#qi-hardware IRC log for Tuesday, 2014-01-28

DocScrutinizer05http://en.wikipedia.org/wiki/X10_(industry_standard)#Radio_protocol00:00
wpwrakthat's pretty much what i have, too. and 250 g will last you one or two lifetimes ;-)00:00
apeletewpwrak: really ? was expecting you'd say no :)00:00
apeletegreat00:00
DocScrutinizer05I'd use stannol00:01
apeleteDocScrutinizer05: only found 60/40 at stannol :-(00:01
DocScrutinizer05no compromises with solder wire00:01
wpwrakdiameter is 0.7 mm, which is the biggest you'll want. it seems to include flux. it seems to be designed for electronics. (not all kind of soldering is for that, and you don't want some of the more aggressive kinds of flux.) and it's 63/37.00:02
apeletewpwrak: let's see if I can find flux now. powder or paste ?00:02
apeletewpwrak: for flux then, better choose powder or paste ?00:03
wpwrak"marker pen" is convenient. else "ballpoint pen". else syringe.00:03
DocScrutinizer05http://www.stannol.de/en/products/solder-wires/00:03
DocScrutinizer05http://wstaw.org/m/2014/01/28/plasma-desktopuM3389.png00:04
wpwrakDocScrutinizer05 only trusts "made in germany" ;-)00:04
wpwrakit's weird that they wouldn't have 63/3700:05
DocScrutinizer05hmm00:05
wpwrakmaybe Sn62Pb35Ag2 is supposed to be a substitute00:06
DocScrutinizer05yes00:07
DocScrutinizer05pretty nice stuff00:07
DocScrutinizer05http://www.stannol.de/en/products/solder-wires/?tx_produktselektor_filter%5Baa_prod%5D=HF32-SMD&tx_produktselektor_filter%5Baction%5D=show&tx_produktselektor_filter%5Bcontroller%5D=Product&cHash=24a7ee35aa8c79ab5af204dbd257cf8d00:07
DocScrutinizer05just searched for it00:07
wpwrakyeah, wikipedia says it's equivalent, and better for anything containing silver00:08
DocScrutinizer05and halogen free no-clean flux00:08
DocScrutinizer05and works great00:09
wpwrakand the wire is nicely thin00:09
wpwrakapelete: do this one would be an option, too. even a little better than the basic 63/3700:10
DocScrutinizer05downside: only 250g available00:12
DocScrutinizer05sidenote: seems all stannol pb solders have red plastic spindle, all RoHS solder wires have green spindle00:14
DocScrutinizer05you usually don't want to go for RoHS00:14
DocScrutinizer05unless industrial00:14
DocScrutinizer05though, rework on RoHS pb-free PCBs may get a tad nasty with SnPb00:15
DocScrutinizer05not relevant for apelete though00:16
apeletewpwrak: which one are you speaking of ?00:16
wpwrakapelete: the one joerg pointed out, http://www.stannol.de/en/products/solder-wires/?tx_produktselektor_filter%5Baa_prod%5D=HF32-SMD&tx_produktselektor_filter%5Baction%5D=show&tx_produktselektor_filter%5Bcontroller%5D=Product&cHash=24a7ee35aa8c79ab5af204dbd257cf8d00:16
DocScrutinizer05the Sn62Pb35ag2 has a melting temperature of only 179°C00:16
wpwrakSn62Pb36Ag200:17
apeleteokay, thanks00:17
wpwrakDocScrutinizer05: 250 g is plenty :)00:17
DocScrutinizer05yeah, WAAAY too much00:17
wpwrakindeed ;-)00:17
DocScrutinizer05and, prolly thanks to Ag, not exactly cheap00:17
wpwrak21.30 EUR. not too bad. http://www.pkelektronik.com/stannol-loetdraht-hf32-smd-sn62pb36ag2-0-5-mm-1-0-250-g.html00:18
DocScrutinizer05:nod:00:19
Action: DocScrutinizer05 idly wonders how many solder (in gramm) I used for the 1600 thruhole PCB of my Nascom-I computer00:20
DocScrutinizer05I'm rather sure, no 50g00:20
wpwrakmuch less i'd say. i have maybe 150-200 g on my 250 g roll of 64/37 and i've been using that for everything i've done in the last 10 years00:22
wpwrakof course, TH uses more than SMT and i had only relatively few TH parts, but still00:23
DocScrutinizer05http://www.ersa.com/media/pdf/fachartikel/fachartikel_englisch/improving_hand_soldering_operational_costs_and_process_control_2.pdf00:25
DocScrutinizer05doesn't answer my question, but a relatively nice read nevertheless00:25
DocScrutinizer05it's amazingly hard to find *any* info on that in aunt google00:40
DocScrutinizer05only about copper tubbing00:40
DocScrutinizer05anyway RoHS is a joke, for DIY00:41
DocScrutinizer05a hunter spreads more Pb to the environment in one night than an electronics nerd in his whole life00:42
DocScrutinizer05but heh, EU is crazy, we know for sure since Energy Saving CF Lamps00:43
DocScrutinizer05eventually they will come up with a procedure how to bend banana to the correct diameter00:43
DocScrutinizer05I doubt environmental advantage of even LED lamps, over incendescant00:45
DocScrutinizer05it's already known they spread toxic gas in your home00:46
DocScrutinizer05not Hg, but all other sorts of shit used in plastic and dunno what00:46
DocScrutinizer05and they have EMI field thet often even exceeds what EU allows for workplace00:47
DocScrutinizer05X-P00:47
DocScrutinizer05while total electric energy savings are zilch. when you heat with electricity anyway00:48
DocScrutinizer05what you save in summer when your own heating doesn't have to burn the amount yiu save on the lamp, you find wasted in fab during production of those lamps00:49
DocScrutinizer05reminds me to finally calculate and get the needed stash of incandescant lamps for next 40 years00:52
wpwrak(banana bending) this will be the ultimate success :)00:53
DocScrutinizer05(solder) I *seem* to recall I once used Weller solder wire that also been quite good00:54
DocScrutinizer05or at least the plastic looked as blue as Weller does00:55
