Machining Thin Parts! Part 2 WW139

[@hettinger1000]

Thanks John. I appreciate these kinds of videos because I often let my brain take me into unknown territory like this. I think you're about where you need it to be and it will work great.

[@xnoitulos]

Machine a similar groove as the Pierson plate under the outer diameter of the part and run that gasket strip. Love your vids!!

[@Bunnyonabender]

I was thinking something like an off the shelf o-ring to get seal as well.

[@PeterWMeek]

Air will still leak in around the spokes and the inner edge of the rim. Your gasket needs to run along ALL edges of the workpiece.

[@warwickben]

If using a screw to hold it in, make the out side of the radius taller so the center in lower . You can level it out by tightening the screw

[@StefanGotteswinter]

Screwing the part down would be my preference - I built vacuum fixtures in the past and they always needed a seal. Normaly I only go trough that trouble for parts that are very sensitive (Had to re-machine 500 painted plastic parts once, for those I made a simple 5-station vacuum fixture out of plastic. For that fixture I cut and glued soft foam rubber into custom seals that fit in a groove on the bottom of the cavities for the parts. I am still not convinced that trochoidal milling slots in aluminum is always the best solution - I absolutely prefer to increase the feed and keep the Doc low on a contour like the fixture you machined in this video.

[@TheWidgetWorks]

I agree, the only other way I would try would be to run a much higher flow pump and increase the depth of the vents. Also making it out of plastic would help with the sealing a lot but you still may have to resort to installing seals. I have a vacuum chuck on my router and even there I tape, screw, or spray glue material to it most of the time as its just so much less chance of it failing.

[@b3nsb3nz]

Totally agree on the slots, so many good tools out there can easily handle slotting and its so much faster even if you have to use multiple depths.

[@barrygerbracht5077]

How uniform was the final thickness?

[@Nicko4x4fab]

if there was no vacuum channel in the jig when u used the fastener approach it may sit flatter because when you pull the center down in the middle the outside edges will naturally try to lift, if the floor of the jig pocket was flat it couldn't pull down the centre of the part.

[@jasonward4480]

did you order the box of fasteners at 18:27 ? if so where from? thanks! big fan of the channel!

[@MrSidiox]

I love it when you show this kind of experimentation. Anyone can do it correct, few have the guts to show the learning process!

[@kristinamckeown5106]

I have used a special 3M double sided tape. Don't use coolant during machining and you will need to use light cuts. Use acetone to remove the adhesive.

[@KenToonz]

Excellent tips on walking it in with cutter compensation!

[@EZ_shop]

Great vid John. Something I noticed when you posted the code... your Brackets is set on "text" mode (lower right corner of the window). If you change it to "Gcode" it will automatically color code your file to make it somewhat easier to read. Ciao, Marco.

[@ROBRENZ]

Hi John, I would bore a set of full pie jaws on a lathe chuck and then bolt the chuck on the mill. ATB, Robin

[@Regalmetalworks]

I am really digging the cutter comp.....I ran a cleanup op on the mill from the plasma table today, and I had to keep adding stock to leave by 10 thou each time until I reached the final correct amount....it ended up being 70 thou over, so 7 trips back to the computer, this would of saved me a lot of steps! Luckily I have quite a few more to make, so I will be implementing this on the next batch! YAY

[@jasonlacoss7834]

extra wide set of shallow soft jaws (full profile) would do the trick too. if you cut the tolerance as close as those first fixture cuts, it should keep the pressure pretty evenly distributed with plenty of surface area to grip and very little if any distortion.

[@bcbloc02]

No problem make part from steel and hold it with a magnet. :-) lol Try heavy grease or if that is too nasty vaseline to get your vacuum to work. That finishing with the sandpaper would eat a lot of time in production. How about a vibratory finisher? I have used my concrete mixer and creek sand to do polishing before too. Just be sure to take the paddles out as they can ding your parts.

[@StefanGotteswinter]

Agree - Aluminium is so annoying, doesnt stick to magnet chucks :(

[@philipthomas9309]

When I use to work with vacuum all the time we would cut a small channel around the outer edges and put in some small dia neoprene rubber almost always got a seal, John did you connect the vacuum channel back to the suction point don't remember seeing a star to centre in vac channel tool path. your 3/32 channel just shy of 3/32 deep 3/32 dia rubber no problem just but fit. If the all outer edges are O ringed will pull down flat all around, Cheers

[@daniels1905]

Thanks for posting John. If I was doing that job in high volume I would try two concentric 1/16" o-rings around the perimeter and move the vacuum port. Maybe add an additional pair of o-rings in the center if it lifts too much.

