I am a chemist looking for a method to prototype electrochemical cells for microscopy. 3D printing would be ideal, but all of the plastics, etc. that can be readily 3d printed have poor chemical stability for these applications. I mill cells out of teflon, hdpe, and/or acrylic, so a CNC mill could be a great addition to my lab.
Would it be possible to machine these materials on a Nomad? In general, the starting materials are no larger than 2"x6" (W, L) and max 1.5" deep cuts into the material. I could provide more details, etc., if that helps. I am not new to machining, but have never used a CNC mill before.
My group studies electrocatalysis. I look at electrochemical process like electroplating or electrolysis at high time resolution, and would like to be able to rapidly prototype cells for different electrode designs.
This is the largest single piece I would like to be able to make – I’ve made it by hand, but our shop is extremely old, so the measurements are not very accurate. The starting piece is 2"x2"x2.5" teflon. The deepest cut is 1.5" for the main chamber. The o-ring cut and through-hole are for attaching another chamber.
The Z axis will be your limiting factor. I would guess, and only guess without seeing drawings or a CAD model, that you will be at the limit for the nomad, but could make it work, possibly with design tweaks. A 2" height, with 1.5" depth of cut is maybe past the limit.
If the configuration is such that there is a pathway to clear the tool entering from the side, you can do a bit better, with tool changes, and of the configuration is such that you have clearance for the spindle end (collet nut on the nomad), which generally will be a sloped or stepped boundary, you can do better.
About the best I have done (and this is not the limit for the machine) is a 60mm tall part (2.5") with internal porting to depth of about 45mm with a full height square face. The part has an open side that provided clearance for tool entry. Without that, I have made 40mm depth through 40mm material without a problem, though tool setting is pretty critical. Yes, this left pecker tracks in the bed. No, it wasn’t intentional. I planned on about 0.25mm clearance, and missed with the zeroing.
Just realized I missed your reply with the PDF (either because I was posting when you submit it or because I am oblivious… I’d bet on door number two…)
The sketch you put up is maybe possible with the nomad as sketched. The through holes are not possible. The well might be, but it is at the limit, and I would need to try to see if it is. There is a little more than the 3" (76.2mm) spec for the Z axis, but not much. I have done 80 mm (40mm material and 40mm tool extension), but it was dead tight. Your sketch shows 3.5" to clear the top edge (you MUST have clearance or there is no way to set zero and to zero for tool length) which is 89mm total. I don’t know, and am not in position to check right now, if there is 9 more mm than I have squeezed out of the machine.
If modification of the design is possible (well depth, at the edge in particular), then you should be in. If not, I’d want to get on a machine to try it first.
As it sits, the order I would use (others might go a different order) would be drill the through holes first on a drill press, as the most likely to mess up operation, then machine the well and tap drill or tap bore (the bore is about 0.150" for 75%, or 0.162" for 50% thread) for the 10-24 holes in the first fixturing, then reset the part and bore the opening in the end through and cut the seat for the O-ring. The tapping is not practical on the nomad. Do it by hand. I would pop the $100 for a hand tapper to make it easy, if I was doing more than a couple, rather than do it with a tap wrench.
The bore for 10-24 is practical with a 1/8" (0.125") tool, but would be better done with either a 3/32" (0.93") tool as a bore, or drilled. The nomad can drill, but it is easiest using 1/8" or 1/4" shank drills (often seen as circuit board drills), rather than standard drills. I would use a 3/32 diameter center cutting flat endmill, myself, with a helical entry to full depth leaving 0.1mm (0.004") for a finish pass. The target thread, for me, would be 60% in acrylic, and 70 to 75% or PTFE or HDPE.
Thanks for the advice. I didn’t expect for the through-holes to be possible. In fact, I didn’t bother with them on the as-built cell either. With a computerized mill, I can easily redesign this to have two pieces with an o-ring between them as well.
It looks like your pocket has sharp corners. That is impossible to mill, however if something with rectangular sharp corners must fit in the pocket that is square you could just mill the corners out to allow it to fit. Another option would be a finish pass with a very small diameter tool but the smaller you go the more likely it is to break the tool and the slow the job will be. Of course truely square corners are still impossible with that method.
For the threaded holes I would just do them on a drill press, if they are not critical for alignment, if they are to be used for alignment it might be better to align with dowel pins anyway as then they would only need precision on a much shorter length and you could mill the dowel pin holes on your nomad.
For the cover is there an oring or gasket too (maybe there needs to be?). It’s just your comment about making this as two parts with a gasket between (great idea) made me think about a gasket or oring for the cover.
Another option may be to not have a cover and just make it as two parts, each with a pocket, and again if allignment is critical you could dowel pin them so you can drill the screw holes on a drill press. Ideally don’t make threaded holes at all and use nuts instead as it saves the time tapping holes You can counter bore for the nuts if you want this thing to sit on a flat surface.
The inset nut idea is a fabulous observation — the PEM nuts on the Shapeoko 3 are a huge time / cost savings and one of the things which makes it possible at its price point.
One other thing — this is a really deep pocket — chip clearance is going to be a challenge to say the least — wonder if you shouldn’t mill the end hole first and connect a vacuum to it to clear chips or something.
Right – the sharp corners are not necessary - just easier to draw. I realize that if you give those to a computer, it will try and fail to make those cuts.
The through holes, etc. are not necessary - I just want to be able to automate the main cuts (chamber, o-ring grove, out port). If I could do those, I could design even more complex cells with smaller-depth cuts and better seals, etc., just like you suggest. The counter-bored nuts are a great idea as well.
Can you also make it in two parts as I suggested or would that fail to contain the chemicals? The pocket might take quite a long time (but maybe not too long as it’s only plastic you are cutting) and as others have said will need a long reach endmill. However If you need to be able to have the full depth, you could still make this in two parts with a gasket/o-ring in between and then just screw it together. You could even machine locating features into the plastic to make screwing it together more precise and removing the need for pinning. But pins are probably the most accurate way to go. I use a handheld dyson to remove chips on deep cuts like your pocket but it requires a fair bit of manual intervention. Wills idea would be much more elegant
Finally the nuts I suggested have the added benefit of having much stronger (and replaceable) threads than if you threaded the plastic.
Another way to do this would be to use flat plates for both the top and the bottom. This would reduce the pocket depth (which would become a through pocket) but would require another seal, another part (simple rectangle), and an additional assembly step. Machining a groove for the seal (a small section, large diameter O-ring) is straightforward.