wpwrak(CF rant) yeah, we have the same madness here. well, so far none of them has failed. but then, since i know that they don't like cycling, i keep the important ones on all night. it may still save energy ;-)00:55
DocScrutinizer05hardly00:55
wpwraki don't think you need to get too picky with the solder wire. if the alloy is the right formula and they put a halfway decent flux, there's not much else that can go wrong.00:56
DocScrutinizer05recent test shown the 8W use 15W actually, and are not as bright as they are supposed to. So it's more than questionable how much saving they actually earn you00:56
wpwrakheh :) more lies then00:56
DocScrutinizer05the flux is muy importante00:57
DocScrutinizer05and I found the 2% Ag do wonders00:57
wpwraki think most of the flux in the solder evaporates when doing SMT00:57
DocScrutinizer05sure, for SMT the flux in solder wire is basically irrelevant00:57
DocScrutinizer05you're supposed to use decent no-clean flux pen anyway00:58
wpwraki solder like this: put flux on the target area, place component, put a drop of solder on the iron, hold down the component, then solder. so the solder stays on the iron for a bit of time, giving the flux time to disappear00:58
wpwrakyup00:58
DocScrutinizer05btw only "poor" flux evaporates that fast00:59
DocScrutinizer05leaving nasty brown blobs00:59
DocScrutinizer05you can tell about quality of your solder from looking at your iron tip01:00
wpwrakyeah, mine's so-so. not super clean but not too evil either.01:02
DocScrutinizer05I hardly ever need to clean the tip (you know, the wet sponge touchdown procedure)01:02
DocScrutinizer05usual resin flux solder wire is a mess01:03
apeleteDocScrutinizer05 wpwrak: does this look like a good flux pen to you ?01:03
apeletehttp://www.pkelektronik.com/circuitworks-kolophonium-flussmittelstift-cw-8200-9-g.html01:03
DocScrutinizer05and some suspicious noname stuff is mere sabotage01:04
wpwraki use a copper "sponge". no need to water the beast :)01:04
DocScrutinizer05kolophonium, nope01:04
DocScrutinizer05Kolo is exactly the resin I just ranted about01:04
wpwrak"Er trägt ein patentiertes korrosionsfreies, halogenfreies organisches No-Clean Flußmittel der Sorte R auf. Das Flussmittel enspricht der Norm MIL-F-14256 E+F Sorte R und RMA." eh ?01:05
apeletehaha :)01:05
wpwrak"no clean" and RMA don't quite resonate :)01:05
DocScrutinizer05get F-SW33 or F-Sw34, better the latter01:09
DocScrutinizer05http://www.conrad.de/ce/de/product/828718/Edsyn-No-clean-Flussmittelstift-FL-88-Flussmittel-Typ-F-SW-34-Inhalt-10-ml?queryFromSuggest=true01:09
DocScrutinizer05stay away from all F-SW1x and F-SW2x. Avoid F-SW31/32 Rosin flux if possible01:10
apeletewas looking at some stannol flux pen, but I'll go with F-Sw34 since you advise so01:11
apeletethanks :)01:11
DocScrutinizer05stannol also has same type of flux pens01:12
DocScrutinizer05http://www.conrad.de/ce/de/product/812062/Stannol-No-clean-Flussmittelstift-830322-Flussmittel-Typ-F-SW-33-Inhalt-10-ml/?ref=detview1&rt=detview1&rb=201:13
DocScrutinizer05you need to test if SW33 or SW34 works better for you01:14
DocScrutinizer05basically all that still doesn't really tell all the truth about Flux. You should get a non-activated no-clean no-rosin flux. but it still needs explicit classification "for SMT2 by manufaturer01:16
DocScrutinizer05s/2/"/01:17
DocScrutinizer05you even might find mildly activated flux that still is suited for SMT01:18
DocScrutinizer05but that's usually no no-clean then, since any activation is generally corrosive01:19
DocScrutinizer05buying/finding the right flux is an art. See http://www.conrad.de/ce/de/product/588431/Stannol-Flussmittelstift-Fine-Fluxer-3000-830335-Inhalt-8-ml/?ref=detview1&rt=detview1&rb=1 vs http://www.conrad.de/ce/de/product/588408/Stannol-Flussmittelstift-Fine-Fluxer-350-830333-Inhalt-8-ml/?ref=detview1&rt=detview1&rb=101:32
DocScrutinizer05nfc what's the difference, except for the price01:33
apeletehaha :)01:38
DocScrutinizer05and while finding a particular solder is like finding a certain kind of flour, you just need to specify the type and grinding and no much other differences between flours, finding a good flux is like finding a tasty cheese since you simply can't go by the composition or other descriptions of physical properties01:38
apeleteI see. for now I went with the Stannol Sn62Pb36Ag2 HF32 SMD solder and Stannol F-SW33 flux pen01:40
DocScrutinizer05will work01:40
DocScrutinizer05:-)01:40
apeletewill see how the flux pen turns out, and try F-SW34 another time01:41
apeletethanks for everything :-)01:41
DocScrutinizer05yw :-)01:41
DocScrutinizer05actually it feels good to help somebody01:43
DocScrutinizer05wpwrak: when you want to make sure "the internet" works while I'm away, would you please keep an eye on ht-tp://monitor.maemo.org/ganglia and may I forward mails like "** PROBLEM Service Alert: firewall-b.maemo.org/SSH is CRITICAL **" to you?01:54
DocScrutinizer05nevermind - "** RECOVERY Service Alert: firewall-b.maemo.org/SSH is OK **"01:55
wpwrak;-)02:45
wpwrakapelete: you could try to get a range of flux pens, e.g., one "no clean", one RMA, one "water-soluble" or (if you can't find one) maybe another "no clean" with a different formula02:48
wpwrakchances are that you'll not use the RMA much. but it'll give you a feeling for how a "strong" flux works.02:48
