DocScrutinizer05 changed the topic of #qi-hardware to: Copyleft hardware - http://qi-hardware.com | hardware hackers join here to discuss Ben NanoNote, atben / atusb 802.15.4 wireless, and other community driven hw projects | public logging at http://en.qi-hardware.com/irclogs and http://irclog.whitequark.org/qi-hardware
<wpwrak> there's also a rough overview in the project wiki: https://gitorious.org/anelok/pages/Home
<DocScrutinizer05> well, you know I always love to read (and comment on) new schematics, but I'm not git's greatest fan
<wpwrak> it works rather well :)
<wpwrak> and especially if you use it for reading ;-)
<wpwrak> but now the schematics are behind whatever ailment the qi-hw servers have this time :(
<DocScrutinizer05> size of device sounds a tad.. huge?
<wpwrak> about 3 x 8.5 x 1.5 cm. a but smaller than my samsung x830.
<wpwrak> the height is determined by battery and wheel. they'd on opposite sides. if i wanted them on the same side, the surface would have to grow
<wpwrak> the width is determined first by the display and then by the wheel. the wheel is already relatively small.
<wpwrak> the length is mainly determined by layout density. for prototyping, i want a DIY-able design, so density is low.
<wpwrak> but even with multilayer, BGA, and so on, there wouldn't be a LOT you could take away. maybe 1 cm, 2 cm if the rest is to be very crowded (wheel touching the display so it'll be hidden behind your fingers)
<DocScrutinizer05> mhm
<DocScrutinizer05> 1.5cm for battery?
<DocScrutinizer05> can't you use CR2532?
<wpwrak> CR2032. that's 3.2 mm plus about 0.5 mm for clip and tolerance. then 1 mm plastic wall.
<wpwrak> above it is the PCB, 0.8 mm. on the PCB is the wheel, 5 mm.
<DocScrutinizer05> ooh
<wpwrak> so that's 5 + 0.8 + 3.2 + 0.5 + 1 mm in total
<wpwrak> hmm. 10.5 mm. lemme check :)
<DocScrutinizer05> 2 walls
<wpwrak> ah yes, 10.4 mm. not 14 mm :)
<wpwrak> only one - the wheel goes to the surface
<DocScrutinizer05> so the battery area is so large it can't stay clear of wheel?
<wpwrak> the wheel is 22 mm, the battery 20 mm, the display 35 mm. you'll want a bit of space between display and wheel. and the antenna needs a bit of room as well, at least 5 mm.
<wpwrak> then add two double walls. so having it all on the same side would make it ... 88 mm plus the wheel-to-display space
<wpwrak> and then there's the power switch. as fat as the battery.
<DocScrutinizer05> err, when wheel and battery are different place then how will their ... I don't get it
<wpwrak> you can either put them side by side or one on top and one on the bottom
<wpwrak> you also have a display that goes on the same side as the wheel
<DocScrutinizer05> when 5mm wheel and 3.2mm battery add up to heght, then battery and wheel are stacked and I dunno why they would be
<wpwrak> correct. that way, i don't get a huge surface
<DocScrutinizer05> I thought you could place battery under display
<wpwrak> else, i'd have to make the device a fair bit longer.
<DocScrutinizer05> assuming that display is <5mm
<wpwrak> the display is quite thin, 1.6 mm
<wpwrak> needs clearance (at least for tolerance) above and below, and also the top plastic (it's too fragile to be exposed)
<eintopf> hello
<DocScrutinizer05> I'm quite sure you know how to stack a ME. But it seems I can't imagine what it looks like
<wpwrak> so that's about 3 mm for the display. i also sneak a few components under it. they have about 2 mm maximum vertical room, preferably a little less
<roh> hm. do you have a cad file?
<wpwrak> nothing complete yet
<roh> maybe something simple like in openscad.. just to visualize
pcercuei has quit [Quit: dodo]
<wpwrak> top has, from left to right: antenna, display, a LED, wheel
<wpwrak> bottom has: inaccessible space under wheel, battery, switch, usb, uSD, keep-out area under antenna
<eintopf> how do I programm the mcu without dfu?
<wpwrak> which mcu ?