[@beachboardfan9544]

Whats the red coating in your 770 bed at 19:28? Glyptal?

[@Dustinmikl]

I'm on the same page with the o-ring idea but I'm thinking radial in the outside wall of the fixture almost like a reverse piston ring. Just allow a one thou squish

[@tridium-go6hw]

John, the part is going to leak. No way around it. you need to be able to flow enough air to maintain atmospheric pressure at all points to maximize holding force. And you need the atmospheric force to act on holding down the part, not overcoming friction to force it into place. Start with a substantial vacuum pump. High flow (cfm) is way more important than absolute vacuum level. you need to evacuate unwanted air fast. open up the fixture until the part will drop into place on it's own, that way the vacuum will be acting solely to hold the part down, not to overcome friction. make the vacuum channel underneath absolutely as wide as possible. only leave the bare minimum shoulder needed to support the part without deforming. this is critical because only the area exposed to vacuum is creating down force. Every bit of shoulder supporting the part is taking away holding force. the thing will leak, so you need to remove air at a very high rate to maintain holding force. In addition to the high flow vacuum pump, you need large passages. make your vacuum channel as deep as practical and use a large fitting and hose to the vac pump, at least 3/8". Air is much thinner than liquid! consider running flood coolant. it will suck through the gaps much slower than air and increase vacuum level under the part. if you try this part you would have to include a sealed container between the vac pump and fixture to catch coolant that was sucked through and prevent it from reaching the pump. the best you can hope for is about 15 pounds force per square inch of part geometry, less depending on your altitude above sea level. if this isn't enough then the whole idea is moot.

[@jkotka]

Thanks for sharing this one, i wonder if adding an o-ring to the fixture on the outer diameter of the part would have helped, most likely not i suppose. would be interesting to understand where the vacuum was leaking off.

[@TheSmartStar]

We have a Datron at work with a vac bed. It uses a special card with a self adhesive grid on it to create small vacuum pockets between the card and the part. It'll hold something like this no problem. And the squish of the card is calculated under vacuum so it stays accurate in Z. Might be a cheaper alternative for an experiment.

[@gregghooper]

Try some oil or light grease on your part to act as a seal for the vacuum.

[@LarryRPark]

Laser-cut rubber gasket w/ perforations on the center lines.

[@billmoran3812]

John, I would try some type of grease to aid holding the vacuum. I think it would have held if you could reduce the leakage. Your vacuum pump just isn't capable of moving a greater quantity of air. Vacuum grease is commonly used in laboratory setups to reduce leaks at fittings. Worth a shot if it became a production run. Maybe with a bigger vacuum pump too.

[@Gaatech]

nice video as always. could you show how you use super glue to hold down parts

[@BradPow]

If you arent using d comps normally, how are you getting parts on size? Reposting? D comps are industry standard i thought

[@VKRenato]

BradPittlike Yes I ask my self same question. For getting h7 tolerance, he probably must generate multiple codes for same operation.

[@suzukichopper]

There's many ways to skin the same cat.

[@StefanGotteswinter]

Yeah reposting - complete Pita and proctivity killer. Had to run a machine once where we didnt have a postprocessor that could output compensation, that was a slight nightmare ;)

[@BradPow]

I'm not knocking him for it. I just wondered how he does it because d comps are very quick and handy to size up a part.

[@cncit]

In some cad cams for example Featurecam only the finish cut is output with cutter comp. The code is generated with the tool on the centre line of the contour and the D is used to call up a radius value from the tool offset table. I use it for every part I machine with my Fanuc controller :-)

[@TheWidgetWorks]

John, my recommendation with cutter comp is use wear or inverse wear not control. This way you will have less issues with lead in and lead outs not being long enough and causing either alarm or gouge. Also it's way less likely that you'll typo a tool diameter and crash, just program everything over sized like you did and then you only have to adjust to fit a few thou with each tool and if you enter the offset in the wrong tool # not problem, ex. no crash! Also then your simulation is still 99.99% valid as your lead in and out potentially is going to be messed with by the cutter comp move generated by the control. The bigger the comp move the bigger the canch of an unexpected change in the tool path that you can not verify till you push the 'hope this works' button.

