On Monday 18 January 2016 11:35:09 Ralph Stirling wrote: > I have an LPKF pcb milling machine with a 62KRPM spindle, but > I rarely use it to make pcb's. I would far rather wait ten days for > double-sided pth boards with soldermask and silkscreen from > OSHPark.com, which are $5/sq.in. for three boards, free shipping. > If I need something bigger than a few square inches, that starts > getting a bit expensive, but most of the quicky boards I would engrave > are small anyway. The OSHPark boards (or any other commercially > fabbed pcb) are so much more reliable and easy to solder than home > engraved ones, that the extra wait is well worth it. I do find the > LPKF machine useful for making solder stencils and depanelizing > batches of boards we sometimes get from Advanced Circuits. > > -- Ralph
When I did finally find that belt driven spindle, it was on a small Sabre machine, with about a 150mm wide moving table for X, which could move (if I read the chinglish right) 200mm and 70mm vertically. Std screws, rated accuracy with anti-backlash nuts was .05mm, and the whole thing, ready to be driven by Mach, was under $800 USD from Taiwan, plus ship I suppose. Looked like a very neat little table-topper one could shelve in between PCB jobs. If I really had a bunch to make, it looked like it could do them at a decent pace, doing 10 grand with that spindles belt in high gear. Even that 10G's it ought to be able to march along at 7 or 8 ipm. The video did show a much different etching bit, I'd estimate as a 85 degree bit, something I have not seen on fleabay yet. It was also throwing white glass for 2-3 inches, so IMNSHO, it was digging way too deep, several thou into the glass. The idea is to remove the copper, not dig a ditch 10 mills deep in the glass. Thats hell on carbide even. If I make a pallet to hold the board flat, I figure I am about right when the copper is cleared, but the glass isn't touched. But that takes eagles pcb-gcode to do that as it can make multiple passes, spaced at 1 or 2 mills apart in order to get a gap that doesn't solder bridge by mistake. If I was doing 50 boards a year, like making quadrature encoders for a G0704 mills spindle for O.P., or if I find there is a market for a ready made charge-pump detector like I'm making, I'd do it in a Texas minute. But making it is less of a problem than fitting it because there isn't a good, precisely located, zero runout, place to put it. Mine works, and works well, but the disk mounting is a bit fragile. As for sending it out to OHPark, I have no clue if pcb's gerber output is as broken as its gcode output is. I think we have gerber viewers, so maybe I ought to have it make the gerber's once just to see if the gerber is all there, because the gcode is NOT all there, thru holes missing and too often the wrong size. Or maybe there's some cruft in my board.pcb causing it. IDK. So much for that idea, I made the .gbr's, but gerbv can't show them, unless I open as a layer. Then the bottom trace is good, but the hole pattern isn't sized correctly. 4 of them are too small. But in .gbr, they are all there, so its in the g-code exporter that they go missing. I see it can mirror, so there is a slight possibility I can use that for copper clearance on top of the board. Food for thought... :) But something else to learn I guess. If I've a mind to. But in the past, I mirrored what I needed to in geany, again because it was a 2 sided board and I was only drilling 35 thou deep, so when I turned the board over to drill the other side, my holes all met in the middle of the boards thickness, with no detectable offsets at the meeting point. But, to do that on THIS mill, I'd have to get camview working. On the small mill, I was using a brass tube set in the corner of the pallet to locate 0,0 XY, then using the co-ordinate system to subtract that tube's location from the patterns 0,0 position. Camview would make that a bunch easier. and handle any rotation that may creep into turning over the board. Must be time to see if I can fix that camera, and make a new mount for it as I knocked it off the small mill with a fixture bolt that was too high. More food for thought. I have a bigger, better camera, but I'd need to disrobe it and see how small I could make it, and design a new mount that actually fits it inside the head casting which will protect it from such operator=me idiocy. That casting has quite a bit of hollow space, with the major problem being one of seeing around the spindle brake I built rather than give Grizzly 90 bucks for theirs, which is just as big & fugly as mine. Good thing I brought in the brass monkey last night, else he'd be emasculated this morning if not. Winter has (finally) arrived in WV, and my diabetic feet are weakly yelping about it. Thanks Ralph. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