<eintopf> oh I thought there is some mcu on this :)
<wpwrak> ah :)
<DocScrutinizer05> hmm, I had placed the battery under display
<DocScrutinizer05> I.E. between display and PCB
<wpwrak> yes, there is. for the initial programming, via SWD (a reduced JTAG) on the large test points
<wpwrak> there's not enough room here :(
<wpwrak> also, access to the battery would be difficult
<DocScrutinizer05> slot-in
<wpwrak> messy :)
<DocScrutinizer05> nah
<wpwrak> and the stack would get too height, so the whell would have to be raised
<DocScrutinizer05> 3.2mm plastic block with 20mm hole in it
viric has quit [Ping timeout: 240 seconds]
<wpwrak> you need something on top that pushes the battery onto the pcb contact
<DocScrutinizer05> makes for nice slide to push in / pull out the battery, and same time creates a flush surface of case
<wpwrak> also, you'll hit the display connector
<DocScrutinizer05> sure, you use a steel spring for that rougjly M shaped
<wpwrak> i can't make such parts :(
<DocScrutinizer05> this isn't too hard. But you can find such thingies in other chinese 1$ toys
<wpwrak> my solution is some unpleasantly soft metal that wraps around the battery and that is supported by the cover. works okay-ish but would fail in a more demanding setup
<wpwrak> yes, all the toys have their very nice custom-made battery contacts
<wpwrak> and all you can find off the shelf are some monstrosities
<wpwrak> with the battery in that area, the layout would probably become unroutable for DIY
<DocScrutinizer05> even more easy: slot in pcb, solder a L shaped spring wire on long end to opposite side of PCB so that short arm of L goes thru the slot in PCB, then push outer small side of battery against it
viric has joined #qi-hardware
<roh> nah.. just use a readymade battery holder. developing your own without a team of seasoned me is madness imho
<wpwrak> roh: i think there's no choice. the ones you find at the usual shops are huge. but you don't have to design it from scratch, of course.
<DocScrutinizer05> make the hole in plastic block a pit (sorry missing the english term for "sackloch") and place some elastic foam or spring in it
<roh> wpwrak: true. most are 'huge'
<wpwrak> a battery-sized slot would again make the design unroutable
<roh> ive seen some quite flat ones too. intended for used standing up
<roh> one could mount them on the edge of the pcb at one end, and use them as slot-in from the side
<DocScrutinizer05> huh, nobody said "battery sized" - just 2 mm to allow for some spring load
<DocScrutinizer05> when inserting battery
<wpwrak> roh: i experimented quite a bit with CR2032 holders. most of them need excessive force. they're basically designed to hold the battery without any mechanical support from the case.
<wpwrak> roh: and even the smallest ones eat a lot more space than just the battery plus a contact
<roh> wpwrak: true. any way.. i guess thats the most difficult part to solve besides trhe display cover
<wpwrak> DocScrutinizer05: you mean a spring parallel to xy for the bottom and a spring parallel to xz for the top/side ?
<wpwrak> then you need a roof against which you can push ...
<DocScrutinizer05> yes, you could put it that way
<DocScrutinizer05> the roof would be the display, no?
<DocScrutinizer05> but I already improved Version1
<wpwrak> no mechanical load on that fragile little thing, please :)
<roh> hehe.. now you know why i asked for a drawing ;)
<roh> what kind of display is it? and where to get it cheap?
<roh> i see.
<wpwrak> it's pretty nice
<roh> ack. bit pricy as a single part.. but thats displays ;)
<wpwrak> if you want brutally cheap, try this one: http://www.buy-display.com/default/oled-display-module.html
<wpwrak> alas, already getting a bit small
<DocScrutinizer05> btw why a CE2032?
<DocScrutinizer05> is that critter eating that much, or running that long?
<DocScrutinizer05> wouldn't a CR2016 do?
<wpwrak> well, display, rf, ...
<roh> DocScrutinizer05: 2032 is nice. everybody has some around
<wpwrak> i calculate a battery life of about two months if used a few minutes per day
<DocScrutinizer05> that RF stuff is still quite mystery to me
<wpwrak> and yes, 2032 has good availability
<DocScrutinizer05> hm, 2016 as well
<wpwrak> MUCH less than 2032
<DocScrutinizer05> LOL, my CR2016 suddenly started moving around, that jumping up into the air
<wpwrak> it hates the truth ;-)
<wpwrak> maybe it's related to keith alexander, head of NSA, these days very busy lying to congress :)
dos1 has quit [Ping timeout: 245 seconds]
<DocScrutinizer05> s/taht/then/
<roh> hm. what is about putting the whole thing into a usb-stick case and using a tiny lipo pack?