[@Freak0NatureX]

Have you considered making your vacuum fixture with a softer material? Perhaps a plastic like nylon? Or perhaps packing it with a grease around the lip

[@jackr166]

Hey John, have you considered just running half of the part depth on one side, then running the other half after you turn the stock over using tabs for the second run​? You get a minimum amount of cleanup and a fully machined part. What do you think?

[@jackr166]

Easier than trying to get a vacuum to work.

[@arthurschroeder8230]

Would adding an o-ring groove to the fixture solve the problem with the vacuum leak? I just used some Harvey O-ring dovetail mills on a part, and they worked great. Wondering if something like that would seal it up. Obviously it would change the design of the fixture a bit, but that might be a way to get the seal.

[@WobblycogsUk]

Assuming there's no hard requirement for the part to be made from aluminium perhaps switch to steel and hold it down with magnets. It's like the parallelism requirement you mentioned, if it's not in the drawing don't put it in. If aluminium is required then perhaps you could design a hold down jig that lets you machine one half and then the other. I'm thinking something that looks like a radiation sign without the centre dot. After the first half is machined you could just rotate the hold down 60 degrees and machine the other half.

[@Jumeres]

Great video John. I've wondered about the G41 tool comp feature. I'm assuming that also works Mach3.

[@leysmetalmecanica1885]

Hi John, it would be possible to put an o-ring on the bottom of the outer circumference. Maybe this will seal the vacuum device. Greetings.

[@pabonrobert5119]

I'd cut o-ring grooves into the fixture, one around the outer diameter of the wheel face and one around each of the wedge shape areas around the spokes. You should seal against the machined face pretty well with o-rings.

[@dansharpie]

just a thought try making a water suction mini bath.

[@MrToolean]

John- why not face off a piece of plate and mitty bite three places on the OD?

[@joshpicard4139]

what about using a small woodruff keyseat cutter all along the walls of your fixture so you could run some o-ring gasket to create a vacuum?

[@qwertydworak]

Can you make the fixturesides slightly conical, and pressfit it down? Obviously with a press for even pressure. And not with alot of pressure of course :) a few holes throu to poke it out of the fixtureplate after.

[@Zwieback1]

Great idea love your videos. Here is what may help holding the wheel in your side two operation: 1. Move the vacuum grooves to the center away from the sides. 2. Use a heavy grease to help getting a better seal.

[@AlexTaradov]

I feel like some sort of soft jaw arrangement might work. It has to apply pressure on the central hub of the part.

[@koloec94]

wouldn't you re-machine the fixture without the added depth for the vacuum?

[@davidfredericksen5933]

liked the video. I have done a lot of machining in vac fixtures just as you did here and it works better if you use an actual vacuum pump to draw the part down and the more CFM'S you move the better. The trouble you had with the comp. erroring out was because you have to have a G01 command at least as long as your compensation value in order for the comp. code to work. I always prefer to use wear comp for that reason. I can use a short G01, an arc lead in and then the part geometry. Really enjoy your videos.

[@jaypierson5955]

David Fredericksen Still not a good part for ANY vacuum pump since the surface area of the part is so low. The max holding power of a vacuum pump at sea level is 14.9 lbs per square inch. It was worth a try to learn.

[@davidfredericksen5933]

Jay, agreed, the surface area is not ideal, but I believe with some "downcut" tooling and specific strategies it could be pulled off. You never know until you try... right. If you would have told me 25 years ago we would be cutting parts at 600ipm I wouldn't have believed it.

[@ytnjw]

I can't remember if you chamfered the first face of the wheel, but if you did, how does the size of the chamfer compare with that of the lands at the bottom of the fixture plate?

[@ExtantFrodo2]

I think I'd sooner go with casting to make the holder rather than machining it. Also, wouldn't you need radial grooves from the center hole for the vacuum to access the "wings"?

[@cag66]

What is the surface area of your part? Would it be useful to machine the fixture out of a hard rubber to help with the sealing?

[@troyam6607]

could you maybe do multi point suction so like 5 - 4 mill suction points on the part?

