4 ms·
Below I suggested it would be easy for EDM or ECM. Have I been misled?
by kragen 5y ago
Below I suggested it would be easy for EDM or ECM. Have I been misled?
- jacquesm 5y agoEDM will take a very long time, you are going to take off a few microns per hit, and ECM will likely take either an insane amount of power or will take forever across such a large surface. Compared to say machining aluminum or steel you will be practicing some patience. Not necessarily a bad thing to have in the world of metal working. ("increased feed speed is not a substitute for bad planning")
- ISL 5y agoEDM can cheese-wire right through blocks of tungsten.
- jacquesm 5y agoThat I readily believe, but to face a sheet is a different matter, which I believe was the application the OP intended. You can also cut tungsten with water jets loaded with abrasives, I've seen a 2" thick block cut like that, it was quite impressive. But that won't get you a surface. You'd need to start with a sheet of the right thickness.
- ISL 5y agoTrue. I bet a quick call to Rembar would put you on your way to a source if you just need sheet. Tungsten foil/sheet isn't that hard to come by.
- jacquesm 5y agoI've only ever used bar stock, but I imagine the same suppliers would have sheet as well though I can see a couple of arched eyebrows when you order that, plus a question or two if you are absolutely sure they have your order down accurately. Cost a small fortune too.
- kragen 5y agoThank you very much!
- tohnjitor 5y agoRembar and Plansee are consistently good.
- chromaton 5y agoCut it off the end of a bar.
- jacquesm 5y agoHehe. Very funny :)
- kragen 5y agoI guess I was thinking that ECM would take less power per mass on tungsten than on things like steel or copper, just because each atom is so much bigger. I'm not very clear on the chemistry, but suppose that we're electrolytically oxidizing each tungsten atom into an orthotungstate ion by withdrawing two electrons from it, with an electrolyte of something like chloric acid or sodium hypochlorite so we don't have to pay more electrons to free up some oxygens. Then at 100% Faraday efficiency we'd cut away 184 daltons of tungsten for every two electrons; contrast with iron where you still need to withdraw two electrons to get a stable water-soluble ferrous ion, but that only buys you 56 daltons of metal removal. Even copper, where the cuprous ion only costs one electron, you only dissolve 64 daltons. There's also the issue of electrode potentials: sucking a coulomb off your anode at 0.5 volts takes only half a joule, but sucking off the same coulomb at 2 volts takes 2 joules. I don't understand standard electrode potentials well enough to predict what voltage you need for efficient tungsten ECM, but on first glance at the reference table, it doesn't look particularly bleak. The tungsten entries are in the middle of the table, near things like lead, tin, and iron, rather than way down on the upper east side with hoity-toity mercury and iridium, much less out in the ethereal districts where gold and ozone dwell.
- jacquesm 5y agoPer unit of mass yes, but there is rather a lot of mass for a given volume. And those are relatively slow processes when it comes to dealing with surfaces. Perpendicular cuts should be a lot quicker (you only have to separate the two sides, in theory one slice of atoms would be enough but in practice you'll end up with some minimal cutting radius). Fascinating processes by the way, and the precision that you can reach with EDM is insane.
- kragen 5y agoHmm, I thought ECM and tap-disintegrating EDM (which is not very precise and does use a huge amount of power, unlike, say, wire EDM) had pretty astonishingly huge MRRs? I think the density ratio for tungsten and steel (19:8, just over 2:1) is actually more even than the atomic mass ratio (23:7, just over 3:1). So I'd think that even on a volume-per-coulomb basis ECM would cut tungsten faster, assuming comparable faradaic efficiencies. Do you have any idea about the electrode-potential question? I'm totally lost.