<roh> so.. basically no removable parts, and use it with a usb-A male to usb-A female cable or plug it in directly
<wpwrak> naw, rechargeable batteries become a replacement problem after a few years
<roh> wpwrak: one-time use is even worse
<wpwrak> if you don't want that problem on your hands you need to pick a very common battery, and these are all large
<roh> make the case with screws and piggypack on some common size
<wpwrak> i mean that it gets hard to find a replacement. i don't expect cr2032 to vanish anytime soon :)
<roh> ive seen some quite tiny lipo-packs complete with protection circuit and wires
<wpwrak> soldering to replace is not nice
dos1 has joined #qi-hardware
<roh> then plug it
<roh> like all the usb-mp3-stick-style players do too
<wpwrak> and you get more problems, e.g., shipping
<DocScrutinizer05> which gives me another idea :-D MAGNET
<wpwrak> dynamo ? :)
<DocScrutinizer05> magnets make GREAD battery contacts for a CR2023
<DocScrutinizer05> GREAT even
<wpwrak> is cr2032 magnetic ?
dos11 has joined #qi-hardware
dos1 has quit [Read error: Connection reset by peer]
<wpwrak> that's a magnet ?
<wpwrak> but a magnet would add height, too ...
<wpwrak> naw, the only other place where the battey could reasonably go would be between display and wheel. no crazy vertical stacking.
<DocScrutinizer05> embedd into PCB!
<wpwrak> ... making the layout unroutable :)
<wpwrak> ultra-thin monolith on "Oe" ?
<wpwrak> in any case, even if you do all the crazy battery stacking, you won't will all that much in the end
dos11 has quit [Ping timeout: 264 seconds]
<DocScrutinizer05> http://maemo.cloud-7.de/share-service/20131006_005.jpg ff - I know there exist even smaller magnets
<whitequark> that's a really strange way of mounting a battery
<wpwrak> there's only about 2.2 mm between pcb and display. so you need to add some 1.5 mm for battery and top contact or whatever
<wpwrak> and you still have components on the bottom. so that's another 1.2 mm to 1.5 mm
<wpwrak> so in the end you may save only about 1 mm
<DocScrutinizer05> 2.2mm seems just fine for a CR2016 with magnet and side-spring
<DocScrutinizer05> so in the end you could even use the blanks PCB lower side as outside of your case
<DocScrutinizer05> when you manage to move those other components between PCB and display as well
<wpwrak> cr2016 is too exotic. and you'd then have to swap batteries every three weeks or so.
<DocScrutinizer05> ohwell
<DocScrutinizer05> dunno what's regarded exotic in your place
<wpwrak> and if that space is taken up by the battery, the other components won't feel too well there :)
<DocScrutinizer05> here you get them in supermarket
<DocScrutinizer05> next to Cr20232 and AAA
<DocScrutinizer05> OK, a bit pricey
<DocScrutinizer05> like 2 EUR per battery
<DocScrutinizer05> CR2032*
<wpwrak> if you buy one at a time, at your pharmacy ;-)
<roh> hm. for 4 you get a lipo cell, for another 2 the charger-parts
<wpwrak> both should be about USD 0.2 apiece if you buy 5 or 10
<wpwrak> so, < 1 USD battery cost per year with CR2032
<wpwrak> 4 batteries or 4 EUR ?
dos11 has joined #qi-hardware
<wpwrak> yup
<wpwrak> see why cr2016 is not such a nice idea ? :)
<DocScrutinizer05> nope, exactly I don't see it
<DocScrutinizer05> it's more than 1/3 of a 2032
<wpwrak> same price as 2032 but less than half the capacity
<DocScrutinizer05> negligible
<wpwrak> actually, 2016 is slightly more expensive
<wpwrak> negligible capacity, yes :)
<wpwrak> you wouldn't need a holder with cr2016 but a belt, like for a machine gun ;-)
<DocScrutinizer05> then your design is faulty, to begin with
<wpwrak> ;-)
<DocScrutinizer05> too energy hungry
<DocScrutinizer05> I mean, it does... nuttin, 99.99999% of the time
<wpwrak> yes, the idle time doesn't matter. i expect just a few uA
<DocScrutinizer05> nah, OLED are greedy, no matter what that site tells
dos11 has quit [Ping timeout: 245 seconds]
<DocScrutinizer05> creating light isn't for free
<wpwrak> about 10 mA for 20% pixels on. less than anything else.