wpwrakoh, and if you're not over budget yet, consider getting a cheap ultrasonic cleaner. try to minimize power / volume, i.e., lowest power you can get.02:50
wpwrakultrasonic cleaning is great for getting rid of flux. you can just use demineralized water for that, the sort you can find in car supplies. low power reduces the risk of damaging components.02:51
DocScrutinizer05only for water soluble flux03:49
wpwraki think it works for all kinds of flux03:50
wpwrakthe "water soluble" isn't all the much about regular water anyway03:51
wpwrak#s/the/that/03:51
wpwrakif you want, you can of course also use other solvents. e.g., alcohol. would be a shame to waste it for that, though ;)03:52
DocScrutinizer05the orthodox way is to use Kontact-WL or similar stuff03:53
DocScrutinizer05Kontakt actually03:54
DocScrutinizer05alcohol may work too03:54
wpwrakoh sure, there are more efficient solvents. but you don't really need them with an ultrasonic cleaner. it's the ultrasound that does most of the work, not the solvent.04:03
wpwrakand things like demineralized water have the additional benefit that you can get them anywhere ;-)04:04
wpwrakwhen you get a chance to set up a proper lab, you may also want to get an ultrasonic cleaner. it's also very useful when you get something and find that its pcb is full of ugly flux residues :)04:05
whitequarkwpwrak: ping07:27
whitequarkroh: which cam do you use?09:32
whitequarkI've tried pycam and not only it wants to work for hours (who the fuck writes a cam in python), but it also generates horrendous toolpaths09:32
whitequarkopencam is a joke, heekscam is what I'm currently building, and... that's all we've got?09:33
whitequarkfoss for 3d09:33
viricheekscam may be a scam09:37
viriche he he09:37
whitequarkokay, heekscnc generates quite decent toolpaths10:06
whitequarkbut it is beyond painful to use10:06
whitequarkfortunately the design of heekscnc (heekscad generates a python script with geometry and tools, python script calls a C lib which generates toolpaths, then outputs customized gcode) makes it vary easy to decouple it from heekscnc10:08
wpwrakwhitequark: pong10:27
whitequarkwpwrak: so I was looking for endmills and somesuch10:28
whitequarkfirst I think I'll get this collet set: http://www.amazon.com/gp/product/B00G9X6TWA/ref=ox_sc_act_title_1?ie=UTF8&psc=1&smid=A388EF2JRGUUQ310:29
whitequarkmy current one only accepts 3mm aka 1/8" shanks10:29
wpwrak3 mm isn't 1/8 in. you'd be off-center almost 0.1 mm. that can't be good10:30
whitequark1/8" is 3.175mm precisely10:31
wpwrakyes. so 0.175 / 2 mm off-center if yours is 3.000 mm10:31
whitequarkand if I'm reading it right, the bits I currently have are ~3.2mm in diameter10:32
whitequark(probably just 3.175)10:32
wpwraki can't make sense of the "spring collet set". how would you mount it ? how do they grip the bit ?10:32
whitequarkoh, such a collet is inserted inside a collet nut (er11)10:32
whitequarkthen it's screwed on the motor shaft10:32
wpwrakah, so your 3 mm are really 1/8 in. okay ;-)10:32
whitequarkand the collet grips the bit as the result10:33
whitequarkokay. so. I think I'll get these endmills for pcb milling: http://www.amazon.com/Carbide-Endmill-0-2mm-Ideal-Engineering-Electronics/dp/B008RE2VJ4/ref=pd_cp_hi_110:33
whitequark0.2mm is about 8 mil, should be good enough10:33
whitequarkand I doubt thinner ones will survive10:34
wpwrak(collet nut) okay. and how do you mount that one on the mill ?10:34
whitequarkwell, the collet nut with a collet inside is screwed on a threaded cylinder, thus gripping the collet and bit10:35
wpwrak10 pieces for only USD 14.99 ? wow10:35
whitequarkand that cylinder has a hole on the other side via which it is affixed to the shaft10:35
whitequarkdoes that make sense?10:36
wpwrak(threaded cylinder) okay, so you need one that matches. doesn't seem to be part of the set, isn't it ?10:36
whitequarkit's already on my CNC10:36
whitequarkthat set of collets looks exactly like the collet I already have, except the CNC only came with one10:36
whitequarkand the collet nut too10:36
whitequarkhttp://www.jinlantrade.com/ebay/3020t/img13.jpg10:36
wpwrakah, i see. pity they don't specify the inner diameter then10:37
wpwrakbut maybe it's a standard10:37
wpwrakokay, sounds good10:37
whitequarkwell, the fact that the collet grips the bit allows for some leeway10:37
whitequarkI'm pretty sure it would grip 3.00mm bits just file10:37
whitequark*fine10:37
whitequarkthe collet is rather flexible10:38
wpwrakno, i mean the other side, where the threaded cylinder enters. not sure if there's one nice standard for that10:38
whitequarkah10:38
whitequarkhttp://ugracnc.com/ER-COLLETS/ER11-Clamping-Nut.html10:38
whitequarkit's apparently a kind of M14 nut10:38
wpwrak(0.2 mm endmill) they'll break very easily. but if you want to experiment with direct pcb milling, then that looks like a great set10:39
wpwrakfor actually cutting a pcb and other excavation work, you'd want larger endmills. about 0.8-1 10:39
whitequarkyes, that's for milling of copper10:40
wpwrak.0 mm is a good size range. they can go reasonably deep (some 3 mm), don't break easily, and still give you a lot of detail10:40
whitequarkwas thinking about these ones: http://www.amazon.com/Carbide-PCB-Mill-Endmill-1-0mm/dp/B008JPCG28/ref=pd_sim_hg_110:40
wpwrakyou'll also want larger sizes, at least one full-size 1/8 in. it's unbreakable and very suitable for bigger objects10:41
