_whitelogger has joined #solvespace
<whitequark> with a better spindle I think 3020 would run alu alright, but then you bump into its tiny Z travel
<whitequark> which is just 5cm
<BusError> ah yeah. wouldn't bother with PCB, I can make PCB quickly enough with chemicals + laser printer, and ultimately, oshpark and the other prototypers
whitequark changed the topic of #solvespace to: SolveSpace--parametric 2d/3d CAD · latest version 2.1 · http://solvespace.com · code at https://github.com/solvespace/solvespace · logs at https://irclog.whitequark.org/solvespace
<whitequark> still need to drill vias or PTH holes
<cr1901_modern> balrog: You wanted to do a hackerspace?
<balrog> no I work with a hackerspace already
<BusError> for DIY yeah, always a pain. that's why now I take the week leadtime and use oshpark ;-)
<cr1901_modern> whitequark: Believe it or not, my local library has a 3d printer lol
<whitequark> I cant do that manually for the life of me. and ultimately I found chemicals too unreliable and messy, not in the sense that I'm afraid of them but just in the sense that it's annoying to actually do
<whitequark> I did get about 50um resolution with photoresist
<whitequark> but then I couldn't get decent etching without access to HCl
<whitequark> I even tried doing plated vias
<BusError> whitequark, even with CNC, you get unplated vias+holes, that's about as useful as a hole in the head :-)
<whitequark> ;)
<whitequark> you can get a primer layer on vias using copper hypophosphite in ammonia solution
<whitequark> and thermal decomposition
<whitequark> so no platinum or formaldehyde
<balrog> an artist collective here that I work with has an xcarve
<BusError> I *did* work a lot on DIY boards, but ultimately, it too SO LONG for a single board that for now, I don't bother designing the board for DIY -- I know it's 'quitting' but hey :@-)
<balrog> so we use that for basic PCB manufacture
<whitequark> BusError: that's very similar to my experience
<balrog> recent versions of kicad are very usable
<balrog> on the software end
<whitequark> this is what I could do
<balrog> on the hardware end? yeah, that gets messy :p
<nats`> balrog I find them usable since 2008 :p
<whitequark> that doesnt describe any of my plated via stuf
<whitequark> you can see the etching problem though
<whitequark> so now I live in HK, which is very close to dirtypcbs fab
<whitequark> I get 2-3 days lead time plus 1-2 days shipping for less than oshpark wants
<whitequark> oshpark is *very* expensive and shipping is also slow
<BusError> whitequark, I had a found a nice way to tin DIY boards tho.. handy for SMT. but again, very DIY..
<whitequark> there's a tinning solution
<whitequark> or you could just do ghetto HASL
<BusError> whitequark, ah neat. I used dirty, and a few others before
<whitequark> or you could use that alloy uh
<whitequark> Rose's metal
<balrog> a friend uses pcbway.com and has very good results
<balrog> they also do things like aluminum pcb at not too high cost
<whitequark> put that in a pan, fill it with acerbic acid in water, heat up to 100 degrees C
<BusError> had invented a really cool method. involving solder paste and plumber flux (!). it's like painting
<whitequark> then dump your PCB in it
<whitequark> instantly tinned
<BusError> ah. copper: http://oomz.net/scaled/IMG_5466s.JPG tinned: http://oomz.net/scaled/IMG_5469s.JPG however that was the pinacle, and likely last-ish of that sort of complexity for DIY just WAY too much work
<whitequark> that's amazingly high quality etching
<whitequark> how did you do it? hcl+h2o2?
<whitequark> heck, even the 3.3 text is legible
<BusError> ferric chloride and a bit of practice. mostly use a sponge, you need *very* little solution, and it keeps forever too as you don't abuse it
<whitequark> ah the sponge method...
<whitequark> yeah thats a lot of manual work. i used FeCl3 and a stirrer, and got godawful results
<whitequark> and with persulphate it's even worse
<whitequark> even though it is cleaner chemically
<whitequark> well
<whitequark> leaves less stains ;p
<whitequark> nice
<whitequark> what's the sot323?
<BusError> it was a 10 bit microchip temp sensor. Still use some of these around actually
<BusError> otherwise it's 0603's stilll had a nice track UNDER one of the diode :-)
<whitequark> amazing
<BusError> but yeah, dirtypcb++ or oshpark++ for that matters
<BusError> in a funny twist, I still use my etching methods for other purposes...
<whitequark> which?
<BusError> brass. same laser masking, electrolysis etching.
<BusError> Did quite a few other projects like that afterward!
<BusError> she was a LOT more impressed than with the regular PCBs :-)
<whitequark> BusError: funny how you mention that...
<whitequark> I'm really proud of that alignment of copper and negative solder mask
<whitequark> even though etching is shit, as usual
<BusError> hehe. well, get some thinish brass plate (not too thick, otherwise it heatsinks and it's **!"&£&£ hard to get the mask on)
<BusError> for etching, electrolysys++
<BusError> water, some sort of salt, a nail, a power supply; aligator clips
<nats`> I do that to mark my tool I bring to work
<nats`> like tweezer etc... :)
<whitequark> nats`: neat
<BusError> nats`, ah with salty water?
<nats`> 30 second for permanent marking :)
<nats`> I do a negativ with capton
<nats`> kapton
<nats`> I use salty water and my lab psu
<whitequark> laser cut kapton?
<BusError> yeah works nice on pretty much anything. But if you want 'deep' ones like that page marker, you need a bit more work
<nats`> whitequark handcut :)
<whitequark> speaking of electrolysis I need to get electropolishing working
<whitequark> in principle you just need a bucket of phosphoric acid
<BusError> I do rust removal a lot actually... I somehow turn out to do that a LOT more than pcbs and stuff these days!
<BusError> dug a hole in the garden, put a tarpolin in, and cleaned a whole wrought iron gate: https://goo.gl/photos/q1NXNVaUfLmVHuCD9
<nats`> I hate bottom view package model in datasheet :\
_whitelogger_ has joined #solvespace
_whitelogger has quit [*.net *.split]
nats` has quit [*.net *.split]
nats` has joined #solvespace
whitequark has quit [*.net *.split]
whitequark has joined #solvespace
<GitHub> [solvespace] whitequark pushed 2 new commits to master: https://github.com/solvespace/solvespace/compare/5462fc6b3c86...2ccf5954d4a8
<GitHub> solvespace/master 2ccf595 EvilSpirit: Allow creating n-dimensional arrays with translate and rotate groups....
<GitHub> solvespace/master 6658b1f EvilSpirit: Allow combining extrude, lathe, translate and rotate group as assemblies....
<whitequark> BusError: huh
<BusError> whitequark, ?
<whitequark> rust removal
<whitequark> not my first guess...
wjIrc has joined #solvespace
<wjIrc> Hello!
wjIrc has left #solvespace [#solvespace]
<GitHub> [solvespace] whitequark pushed 2 new commits to master: https://github.com/solvespace/solvespace/compare/2ccf5954d4a8...ea52fcbce153
<GitHub> solvespace/master ea52fcb whitequark: Remove unnecessary GenerateAll/ZoomToFit calls in AfterNewFile....
<GitHub> solvespace/master 62f5f69 whitequark: Print a debug message for generations that are taking a long time.