<wpwrak> besides, a traditional LCD would be much thicker ;-)
<DocScrutinizer05> exactly
<DocScrutinizer05> e-paper
<wpwrak> we've been through that. you really should read the list :) epaper is about the worst possible choice for a password safe ...
<DocScrutinizer05> though, there been that "electronic tag" thingie. I can't recall what it been
<roh> hmm.. some holder like this could work: https://www.sparkfun.com/products/10592
<wpwrak> well, they also go under "e-paper"
<roh> the other contact is the pcb itself
<wpwrak> roh: HUGE
<roh> usually one does hasl finish then, adding gold
<wpwrak> roh: and you haven't seen the insertion force of these things :)
<DocScrutinizer05> roh: I already suggested that, he doesn't like it
<roh> wpwrak: well.. smaller than the ones with a plastic holding the other pin
<wpwrak> roh: yes, a medium-sized dinosaur is smaller than a whale ... ;-)
<DocScrutinizer05> how's insertion force a problem?
<wpwrak> DocScrutinizer05: you need a crowbar to get that battery out ...
<roh> less insertion force means bad contact.
<DocScrutinizer05> usually not
<DocScrutinizer05> (whale > dino)
<wpwrak> DocScrutinizer05: i've experimented quite extensively with these holders. just trust me ;-)
<roh> there is a reason for the one there. it doesnt fall out. so one calculates the minimum at rated current, and then ramps that up to whatever is needed to hold the weight of the battery on given spec gfoces.
<DocScrutinizer05> I know those holders. I always thought it quite assuring when battery sits tight
<wpwrak> roh: naw, they need all that pressure not for electrical contact but because they don't have a case
<wpwrak> there's some nice strong plastic around it ...
<wpwrak> ... unless it figures out how to quantum-tunnel though it, it really won't go anywhere else
<DocScrutinizer05> *shrug*
<wpwrak> roh: the SMT ones also don't stick too well to the PCB (especially considering the enormous force). better use a through-hole version.
<whitequark> wpwrak: I (still) don't really get your choice of OLED. don't LCDs eat hundreds of uAs instead of tens of mAs?
<roh> wpwrak: ive heard about that.. but never had an issue on proper fr4 boards
<roh> i guess the homemade ones delaminate more easily
<wpwrak> almost twice as fat as the cr2032 ;-)
<wpwrak> whitequark: it's the backlight that kills you ...
<whitequark> OLED probably eats less than LCD with backlight, but do you really need that?
<whitequark> and if you do: can't you add a button which turns on backlight? like they do on watches
<DocScrutinizer05> >>Awesome, it looks like a cartoon cat!<< LOL
<wpwrak> non-backlit lcds are often a pain to read. specially if you have something more complex than huge 7 segment digits
<whitequark> wpwrak: dunno, I find non-backlit monochrome LCDs pretty readable. I may be biased though
<whitequark> dot matrix
<wpwrak> another issue is that panel sizes go up when you move to this kind of display
<wpwrak> that is, for off the shelf parts
<whitequark> w/h?
<wpwrak> and everyone seems to integrate a backlight
<wpwrak> mainly thickness, but also junk around the active area
<DocScrutinizer05> dang, I moved the N900 to the other room again, so no photo of my homematic CCU LCD display
<wpwrak> of course, LCD can be very think. but it seems that these are custom designs again.
* whitequark looks at a nokia3310 display lying around
<wpwrak> err, very THIN
<DocScrutinizer05> OLED start to get exciting when I can *roll* then out of the device, to get a 10" screen
rz2k has quit []
<wpwrak> get a laser beamer ;-)
<DocScrutinizer05> meh
<DocScrutinizer05> TI blaze already *has*beamer integrated
<DocScrutinizer05> doesn't sound too good for standard usecase though
<DocScrutinizer05> it's more about idiots drooling over a 15" video projection at the wall, in an absolutely dark room 8maybe the videos that would be siutable for such usecase line up nicely with dark rooms)
<whitequark> this one is 1.8mm thick
<whitequark> dunno if you can source them
<DocScrutinizer05> whitequark: and probably even has backlight
<wpwrak> backlight = power-hungry
<whitequark> DocScrutinizer05: nope, backlight was on the pcb
<DocScrutinizer05> ohwell
<whitequark> LEDs mounted so they emit light in parallel with the PCB
<wpwrak> with oled, i can control the power drain by keeping as many pixels as possible dark
<wpwrak> with lcd it's all or nothing
<whitequark> wpwrak: but really, how much of the time will you *need* (back)light?