whitequarkoh, there's one with the CNC: http://www.amazon.com/Steelex-D2699-8-Inch-2-Flute-Carbide/dp/B001UE7IVO/ref=sr_1_8?s=hi&ie=UTF8&qid=1390905713&sr=1-8&keywords=endmill10:42
wpwrakinteresting, the 1.0 mm are USD 13.88 each. i bet the 0.2 mm have the unit price, too. i've never seen endmills at USD 1.410:43
whitequarkappears to be @10pcs10:44
wpwrakokay, that also solves the mystery what you used for your engraving experiment :)10:44
whitequarkyou think I should buy 20? :)10:44
whitequark(engraving experiment) no, the CNC also had 5 engraving bits with it10:44
wpwrakat that price ? certainly. but i doubt it's for 10 units :)10:44
wpwrakoh wow. the bits must be worth more than what you paid. no surprise they had nothing left for a decent transformer ;-)10:45
whitequarkok. so, then drill bits: http://www.amazon.com/gp/product/B00569X3JU/ref=ox_sc_act_title_2?ie=UTF8&psc=1&smid=A3CXZ32ML2H79F10:46
whitequarkagain at a ridiculously low price10:47
wpwrakyou may also want to get an endmill in the 0.8 mm range. there are a lot of small mounting holes or slots of about that size. 1.0 mm would already be quite large for such things10:47
whitequarkhm10:47
wpwrakHSS, that explains the price10:47
whitequarkHSS is bad?10:48
whitequarkhuh, this carbide one: http://www.amazon.com/Carbide-End-Mill-0-8-Cut/dp/B007SUEI1C is at $4110:48
wpwrakthe endmill is expensive because it has a weird head ;-)10:49
whitequarkoh10:49
wpwrakHSS is okay for drills. they don't suffer nearly as much wear and endmills.10:49
wpwrak#s/and/as/10:49
wpwrakdoes all that stuff ship from amazon ? or from 3rd parties ?10:51
wpwraki didn't realize amazon had this sort of things. it's an amazing collection10:51
whitequark3rd parties, but some from amazon warehouses10:51
whitequarkamazon has a lot of surprising items at surprising price10:51
whitequarkit's espeically good if you use a remailer in USA. free shipping in 1-2 days10:52
whitequarkok, amazon doesn't have any endmills under 1.0mm cheap10:53
whitequarkstarts at $20/1pcs10:53
wpwrakamazon warehouses sounds great. that way it's just one shipment.10:54
whitequarkyep!10:54
whitequarkok, there's a set: http://www.amazon.com/0-6-1-5mm-Tungsten-Carbide-Milling-Machine/dp/B00GEC9X10/ref=sr_1_9?s=hi&ie=UTF8&qid=1390905827&sr=1-9&keywords=1mm+endmill10:54
wpwrakaaah, THAT explains it ! "Please note,These bits are Very Sharp and lightly used and Each Piece has Been Cleaned ."10:56
whitequarkhm?10:57
wpwrakthey're refurbished, not new10:57
whitequarkoh10:57
whitequarkhow bad is that?10:58
wpwrakdunno. try it. at that price, you can't go wrong ;-)10:58
Action: whitequark looks at the cart containing $150 of bits10:59
whitequarkwell, I guess it's ok, considering the amount I get10:59
whitequark80 bits by the cost of 12 or something10:59
wpwrakthe more worn a bit is, the more burr you get and the more unwanted forces it applies to the piece. before i broke my 0.8 mm endmill, it had produced considerable burr (it's been in use for a good while, mainly cutting pcbs, which are nasty)11:00
wpwrakthen the new one cut a pcb without any burr. perfectly clean. that's when you notice just how long overdue the previous one way ;-)11:00
whitequarkah11:01
whitequarkbtw, the 1.0mm endmill description mentioned it can be used for _drilling_ pcbs11:01
whitequarkdoes that make any sense?11:01
wpwrakthe endmill set may be a bit fragile. normally, the length of the cutting part is a function of the diameter. if it gets very long and thin, it'll break more easily. but okay, you can always compensate by going very slowly11:02
wpwrakendmills work great as drills, too11:02
whitequark0.2mm holes... vias!11:03
wpwrakthat's very thin even for that :)11:03
whitequarkwell, a bit11:03
whitequarkwhile we're at it, how do you fix the pcb onto the table?11:03
wpwraki use 13.5 mil holes, 0.35 mm.11:03
wpwraki don't11:04
whitequarkoh?11:04
wpwrakah, you mean when milling, sorry :)11:04
wpwrakmy mind had wandered to soldering ;-)11:04
wpwrakwait, i'll get me some coke. this is no condition to be in ...11:05
wpwrakokay, that's where our ways part. i use double-sided tape. you can do that too, but you have better ways since you can use screw-down fixtures11:08
wpwraki don't know too much about them but roh probably does11:08
whitequarkdouble-sided tape I have, yes11:08
whitequarkthe cnc machine has some parts with it which *seem* to be kind of a screw-down fixture11:08
wpwrakexcellent11:09
whitequarkbut the nuts are slightly, about 0.2mm bigger, than the slits on the table11:09
wpwrakthe slits should be tiangular. maybe they only fit if inserted from the side ?11:09
whitequarktriangular?11:09
whitequarkwell, I thought you slide in the nut from the side to the position you need, and then screw the plate with a bolt or something11:10
wpwrakor maybe even rectangular. wider inside than at the top11:10
wpwrakyes11:10
wpwrakah, and you want some reasonably level wood. or maybe some FR2 board you don't like. they make great victim material.11:11
whitequarkhttp://www.jinlantrade.com/ebay/3020t/img14.jpg11:11
wpwrakah yes, perfect11:11
whitequarkand http://www.jinlantrade.com/ebay/3020t/img21.jpg11:11
whitequarkthe nuts are slightly bigger horizontally than the slits11:12
whitequarkthe bolt heads are slightly bigger vertically than the slits11:12
whitequarkso, yeah, there's nothing I can insert there :D11:12
