter 0.5 millimeter please stop judging my serve or window cutout 0.75 millimeter and that 3d printed bearing block is also pretty gnarly isn't it 1 millimeter let's machine a new one later one and a half millimeter two millimeter I should probably calculate and adjust my parameters for optimal chip load in each test but I'm not even sure how relevant these tests are for the automatically generated adaptive tool paths that I'm going to run on this thing three millimeter starting to chatter a bit [Music] now I could add more flutes or more than double the RPM to make this easier but realistically I don't think I will ever need such heavy cuts not exactly steaming hot a bomb 79 size chips but still fantastic I don't have the right tool for that central hole I'm afraid the normal hardware store twist drill certainly is not it it would be comfortable around one to three thousand rpm while the spindle has most of its power at ten times that speed still with extra slow manual jogging we are getting through and that's okay for a one of low-tech part did I hear adaptive toolpaths let's make one more aluminium part but microscopic this time with microscopic tools high rpms and high-tech optimal load tool paths I have no experience with this whatsoever so there will be some microscopic casualties I'm afraid that final tap was game-winning check this out I want to try and use these solid tungsten carbide PCB bits they may not have the perfect geometry for metalworking but they are cheap in large quantities and they break easily when something is going wrong so my lack of experience doesn't pose a threat to that fancy spindle [Music] in fusion 360s chem I've configured no less than five operations to make this part first 3d adaptive clearing to remove a lot of material quickly but with 0.2 millimeter stock to leave for a later finishing pass this will utilize the entire nine millimeter length of the flutes with their two millimeter diameter that's certainly pushing the envelope a little bit next another 2d adaptive that spirals out a bit for other than the parts actual silhouette as a preparation for the 2d contour cut out then while I still have a good grip on everything I'm finishing this internal surface by removing the last 0.2 millimeter doing roughing and finishing with the same tool list not a very good idea at all but you know this is all still very experimental then I'm going to try and cut out the part with a 2d contour operation that is not a modern a smart strategy at a traditional lateral slot milling and then finally I'm going to Peck drill that central hole was the 200 micrometer drill bit and hopefully not end up having to watch it Bend okay here goes nothing the entry helix is way too slow and the optimal chip load setting is way too aggressive rip my first tool ever on this machine okay better whoa whoa won't calm down dude [Music] oh that still feels like way too much material for such a tiny end mill but in general the utilization of the entire tool length that one of Megatron's experts has recommended to me works is really well for better chip evacuation I've gotta avoid to these narrow deep holes though oops what was perfectly obvious but completely under my radar until now is the kind of chips that a small tool with a high rpm spindle produces in collaboration with a mist cooling fluid this fine aluminum powder likes to stick everywhere and well enough to not just be vacuumed away it's really unpleasant actually and it makes me want flood cooling even though that would be incompatible with my oil lubrication and just needing a lot more work in general here's what I originally wanted to be a finishing pass I generated that with a relatively new steep and shallow strategy in fusion 360 and as it turns out it does more work than I would like a gentle finishing pass to do it makes the first breakthrough us in these spokes even and then kind of hastily cleans them up that's not a catastrophe though and the next time I'll know more and to know the sound of 30,000 rpm for that 200 micrometer drill bit oh yeah it feels almost as if I've got to be careful with a missed cooling too much pressure and we'll send that dainty tool flying not really this turned out to be the easiest operation in the end half a millimeter per peck with full retraction and I didn't break a single one of these bits this kind of micro machining really benefits from Megatron's world-class run-out specification of less than 2 microns for this model otherwise such a tiny drill the shelf life is about the time it takes to travel from tool changer to material but here we just made a four millimeter deep hole without a problem alright how and especially what did we do is this some kind of bling alloy wheel for a 200 micrometer axle no it's a fiber lapping pack in a long ago fiber coupled laser video I tried to polish a termination handheld that resulted in rounded corners and subsequently a burning termination this thing will hold fibers perfectly square while lapping and that'll hopefully result in much better terminations and a few more fiber laser adventures okay that's no fun anymore this finger numbing cold doesn't deserve happy Christmas tunes either there's actually another conceptual problem that I haven't