<whitequark> you have a freakin' big light right in front of you (computer display) at all times, for that matter
<whitequark> or phone display, though that works somewhat worse
<wpwrak> depends on the display type. some need it all the time. some can work with ambient light, but have poor contrast
<wpwrak> that big light is more likely to blind you than do anything else ...
<whitequark> the picture does it a favor, but contrast is usually pretty good on the monochrome ones
<DocScrutinizer05> the standard LCD like used in r0ket is ~3mm, *with* backlight
<wpwrak> these are the ones that you can't find in a reasonable size
<whitequark> DocScrutinizer05: huh, r0ket is cool. like, really cool. thanks for the accidental heads-up
<whitequark> I needed something like that a month ago
dandon_ has joined #qi-hardware
<whitequark> ... actually, it is exactly what I need
dandon has quit [Ping timeout: 240 seconds]
dandon_ is now known as dandon
<DocScrutinizer05> they do an awesome trick: using photoelectric voltage of a LED for ambient light sensor. Could almost be one of my kinky ideas
<DocScrutinizer05> "either let it emit light, or use it to detect light" AWESOME!
<DocScrutinizer05> dang, you probably could detect the own photon echo during power down phase of PWN operation ;-P
<DocScrutinizer05> nah, not really, for that they are a tad too slow
<wpwrak> flash the LED, then measure how long it takes until the light is reflected by the moon :)
<DocScrutinizer05> that could work, by timespan between saturation decay and echo coming in. But _only_ for the timespan
<wpwrak> do it enough times and you can nicely filter the data out of the noise ;-)
<DocScrutinizer05> though, a pulse of THAT energy will take quite a while to dissipate all the free electrons in the bandgap ;-P
<DocScrutinizer05> I actually think LEDs are max in the MHz range, for 10/90
<DocScrutinizer05> for the high-bandwidth glasfibre you need special LEDs, no?
viric has quit [Ping timeout: 240 seconds]
viric has joined #qi-hardware
<wpwrak> hmm, board is about the maximum size i can still tin-coat manually. much wider and i can't reach things properly
caLico_jAcK has joined #qi-hardware
caLico_jAcK has quit [Quit: Leaving]
<wpwrak> via wires soldered. hope i got all of them - there's a lot
<wpwrak> now .. let's add some components ..
<wpwrak> blargh. transceiver clock output shorted. that was messy :(
pcercuei has joined #qi-hardware
xiangfu has joined #qi-hardware
kilae has joined #qi-hardware
viric has quit [Ping timeout: 240 seconds]
viric has joined #qi-hardware
xiangfu has quit [Read error: Connection reset by peer]
pcercuei has quit [Quit: brb]
pcercuei has joined #qi-hardware
pcercuei has quit [Quit: brb]
pcercuei has joined #qi-hardware
dos11 has joined #qi-hardware
rz2k has joined #qi-hardware
arossdotme has quit [Ping timeout: 252 seconds]
<wpwrak> hmm, qi-hw servers still down :-(
<wpwrak> i wonder if the ghost of wolfspraul reads this ...
dos11 has quit [Ping timeout: 246 seconds]
<larsc> let me pull out my ouija board, maybe I can reach him
<viric> ip-over-ouija?
viric_ has joined #qi-hardware
viric has quit [Ping timeout: 240 seconds]
viric_ is now known as viric
<larsc> I'm only getting a "No route to ghost" error...
<mth> did you configure a gateway to the spirit world?
qi-bot has joined #qi-hardware
arossdotme has joined #qi-hardware
<larsc> just a long roundtrip time, server is up again
arossdotme has quit [Ping timeout: 240 seconds]
arossdotme has joined #qi-hardware
viric_ has joined #qi-hardware
viric has quit [Ping timeout: 240 seconds]
viric_ is now known as viric
<wpwrak> whee, the ghost did read it ! :) thanks, ghost of wolfspraul !