whitequarkand there is no manual, of course11:12
wpwrakah well, find smaller nuts then :)11:12
whitequarksmaller bolts maybe? the bolt would be too long to completely go into the nut11:13
whitequarkcompare bolt length and slit depth11:13
wpwrakthey have to go through the rectangular piece as well11:16
whitequarkyeah, but they're still 2x longer than needed11:16
wpwrakthe stack should be: nut, table, workpiece, rectangular piece, head of screw11:17
wpwrakoh, or maybe it the other way around11:17
wpwrakdoes the head of the screw fit into the grooves ?11:17
whitequarknope!11:17
wpwrakdamn :)11:17
whitequarkabout 0.2mm higher than the groove11:18
whitequarkit really looks like they've simply screwed up with tolerances11:18
wpwrakwhat do people in the fora say ? same problem ?11:18
wpwrakor maybe just bought the wrong parts ;-)11:18
whitequarkcan't google anything useful11:19
wpwrakmaybe ask in the forum where you found that other guy's report about the flaming transformer ?11:20
whitequarkmmm, easier to find the right screws :)11:21
whitequarkit's not like there is a shortage11:21
whitequarkbolts*11:21
apeletewpwrak: to clean flux on a pcb you actually put the pcb in the ultrasonic cleaner *with* demineralized water in it ?11:28
wpwrakyup11:32
wpwrakthe water carries the sound waves. and works a little bit as solvent, too11:33
nicksydneywpwrak: http://dmitry.gr/index.php?r=05.Projects&proj=11.%20Bluetooth%20LE%20fakery11:36
wpwraknicksydney: very very old news ;-)11:43
nicksydneywpwrak: ahhh...ok :)11:43
wpwraknicksydney: i even mentioned it in one of my posts. twice, i think ;-)11:43
whitequarkha, the second version of 3020 uses a SMPS: http://img02.taobaocdn.com/imgextra/i2/900208505/T2Shq3Xj8bXXXXXXXX_!!900208505.jpg12:36
whitequarkbut it would also set me back $1300 so... not economical either way12:37
CYB3RHi all! Anyone here uses UBIBoot or nandboot?13:06
whitequarkwpwrak: I think I found the catch with those cheap bits... Shipping: Standard (17-28 business days)13:25
whitequarkwpwrak: where do you buy your endmills?13:26
wpwraki got mine from www.pmtnow.com13:28
wpwrakof course, those are in the USD 15 / each price range. all new, though. and they come in nice boxes :)13:28
whitequarksee, I can't find 0.8mm, much less 0.2mm endmills here at all13:29
wpwrakhave you fixed the power supply yet ? if not, consider that you'll spend some time waiting for that as well13:29
wpwrak"here" being locally ?13:30
whitequarkyes13:30
whitequarkoh, actually found them, yeah, about $15 each13:30
wpwrakyeah, same here. there's virtually nothing. there's a bunch of jeweler's tools shops so i tried these. all i had at the end of the day was another 1/8 in endmill. a weird piece, something like 4-flute. but works :) i keep it as a backup13:31
wpwrakwell yes, USD 15 is the usual price for such things ...13:32
whitequarkbigger ones up to $60 !()13:32
wpwrakthe really cheap ones probably work like this: a factory uses new endmills. they have high standards, so as soon as the mills begin to perform less than perfectly, they replace them. but the mills are still good for a lot more use, you just wouldn't get perfect cuts. for PCBs and such, you can tolerate quite a lot of imperfection. e.g., you just scrape off the burr without even thinking much of it.13:33
whitequarkI can see how that would work, yes13:34
wpwrakso they sell their "spent" endmills to a company that cleans and resells them. thus those companies depend on the inflow from factories, which may not be continuous.13:34
wpwrakthus the long lead times. maybe they also don't stock but just build up a big order, then go asking around.13:35
whitequarkanother guess: they only ship from HK via ePacket and amazon lists the longest epacket delivery time, that's 17-28 BD13:35
whitequarkin reality ePacket arrives to US for 5-7 days13:35
whitequarksometimes less13:36
whitequarkok, we'll see soon enough13:36
wpwrakyeah, just get the stuff. it's so cheap, even if it's an almost total writeoff, you still win ;-)13:36
wpwrakwhitequark: seen this kinda sci-fi story yet ? http://www.tor.com/stories/2012/07/a-tall-tail14:27
whitequarkhahahaha, hot HF exhaust plume14:38
wpwrakit gets better :)14:39
whitequarkhehe, ClF314:59
whitequarkdidn't we discuss it here already? together with F2O214:59
wpwrakpossibly :)14:59
whitequarkhttp://pipeline.corante.com/archives/2010/02/23/things_i_wont_work_with_dioxygen_difluoride.php15:00
wpwrakyes yes FOOF spells fun :)15:01
DocScrutinizer05((<wpwrak> the stack should be: nut, table, workpiece, rectangular piece, head of screw)) maybe ... no... nevermind, it's strange. anyway the square nuts go to the grooves in table15:05
DocScrutinizer05get a decent file, file the nuts to fit15:06
wpwraksays the pragmatist ;-)15:06
whitequarkwhat about bolts? still too long15:06
wpwraki guess it's file + long bolt = short bolt15:07
DocScrutinizer05I'd consider using washers or similar stuff15:08
DocScrutinizer05eventually you'll want long bolts15:08
wpwrakeven better. non-destructive :)15:09
Action: wpwrak nods15:09
DocScrutinizer05make sure mill head doesn't run into bolts!15:09
DocScrutinizer05alternatively use a piece of wood cut to shape and thick enough to lift PCB to the level you need for the bolts to fit15:10
DocScrutinizer05also fine for drilling15:10
wpwrakbolt and head will generally survive. the bit was cheap :)15:10