thought about before moving here we haven't even reached sub-zero temperatures but my oil viscosity is already honey year and the central lubrication pump is stalling let's work really hard to warm up this shed time for some steel I'll keep my phone in a somewhat safe distance here so that I can at least document the aftermath if things go horribly wrong this is the same sharp - flute solid carbide end mill theoretically with its geometry it can plunge and cut full slots to a certain degree so some superficial scratching on this piece of carbon steel should be alright I hope we're not really cutting steel here but mostly that delicious brown crust with my old aluminum CNC I couldn't even think about brown crusts without it losing steps so this is already quite the accomplishment but after working on this thing for a year I am expecting a little bit more hold on to your heads I'm taking an entire millimeter now Wow so quiet and effortlessly fantastic for me and those rare steel occasions that would already be good enough but for the sake of flexing I think we've got a lot more potential although flexing is kind of the opposite of what we want here yeah 1.5 millimeter depth is still easy but it doesn't sound as nicely anymore oh no the chips are even nastier they embed themselves into whatever you're wearing or skin if nothing else is available not that I have conducted such an experiment mind you it's still way too few degrees C in here but the beauty of a freshly milled piece of steel s worth it so Island or the chips today and for the next four days as well and whenever I want to wear this sweatshirt again I can even run such a spiral plunge right into the material with reasonable parameters but I had to pull the emergency brake there and that sounded terrifying I bet the neighbors in a 500 meter radius will agree that was the second appearance of the overly enthusiastic default values in fusion 360 let me try again over here with one millimeter optimal load instead of 2.8 [Music] Wow look at that grown up cnc performance right there this is such a big moment I kind of knew that my machine frame was pretty strong but pretty strong was more or less all the data I had I never simulated anything so seeing this steel shredding performance as a huge relief it's almost fireworks worth it don't you think oh look we are getting happy colourful chips no temperatures must be rising a what you mind not doing that please what what the hell was that did they put stone in their steel there isn't even anything out of the ordinary down there maybe my mists cooling didn't reach all the way into the bore or it was the Revenge of the infamous brown crust and whatever case hardening happens to go along with that at any rate or the conclusion that I'm going to draw here is exactly the one that I was hoping for the dream team that is my cnc machine and the maker trans bindle can take on steel without slowing down too much by doing so we can create beautiful surface finishes however and almost perfect geometric accuracy almost I wanted to finish this video by making my own bearing blocks ideally with the same fit as this one but somehow my solid carbide end mill got dull all of a sudden so before mauling this beautiful piece of pre ground steel I've got to get that sorted overall this spindle is an absolute highlight and utilizing its tool changer will keep me busy for a little longer if your cnc machine needs an upgrade as well check out the manufacturer megatron who has high-quality spindles in all price and power ranges in their standard line they have liquid and air-cooled models this one being one of the latter none of them have fake bearings or black out causing defects like the other one that I've taken apart in the beginning of this video did but they are still reasonably priced for Hobby machines and specified for more than that they also have these tool change adaptors with which you can transform another spindle and give yourself a huge productivity boost running multi-tool programs autonomously know long having to measure offsets and so forth I am also working towards that but for now one last silly experiment this blue bit is just as mean as it looks it's a six flute solid carbide heart milling tool so let's see if we can take a bite out of that tuff machine tab oh of course we can I think when working slowly enough we can take small bites out of everything even completely innocent er 32 chucks all right I think that was enough destruction for today thank you for watching and see you soon

Channel: Marco Reps

Views: 590,850

Rating: 4.8996096 out of 5

Keywords: marco, reps, cnc, spindle, diy, electronic, steel, 2.2, kw, omron, water, cooling, atc, hsk, vfd, phase, repair, teardown, cad, review, carbide, hss, aluminium, acrylic, chiller, mechatron, igus, cables, tool, holder, changer, control, cabinet, safety, circuit, pneumatic, clamping, milling, machine, metal, working, servo, arduino, raspberry, pi, relay, angular, contact, ball, bearing, linear, rail, screw, rotor, motor, induction, tramming, micron, µm, micrometre, alignment, way, cover, facing, vise, fog, mist, flatness, finish, feed, speed, chip, load, fusion360

Id: vB_FvkllLPc

Length: 30min 32sec (1832 seconds)

Published: Thu Nov 28 2019