[@liveepically]

a groove around the outside of the diameter of the wheel with a thin O-ring. The vacuum will compress the o-ring to where the part fits flat with bottom of the fixture.

[@mattyfromlondon]

drill holes along the axis of the spokes of the wheel with a counter bore say .5 mm deep then use small o-rings in the holes then you can use the smartvac to hold down the part without loss of suction :)

[@jasonlacoss7834]

what about a light coat of Vaseline or grease around the bottom perimeter? might be just enough for the vacuum to pull a seal like a gasket.

[@JackDanos]

Hi John! Love your videos - I just got a 770 and made my first cuts in Aluminum yesterday. I'm using the Tormach flood coolant system with the Durakut 4010 they sent with it. I left it wet last night and woke up to rust on the table this morning! I've read that I did a few things wrong like I should be using distilled or RO water instead of tap, and I shouldn't leave the table wet overnight. Do you have any suggestions on how to prevent rust? I've read that coating the table in WD40 after using it will help - any suggestions would be greatly appreciated!!!

[@MindAkerKnives]

I think using tabs would be a good way to go (thick enough to allow for bottom side ops), than flip the material before cutting the tabs, than mill to thickness, than chamfer, than add hold down bolts through center hole to the fixture with thick washer disk to help, than finishing contour to remove the tabs. You could use the center of the stock as the origin so when you flip the part the toolpaths "match" up no matter if the stock isn't precisely what you entered in CAM (learned this the hard way when flipping parts), just make sure you locate it parallel with the X axis with dowels or something in the pallet. A table sized pallet with the needed tapped holes could be made to help production along with the washer disks.

[@royletterle2082]

Spray fixture with plasti dip -> Re-machine top of fixture, side walls and the groove. Just leaving a small ridge in corners of the spooks may create a minimal seal

[@williamsquires3070]

Okay, now what the frip are you going to do with that little wheel?!? I vote for putting gear teeth on it and using it to make a fancy, mechanical clock! :)

[@AJMansfield1]

Here's a way you could do it, although it would involve reworking the earlier steps: - drill a hole in the center of each of the spoke holes, and four holes into the corners outside the part (used for holding the part down later) - cut all the edge chamfers on the current side (essentially, you'd just be cutting a v-groove into the surface where the part's edges will later land) - flip the part over, and mount it with cap screws using the holes you drilled earlier into standoffs of some kind - cut all the chamfers on this side - for each spoke hole, plunge the mill all the way through the blank and cut out the perimeter - plunge the mill through the blank at the outside edge, and cut out the outside perimeter

[@AJMansfield1]

Another way would be to take care of _just_ the center holes and spokes on the mill, then put it on a lathe and turn the outside edges.

[@mrc4nl]

Put a small layer of silicone sealant around the outer edge of the fixture plate. let it dry first so it wont stick to the part. the vacuum itself shoud press the part harder against the seal making the vacuum stronger.

[@christurley391]

Does high viscosity oil fix the lack of vacuum?

[@apokefaleart8918]

just an idea by me. you could try milling your cuttings of your fixture plate bigger, so you have more space between fixture and machined part. and the use mats you use on a flat vacuum table for the gaps. I hope you understand what I mean. greetings from germany.

[@chrisdemchalk3491]

Crazy idea I tried on similar project. Use a very small amount of shaving gel. It seals and washes away with water. The beauty is its change between gel to foam that seals the small void.

[@swolath]

what about some sort of sticky spray that can hold the seal?

[@hockeygeek21]

Laser engraver would be great to cut a custom rubber gasket.

[@r.j.sworkshop7883]

As others have said, apply something to create a seal, however if you were going to do 100 or 1000 of them you would need something that is easy to clean off in a batch and something that doesn't build up on the plate as well. I don't have a specific answer, but I am thinking of the tire mounting paste that is used to install car tires. It has some body to it to create the seal, and to remove it you could put 50 or 100 parts in a warm water bath. The difficulty would be not getting chips stuck in the vacuum plate during the part change. Good Luck.

[@jarisipilainen3875]

19:55 your part not flat. you mill it between vise jaws it warped. so suction didnt work also. clamp it down

[@gustav901]

On the fixture, try making a lip around all the edges and then use silicone to fill that lip. If you make the edges of the lip spotted or knurled, the silicon should stay in place.