DocScrutinizer05with bolts that long and high above table level, you can possibly run the motor into them15:11
whitequarkyep, they surely didn't think the bolts thorough15:12
DocScrutinizer05alternatively, get a collection of bolts of different length, 5mm stepping15:12
DocScrutinizer05I wondered if each fixture consists of one flat rectangular bit and TWO bolts, but it doesn't make much sense in the end15:13
whitequarkDocScrutinizer05: apparently it does somehow15:13
whitequarkthere's nothing else to put bolts into15:14
wpwrakDocScrutinizer05: it does if you cross a corner15:14
DocScrutinizer05OOOH!15:14
DocScrutinizer05:-)15:14
DocScrutinizer05now that makes some sense, indeed15:14
whitequarkDocScrutinizer05: you should perform a reaction of O2F2 with H2S15:26
whitequarkwith 1 mole of H2S, the thermodynamic output is equivalent to about 1.5 sticks of TNT15:26
whitequark... at 160K15:26
whitequark... and the result includes hot HF and SF6 at several Mach15:27
whitequarkwpwrak: oh, that story actually mentions FOOF15:29
whitequarkwait15:36
whitequarksomeone *actually* thought dimethylmercury would make a rocket propellant?!!15:37
whitequarkhttp://pipeline.corante.com/archives/2009/10/23/things_i_wont_work_with_straight_dimethyl_zinc.php#44157815:37
DocScrutinizer05wow, nice story16:14
whitequarkyeah16:15
larscas long as you are in the rocket everything is fine ;)16:18
DocScrutinizer05nah, one of the nonsense bits, any gamma radiation from the propulsion motor wouldn't be directed like the yet but omnidirectional and this fusing the remaining red mercury in the fuel tanks16:26
DocScrutinizer05s/yet/jet/16:27
DocScrutinizer05s/this/thus/16:27
whitequarki wonder if it's theoretically possible to make a jet of gamma radiation16:28
whitequarksay some nuclei have magnetic moments, maybe that can be used to orient them or somesuch16:28
DocScrutinizer05just like it's theoretically possible to create a jet/beam of hard x-ray16:28
DocScrutinizer05you know how x-ray laser works?16:29
DocScrutinizer05lead rod, nukler explosion to pump16:29
DocScrutinizer05good I googled before ranting, I thought dimethyl-mercury was used in flu vaccine, but that actually been http://en.wikipedia.org/wiki/Thiomersal16:34
DocScrutinizer05which metabolizes to ethyl-mercury, not methyl-mercury16:35
whitequarkx-ray yes16:39
whitequarkbut gamma?16:39
whitequarkgenerally gamma is not in the energy range of electron orbitals but nuclear stuff16:39
DocScrutinizer05gammy is hard hard x-ray, no?16:39
whitequarkthere is no way a gamma laser is possible, pretty much by definition16:39
whitequarkactually16:40
DocScrutinizer05yep, quite possibly correct16:40
whitequarkthere is no way a gamma laser would work if you use electron shells16:40
whitequarkbut!16:40
whitequarkatomic nuclei can get excited too.16:40
whitequarkis it possible to make a laser out of atomic nuclei?16:40
DocScrutinizer05interesting question16:40
whitequarkwow16:41
whitequarkhttp://en.wikipedia.org/wiki/Induced_gamma_emission16:41
whitequarkapparently this was investigated closely16:41
DocScrutinizer05could a gamma photon interact with an atomic nucleus on a quantum level16:41
whitequarkyes16:41
whitequark"Aircraft power[edit]16:42
whitequarkIn February 2003, the non-peer reviewed New Scientist wrote about the possibility of an IGE-powered airplane, a variant on nuclear propulsion.[8] The idea was to utilize 178m2Hf (presumably due to its high energy to weight ratio) which would be triggered to release gamma rays that would heat air in a chamber for jet propulsion."16:42
whitequarkby the way, DocScrutinizer05, directly answering your question: http://en.wikipedia.org/wiki/Hafnium_controversy16:43
whitequarkthat Hafnium thing (and presumably mercury thing too) is an example of IGE, therefore it will exactly make a jet16:43
DocScrutinizer05IGE alone doesn't create sufficient preconditions for a laser type directed beam of photons16:47
DocScrutinizer05actually, if it did, the whole chain rwaction idea was a dream16:47
whitequarkwhy?16:48
DocScrutinizer05since you would need to provide infinite amount of triggering photons due to the secondary photons not seeing any diversity in vector16:48
DocScrutinizer05a laser always relies on spontaneous emission of primary photons in random directions, while a set of reflectors makes a certain vector go thru the medium multiple times to amplify16:50
DocScrutinizer05except laser boosters16:51
whitequarkright16:51
whitequarkno gamma reflectors16:51
DocScrutinizer05that's the point16:51
DocScrutinizer05particularly not direction of jet16:52
whitequarkokay, there are gamma ray collimators16:53
whitequarkfiltering16:53
DocScrutinizer05the whole story is quite nicely featured with real background, but has a few elementary deficiencies16:53
whitequarkperhaps you could provide enough triggering gamma rays created with usual means, and then amplify16:53
DocScrutinizer05all in lime with Lem, who as well did his stories like "what would result from ignoring one physical fact"16:54
DocScrutinizer05line*16:54
DocScrutinizer05well, yes, amplification was possible16:54
DocScrutinizer05but then, how is this whole red mercury supposed to be stable for a halflife of >1h despite the spontaneous "decay", but still amplifies a trigger gamma ray that much?16:56