<wpwrak> DocScrutinizer05: now you can see all the goodies of anelok
<wpwrak> a display and wheel prototype (slightly smaller than the latest design): http://downloads.qi-hardware.com/people/werner/anelok/tmp/anelok-dui.ogv
arossdotme has quit [Ping timeout: 260 seconds]
<wpwrak> (that one has the wheel very close to the display. i've now increased that space by a few mm)
<wpwrak> and you'll like this: a little box that implements a USB y-cable: http://downloads.qi-hardware.com/people/werner/pwsafe/ybox.pdf
<wpwrak> but most of september was really about anelok: http://lists.en.qi-hardware.com/pipermail/discussion/2013-September/thread.html
<larsc> wpwrak: my psychic skills needs some work, it took some time to wake to spirit ;)
<wpwrak> larsc: it still worked impressively quickly :)
garlp has joined #qi-hardware
<garlp> Hi. My codepage is right?
arossdotme has joined #qi-hardware
<garlp> halloooo
<wpwrak> hmm, where's a klingon example in proper unicode when you need one to copy & paste ?
<wpwrak> (anelok) meanwhile, the key components (transceiver, MCU, memcard, usb, a few caps) are soldered. should be able to try SWD soon.
<wpwrak> kinda curious whether the mcu will be willing to talk ...
<wpwrak> mcu was relatively easy to solder. positioning is hard, as usual with QFN in a DIY process. transceicer had a mystery short between ground and clock out. memcard may have some footprint issues - the contacts don't seem to be quite where they should be. needs checking.
<wpwrak> usb looks intimidating (the shell covers the signals almost completely) but was surprisingly easy to solder.
<larsc> the size is that of a usb mem stick?
garlp has quit [Quit: Miranda NG! Smaller, Faster, Easier. http://miranda-ng.org/]
garlp has joined #qi-hardware
<garlp> hi. My codepage is right?
<wpwrak> there are probably some very large ones with roughly that size :)
<larsc> wpwrak: yea, I was thinking about those back from a few years ago
<larsc> I find it amazing how fast you were able to put this together
<garlp> hi from Russia
<larsc> hello
<garlp> larsc: my codepage is right? ok
<larsc> at least the ascii part of it
<wpwrak> it's about half the size of a PCMCIA card
<wpwrak> couldn't find any of my ancient (and huge) usb memory sticks. i remember one from sony that was particularly oversized.
<wpwrak> larsc: it's till too early to know whether i actually got it right ;)
<larsc> even for an almost working prototype it was still pretty fast
dos11 has joined #qi-hardware
dos11 is now known as dos1`
dos1` is now known as dos1
viric has quit [Ping timeout: 240 seconds]
<DocScrutinizer05> wpwrak: C20 and C22 look *very* strange
<DocScrutinizer05> if those are correct, I'd not really know why POWERED at B1:2 and why C23. I'd rempve POWERED then and make C23 0R
<DocScrutinizer05> or that at86rf232 is a really strange chip when it needs that circuit
viric has joined #qi-hardware
rz2k has quit []
<wpwrak> well, they say it needs those DC blockers
<wpwrak> "POWERED" just tells kicad that this is a power input. otherwise, i'll get an ERC warning
<wpwrak> (power input for declaring the middle tap as GND)
<wpwrak> maybe i should make it passive. it looks inded a bit odd the way it is now.
<wpwrak> the baluns have rather conflicting descriptions of what should go there. some say "just ground it". that's where the "power in" comes from
<DocScrutinizer05> R9 what the heck?
<wpwrak> everybody loves R9 ;-)
<wpwrak> it's to supply the RTC. the basic ideas is to limit battery current such that the device can be in deep sleep (with RTC running) but can't do anything else
<wpwrak> especially that it can't turn on RF
<wpwrak> well, it could try to. but then the power supply would collapse.
<wpwrak> hmm, the bad think about these tiny components is that, when you divert your attention for a second, you have a hard time finding them again ...
<DocScrutinizer05> when you want to disable RF, switch off AT86Rf232
<DocScrutinizer05> R9 makes the whole circuit behave undefined
arossdotme has quit [Ping timeout: 240 seconds]
<wpwrak> the MCU is supposed to catch power dropping and shut things down
<DocScrutinizer05> that's not a stable design
<wpwrak> and the purpose of this is to protect against a rogue MCU
<wpwrak> yeah, i guess it may bounce around in power-on-reset
<DocScrutinizer05> there's no way for the circuit to detect state of the switch, except try to draw huge amount of current, in which case VDD drops and probably RTC doesn't work either. While it does NOT reliably stop RF since you got buffer Cs
<wpwrak> my idea is to set a high brown-out level and trigger on that. this code would shut down things very quickly, before reaching the low brown-out level
<wpwrak> the caps are way too small to feed RF for anything. so they actually help, buying us a little more time for the cleanup.