[@jwardWWOPD]

you should have the vacuum seal only on the rim, not the spokes. and for the fastener you should relieve the center so the part is pulled against out side rim just before it hits on hub in center, or spokes

[@Sketch1994]

I would try a tool with an as negative as possible back rake angle so that the cutting forces push the part to the fixture.

[@adithmart]

Well, is the skin on the backside warping your part?

[@886014]

G'day John, with all due respect I personally think you went for a complex solution when an easier one was available. I'd suggest when fixturing use the features available. You have a centre hole plus 5 other "holes" available. Fix to features that will later be machined away. PS If you wet your abrasive paper it will stick to the plate and you don't need to hold it.

[@johnalexander2349]

Amateur here: I feel that the exact shape of the vacuum fixture is working against you. The part should touch the fixture as little as possible - around the edge, a few supports (to stop the vacuum dishing the part) and a few pins for indexing the spokes - everything else should be milled away. This has the added benefit of increasing the surface area that vacuum is applied to. I'll try draw it if this isn't clear, but I suck at Fusion.

[@johnalexander2349]

Disregard that. You'll lose vacuum as soon as you start cutting. But I still think that a less exact fixture will allow the part more room to conform.

[@MrBastor87]

Hey nyccnc, can I ask why are you making these out of sqaure stock and not just round?

[@kenfetter8445]

would it be possible to wipe some silicone in the lip it rest on to help seal the vacuum but not enough to through off all your other work

[@joshuwa88]

High production of thin parts like this are perfect for a cnc router. You can get one with a 24-30,000 rpm spindle and blaze an "O" flute through there at a high chip load. Maybe your next machine?

[@playdav485]

hi would some sort of grease seal the vacuum fixture

[@rdstngry]

What about a thin bead of silicone on the vertical faces, too seal the vacuum fixture? Maybe something in the RTV family like Permatex Gasket Maker. And it is entirely possible that the only reason this came to mind is that I am a Gearhead.

[@gnather]

I am a absolute beginner to machining. However I am a longtime user of Cyanoacrylate Glue and there is a caveat to using accelerator. Ithink that it weakens the bond.For maximun strength it needs to set on its own. I use it to fill seams in plastic and if you use the accelorator then it seems to sand out much easier than if you allow to set on its own which causes it to be harder than the plastic around it. If you are going to be using it in a high heat application I might be concerned about using Zip Kicker. But you guys probably knew all that anyway. I did say I was a total begginer.

[@paisleypcdoc]

I love your videos. Your sharpy demonstration of tool deflection raised an eyebrow...If, for talking sake, it milled perfectly (no deflection, 100% accurate), and then you added a thin film of sharpy. If you ran the same gcode again, it would remove the sharpy, obviously. Same idea as when you ran superfly over your powder coated sign.

[@robertpoirier5157]

Wouldn't be possible to cut a grove for a gasket around the edge of the part holder then find a flat rubber gasket material and cut it to size with the Tormach to fit that grove? Kind of the same principle as a regular vacuum plate but using a custom cut gasket instead of regular o-ring like material. I wish I had a Tormach mill to try this out. It would sure be fun to try ! The challenge would be to hold the gasket material for the cutting operation but it would achieve your goal.

[@robertpoirier5157]

Just building on this same idea... There has to be some "pourable" gasket material that you could lay down in that grove then machine to size so it creates the seal you need. Worth a try since if that works, there would be almost no limitation to what you could hold with a vacuum.

[@jeffcostello6747]

If this was a production part and the order was for a few hundred, rather than just a few, it would be nice to have a part tumbler to break the edges and create a uniform surface finish after decking the back side, rather than worry about clocking the part and creating a custom fixture. Could this perhaps be made on a lathe? Yeah probably. Would that be faster? Maybe. Probably would need a few operations on the mill to cut out the spokes for instance if you didn't have live tooling. How about a die to just press the part out. Would that be faster? Surely. But would the parts meet spec? Depends. Which means you might require more finishing operations??? What quantity of parts would justify making a custom die? The rabbit hole is practically endless. I love videos that raise far more questions than answers. It's great to get thinking and awesome to see your approach. Thanks for sharing.