DocScrutinizer05somehow you had to increase the number of photon-nukleus "quantum collisions" significantly above what happens in reactor during charging the material and the natural decay that happens there all the time16:58
DocScrutinizer05that's why you need the reflectors in a laser16:58
DocScrutinizer05the reflectors don't work ideal, creating a photon that always hits same atoms on its way back and forth16:59
DocScrutinizer05thus the number of atoms interacting with one photon is multiplied17:00
DocScrutinizer05actually shape of reflectors is an own science, there are bi-convex, plae-convex and whatnot else setups17:02
DocScrutinizer05plane*17:02
DocScrutinizer05even when you assume the number of excited7charged atoms in that red mercury was next to 100% (which is impossible), it still is a pretty short distance that your igniting photon can travel through the booster medium17:04
DocScrutinizer05admittedly along the jet it might be much longer though ;-)17:04
whitequarkmmm, not necessarily17:05
DocScrutinizer05effectively making for a good gamma boost of the radiation following the missile trajectory backwards17:05
whitequarkit would depend on the capture cross-section for the mercury nucleus for gamma rays17:05
whitequarkand nucleus is pretty damn small17:05
whitequarks,capture,absorption,17:06
DocScrutinizer05one thing's for sure, the liguid(?) red mercury would be magnitudes "better" at boosting own decay than the gas in the propulsion engine17:07
whitequarkthere is actually a reference: http://www.nist.gov/pml/data/xcom/index.cfm17:07
whitequarkso for gamma rays at 1MeV it's 7e-02 cm^2/g17:10
whitequarkI wonder how that compares to lasers17:11
DocScrutinizer05leading nowhere for me right now. I fail to understand how they trigger spontanoeus release of all stored energy at all17:11
whitequarksame as for lasers?17:12
DocScrutinizer05for lasers the pumping energy is magnitudes higher than the output17:24
DocScrutinizer05in this case they "pump2 the medium in a nuclear reactor over weeks, aiui.17:25
DocScrutinizer05for a laser the lase action is started by photons from spontaneous decay17:25
DocScrutinizer0599.999...% of those are lost due to incorrect vector17:26
DocScrutinizer05for laser boosters you send in a laser beam, and after travesing the boost nedium, it got amplified by a factor of maybe 2 or 5 or 10, afaik17:27
DocScrutinizer05which in our case would mean you need to create at least 10% of the energy by a trigger gamma source17:27
DocScrutinizer05I still don't get it how they plan to use this flourescence effect to e.g. trigger hydrogen thermonuclear fusions17:28
DocScrutinizer05without a fission gamma ray initial source17:29
DocScrutinizer05you probably could build a critical-mass engine like for fission bombs with that Haffnium17:30
larscIf this channel wasn't monitored before after today it certainly is ;)17:31
DocScrutinizer05which I guess been the purpose of TLAs making this guy publish that story ;-)17:34
DocScrutinizer05we should ask about our paycheck ;-P17:34
DocScrutinizer05(critical mass canon/implosion type engine) this obviously defeats transporting hige amounts of same substance in a tank and "burning" it in a propulsion engine17:56
DocScrutinizer05huge*17:56
DocScrutinizer05just lije you can't transport 2 tons of plutonium in a tank and then use chunks of 10kg out of that to create a critical mass in any propulsion engine17:58
whitequarkProject Orion did that? :D17:58
apeletelarsc: is there a way to inspect data transfered by the dma ?18:26
apeletelarsc: flshing before write + invalidating after read didn't cut it yesterday:18:27
apeletehttp://paste.debian.net/78723/18:27
apeletehttp://paste.debian.net/78725/18:27
pcercueiapelete: md5sum18:27
apeletepcercuei: what do you mean ?18:34
pcercueiread a chunk of data using DMA, read the same chunk without DMA, and compare them18:36
apeleteok, let's see how I can do that in the mmc driver code18:51
wpwrakwhee, just saw that neo900 crossed the 300 units mark !19:03
pcercueiunits produced?19:25
wpwrakunits "donated for"19:26
pcercueiapelete, no response so far to your mail?19:28
apeletepcercuei: not yet, justin may be busy these days, let's just give him some time19:30
nicksydneywpwrak: ping !21:41
wpwrakmoo21:47
nicksydneyhahhah....so you have a cow now :)21:49
nicksydneyhahaha21:49
nicksydneywhat's happening with anelok...what you working on ?21:49
nicksydneyany riots happening there ?21:59
wpwrakno riots just yet. today was quiet. yesterday was ... interesting: http://www.argentinaindependent.com/currentaffairs/newsfromargentina/argentina-news-roundup-27th-january-2014/#comment-21320822:00
nicksydneyinteresting22:02
wpwrakanelok ... still have making the y-box board in the queue. solved the cnc mill issues but still 'need to etch and then solder.22:02
nicksydneyyeah for etching :)22:03
nicksydneyhttp://www.zerohedge.com/news/2014-01-28/chart-day-orders-computers-and-electronic-products-plunge-1993-levels --> seems like people are spending less buying computer...more tablets i presume :)22:03
nicksydneynot sure if this is true anybody from Russia can confirm this http://www.zerohedge.com/news/2014-01-28/russian-bank-halts-all-cash-withdrawals22:04
nicksydneywas reading through this forums http://forum.odroid.com/viewtopic.php?f=77&t=3342 ... sounds like a hardware issue ?22:08
larscapelete: so one thing you can try is to mark the memory with a pattern e.g. 0x5522:09