<wpwrak> the main cap there is C15 anyway
<larsc> can't you connect that switch to reset, for both the mcu and rf?
<DocScrutinizer05> yeah
<DocScrutinizer05> this is insane design
<DocScrutinizer05> why would the CPU work with "brownout" when it gone rogue regarding any more sane design?
<DocScrutinizer05> and duh, this MCU doesn't need a xtal for rtc??
<DocScrutinizer05> and no separate power VDD pin for rtc?
<wpwrak> it has no separate power pin for RTC
<wpwrak> (unfortunately)
<wpwrak> hence that hack
<DocScrutinizer05> then see what larsc said
<wpwrak> (rogue) i don't expect it to work with brownout if it goes mad. all bets are off then. but i want to be sure that the system can't do anything "interesting" when the switch is in the off position
FrankBlues has joined #qi-hardware
<wpwrak> holding everything in reset may be an option. need to see about that.
<DocScrutinizer05> then switch off the RF chip
<wpwrak> alas, the MCU does not define a reset state for the GPIOs. so they can be anything. i'm sure it makes perfect sense in some parallel universe ...
<wpwrak> properly switching off the RF chip would still require the MCU to cooperate
<wpwrak> but holding it in reset may be an option. need to check the current.
<DocScrutinizer05> kill the clock -> no more RF
<DocScrutinizer05> just a brainstorming idea, when reset doesn't work
<wpwrak> killing the clock means messing with the crystal
<DocScrutinizer05> yeah, sth like that
<wpwrak> well, i guess one could forcible ground it but ...
<wpwrak> but it does sound a bit like asking for trouble
<DocScrutinizer05> probbaly
<DocScrutinizer05> switching off VSYS not feasible?
<DocScrutinizer05> maybe only for one of EVDD DEVDD
<wpwrak> again, power through GPIOs
<DocScrutinizer05> I guess one is logic, the other analog?
<wpwrak> yes
<DocScrutinizer05> doesn't it define what happens when analog is unpowered?
<wpwrak> of course not :)
<DocScrutinizer05> :-S
<DocScrutinizer05> I'd go for pulling RF_nRST to GND. Can get detected by MCU by using PTC2 as input
<DocScrutinizer05> If you want or need that, you also can connect PTC2 to MCU nRESET via a R-C to give a pulse to nRESET when the switch gets closed
<DocScrutinizer05> ooops, no :-P
<wpwrak> hmm yes, that could be a possibility (forced reset)
<DocScrutinizer05> otherwise MCU resets itself
<wpwrak> heh :)
<DocScrutinizer05> needs a tiny bit more birdseed
<DocScrutinizer05> 2 diodes or sth
<DocScrutinizer05> maybe only one
* DocScrutinizer05 heads out, getting "breakfast" coffee
<wpwrak> i think the PTC bank is interrupt-capable, so detection shouldn't be a big deal
<DocScrutinizer05> I anyway *strongly* recommend _not_ to use that R9 "solution". Might work, when you're extremely lucky, but yuo never can completely evaluate and verify the circuit behaviour
<DocScrutinizer05> s/yuo/you/
<qi-bot> DocScrutinizer05 meant: "I anyway *strongly* recommend _not_ to use that R9 "solution". Might work, when you're extremely lucky, but you never can completely evaluate and verify the circuit behaviour"
<wpwrak> and it would be for "cleanup" anyway (set GPIOs, stop expecting RF to work, etc.)
<wpwrak> yeah, it's a bit chancy
<wpwrak> need to see how much RF draws in reset. that's of course undocumented.
<wpwrak> SLP_TR (sleep mode) may or may not have an effect there, too
<wpwrak> in-circuit programmer is unhappy with what i've done :-( let's find out why ...