[@gremlinsports]

same thought an oring simple cheap. might also use soft jaws or simple expanding mandrel.

[@underourrock]

Scaling production: Machine your first op like you show in your first video. For the second op, make a fixture that has 2 pins set to locate the same machined corner of the stock. The pins will be located on the bottom left or the top right of the part depending on which direction you decide to flip it. Use the pins to set your part location. Zero is found off the same machined corner that you used on your first op. Use superglue to secure the part in place. Include the cost of the superglue in the cost of the part. Profit. Any other clamping method I can think of would be way too expensive in terms of mitee bite clamps or components and time. I mean if you had a crap ton of the right kind of mitee bite clamps, you could clamp against the sides of the flat features for the second op, but loading and unloading that fixture (from the bottom, no less) would be a royal pain. The labor it would take to load and unload could be invested in superglue and still save you time.

[@samgerlof9848]

Maybe try some sort of silicone sealant on the plate? or vaseline..

[@jesseblack0713]

I would put it in a lathe and face it off.

[@EUnit111986]

Jesse Black I agree. Easiest solution with minimal effort. As long as your jaws in the lathe indicate flat, it's just a matter of how fast can you manually swap the parts for quantitie runs

[@nraynaud]

I was thinking the same, but there is still the issue of chamfering the back side of the spokes (45° very small dovetail cutter? that seems shady).

[@RedneckTech]

I think you should try some grease or similar

[@RambozoClown]

Another option would be to use soft jaws milled close like your plate, but then open up the OD a little and instead of clamping, pull the vise open to give tension and clock them on the spokes.

[@collinsimpson4933]

maybe try teflon tape for the vacuum seal.

[@robertroy3560]

Another option is to use a lathe chuck. Cut your fixture to .010 less than the part in diameter and use it as a backer for the part when you chuck it.

[@boxoftj1]

that looks pretty good! Why not use your Fixturing plate, though? no extra fixturing needed. that's how we make flat thin parts where I work, multiple holding points if flatness is a big deal. Definitely not fast, tho, and I know that was kinda the purpose in this...

[@ocmooz]

Just make blanks on the lathe (finish o.d. and thickness) and cut out the windows on the mill. Quick circular soft jaw with light torque and you're all done.

[@MattOGormanSmith]

How about making the whole fixture out of HDPE or Acetal? (the copolymer is slightly softer than the homopolymer Delrin, and half the price)

[@prodesign8189]

If you didn't have a slide fit or looser, you are not helping the vacuum. I think the interference was fighting the vacuum. I like that you at least tried. That is the way we find better ways to do things. Thanks for your vids and time dude.

[@ikcalB]

I'd suggest a looser fit too. it slip in/out by itself

[@meansq]

I would use a small amount of gasket silicon glue to make the vacuum setup work

[@endrachwer]

when sanding on a flat like that try moving it in an " 8 ". that way you get less scratches and a more flat surface :)

[@learningeveryday5198]

you could surface the part and in the same setup machine the countour then flip it surface the stock and machine the pockets so easy

[@JayDee28]

I would have cut a plate to thickness, then the inside contour, then the outside contour after refixturing, then chamfering on the backside. I would guess that gives you much more repeatable results and better thickness consistency. imho you will only get the vacuum to hold with o-rings underneath the spokes.

[@zeevyehuda2501]

I'd just pour some flowable silicone around the edges, and use the vacuum fixture. That way you have a tight seal, and the silicone comes off pretty easy

[@OriginalJetForMe]

Route a channel for a gasket?

[@paulgrep3193]

Well after watching so many of your videos I never noticed that Fusion360 was doing all the tool offsets! Always done at the CNC end in my day.....

[@zuknivek]

When I use circle interpolation to cut a diameter, its never exactly round. Depending on the machine I use, its anywhere from .0005" -.003" oval. I wouldn't think .0005" would cause an issue for the vacuum but I've never used one. I would try cutting a diameter in another piece of stock to see if your machine cuts an oval like my haas.

[@jjflounder1]

small "O-Ring" around the OD and each hup .... McMaster-Carr sells small o-ring stock . that can be cut to length

[@jjflounder1]

McMaster-Carr Ft.1/160. .070"Durometer 50A (Soft)ASTM D2000-25° to 212°Black 9864K23. $0.50 per ft