larscand then check if the content of the memory has changed after the transfer22:09
larscto figure out whether no data was transferred or whether garbage was transferred22:09
apeletelarsc: so I put the pattern in the data buffer before passing it to dma ?22:11
larscyes22:11
larscbefore startingthe dma transfer22:11
apeletelarsc: ok, but I was actually wondering *how* to check the content of the memory22:12
wpwraknicksydney: what on earth does "top 200" mean ? really really tiny ?22:12
apeletelarsc: I think the data is held in data->sg, so I just print the content of data->sg ?22:13
nicksydneywpwrak: when it comes to bank it's always "Top xxx" .... it's like the bigger your "assets" are the bigger named and famed you are :022:15
wpwraknicksydney: did you know that you're in the Top 22 M richest people of Australia ? ;-) the bad news is of course that you're probably in the Top 22 M poorest as well ...22:16
wpwrakwell yes, the more play money you have, the more fun22:16
nicksydneywpwrak: analyst love to play with stats...just look at housing analyst they love to say prices has risen by 1.8% or some numbers and it sounds sooo big !22:17
wpwrakwell, if it's a daily increase ...22:18
nicksydneydown here the only conversation that people love to talk is about house prices and how "rich" they are as they have BIG debts and have a house....if you don't have a house or you renting you are considered "poor" i'm one of those that never wants to buy in this crazy environment22:19
wpwrakowning a house gets kinda pricey if you like to live in a city22:23
nicksydneywpwrak: over here even if you live 50-55km away from the city it's costing you a good 600k for a 300-350m house :)22:24
nicksydneythat's the minimum 22:24
nicksydneyso pretty much you are working as a slave to the bank until you die or you are old !22:24
pcercueidollars?22:26
nicksydneyyup22:27
wpwrakpricy. in buenos aires the square meter of an apartment costs 1.5-2.5 kUSD.22:28
nicksydneywpwrak:  to give you an idea of madness.. check out this http://www.realestate.com.au/property-house-nsw-epping-115940927 .. listing for a cool $1,000,000  ... and see how far it is from city https://maps.google.com/maps?saddr=7+Boronia+Ave,+Epping,+NSW,+2121,+Australia&daddr=Sydney+NSW,+Australia&hl=en&sll=-33.77452,151.075722&sspn=0.008017,0.013239&geocode=FUik_P0dijsBCSnZ6022IKQSazHRoUhtG3Z7yg%3BFSE5-22:29
nicksydney_0dTjwDCSk_dJrzOJgSazFQqDIWaH0BAw&oq=sydney+cb&t=h&mra=ls&z=13 .... the house in that listing is  circa 1925 !22:29
nicksydneywpwrak: this same apartment in that area http://www.realestate.com.au/property-apartment-nsw-epping-115881555 is costing $775,000 for 3 bedroom :)22:30
pcercueimove to Argentina :o22:32
wpwrakwell, it looks rather fancy. of course, the really rich have their mansions far from the riff raff ...22:34
nicksydneyonce 3d printing for concrete is available building house will be cheaper :)22:38
nicksydneyuhh nice .. https://play.google.com/store/apps/details?id=com.stackexchange.marvin22:38
wpwrak(apartment) i think i can compete with that ;-) triplex, 3 bedrooms, living room, large kitchen, two terraces, one large bathroom, one small bathroom, one toilet, two balconies, one parking space. bought it for i think USD 120k + fees and taxes. had to put some ~20k into renovating, though.22:39
nicksydneywpwrak: (apartment) in Argentina ?22:43
wpwraklocation isn't the poshest area but well-connected. subway is at ~150 m, countless bus lines at ~50-150 m, train station - if i should ever feel especially suicidal - some 350 m22:44
wpwrakyes, in buenos aires22:44
wpwrakhmm, city map says i should double my distance estimates22:47
nicksydneywpwrak: that's very good and makes more sense in terms of buying 22:47
nicksydneywpwrak: hell even buying in the US makes more sense than here :)22:48
wpwrakwell, wait for the bubble to burst. then snatch up some nice sprawling mansions some poor sob can't afford anymore :)22:49
nicksydneywpwrak: that's the only thing that I can do..been waiting for 4 yrs now :)22:52
nicksydneywpwrak: it went down from 2009-2011/12 but then it picked up because the F**** morons in central bank cut the interest rate !22:52
wpwrakhehe :)22:53
nicksydneyrathen than stressing about it..better do some MCU programming :)22:56
nicksydneyhehe22:56
wpwrakyeah, that's relaxing22:57
nicksydneyreading on AVR this 2 nights and understood now that programming it is a chicken-and-egg situation....you need an AVR to program an AVR 23:03
nicksydneythinking i'm falling in love too with Microchip PIC MCU :)...tiny but yet powerful 23:03
wpwrakjust forget all that old junk. arm is the future23:04
nicksydneyof course that's the future.23:06
nicksydneythis guy was able to program the AVR using a button http://hackaday.com/2012/09/24/programming-a-microcontroller-one-bit-at-a-time/ :)23:06
nicksydneyhahaha23:06
wpwrak3 buttons :)23:11
nicksydneyhey perhaps we can try something like that for SWD :)23:13
wpwraki wish thee luck :)23:17
nicksydneywpwrak: you see this before right http://www.warpboard.org/ ?23:20
nicksydneywpwrak: how does normally MCU like warpboard communicates with another processor like a Cortex-M4 ? does it use like spi ?23:22
nicksydneywpwrak: from that project we can borrow the 'wireless charging' solution...it uses M4 for that23:29
--- Wed Jan 29 201400:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!