FrankBlues has quit [Remote host closed the connection]
<wpwrak> btw, there's another "chancy" mode: when removing the battery, the MCU needs to catch the imminent complete loss of power and bring down things, so that C15 can keep the RTC alive during the battery change
<wpwrak> will be fun to make this work :)
rozzin has left #qi-hardware [#qi-hardware]
<DocScrutinizer05> forget it
<DocScrutinizer05> lock the battery lid with the slide switch SW2
<wpwrak> naw, switches are too much of a pain
<wpwrak> they're huge, expensive, and never quite fit
<DocScrutinizer05> if you can't do that, then get two Vbat+ contacts on the PCB, the detector one opening earlier than the main one
<wpwrak> interesting idea
<DocScrutinizer05> or a Reed or Hall or IR-proxy switch on your case, "intrusion detector"
<wpwrak> satellite surveillance ;-)
<DocScrutinizer05> for that your refusal to make battery slot-in actually is beneficial
* wpwrak feels victorious :)
<DocScrutinizer05> adding a tiny magnet to the removable "lid" and a Reed switch on PCB next to it is dead simple
<DocScrutinizer05> you still didn't explain how that magic MCU keeps realtime without an XTAL
<cde> `d
<wpwrak> it has a reasonably accurate RC
<larsc> DocScrutinizer05: built-in xtal
<DocScrutinizer05> nifty
<wpwrak> and we can calibrate it against the very precise xtal
<DocScrutinizer05> `d ?
<wpwrak> the rtc is for things like security tokens. so it doesn't have to be extremely accurate.
porchaso0 has joined #qi-hardware
<DocScrutinizer05> errr
<DocScrutinizer05> sorry, you lost me
garlp has quit [Quit: Miranda NG! Smaller, Faster, Easier. http://miranda-ng.org/]
porchao has quit [Ping timeout: 252 seconds]
<wpwrak> the code they generate normally depends on time (and usually on the number of activations)
<wpwrak> the little critters that generate a code when you turn them on. then you type it into some authentication system. for two-factor authentication.
hellekin has quit [Ping timeout: 245 seconds]
<cde> DocScrutinizer05: sorry, typing mistake
<wpwrak> weird. if i connect the board to the programmer, the programmer doesn't even enumerate on usb
<wpwrak> all the signals look correct, though
<larsc> does the programmer power the device?
<DocScrutinizer05> VBUS
<wpwrak> yes, power looks normal
<DocScrutinizer05> err wait, what exactly are you saying?
<DocScrutinizer05> is the programmer connected to anelok via USB?
<wpwrak> no, via GND, VDD, and SWD
<DocScrutinizer05> aah, and what does it do? is it a own system or just some sort of level shifter to USB?
<DocScrutinizer05> ftdi?
<wpwrak> an evaluation board with built-in (fancy) usb to swd gateway
<wpwrak> tried different firmware. same problem. let's recheck power ...
<apelete> Hi
<DocScrutinizer05> errr
<DocScrutinizer05> you call that a programmer?
<wpwrak> (firmware) programmer firmware i mean. not target :)
<wpwrak> well, it contains one :)
<wpwrak> the opensda critter
<apelete> larsc mth: but something is still going wrong -> http://paste.debian.net/51955/
<DocScrutinizer05> hmm, no idea
<larsc> apelete: lots of a's
<cde> it's the aaaaapocalypse
<apelete> larsc: seems like the driver is not being loaded, and I don't know why it's printing "musb-hdrc: version 6.0, ?dma?, otg (peripheral+host)"
<apelete> in platform data I set it to peripheral only....
<apelete> cde: not funny :D
<wpwrak> in the movies, when someone falls to their death, they usually shout "aaaaa...". maybe that kernel found the choice of last words of these people inspiring.
<cde> in kernel land, no one will hear your scream
<roh> hrhrhr
<larsc> apelete: hm, i can't even find the code that prints that message
<larsc> ah, it was removed
<larsc> it prints that message when the module is loaded
<wpwrak> someone's been hiding evidence
<apelete> larsc: so the module is being loaded ? it doesn't pring any of the printk I left in the glue layer...
porchao has joined #qi-hardware
<larsc> those will probably run only when the device is created
<larsc> or maybe the glue code is build as a module
<larsc> but the core code is built-in
porchaso0 has quit [Ping timeout: 248 seconds]
<DocScrutinizer05> 0xaa is a very common pattern for unititialized RAM
<apelete> larsc: hmmm, I thought there was something amiss with my kernel config file, I checked it but didn't find anything unusual -> http://seketeli.fr/git/~apelete/qi-kernel.git/commit/?h=jz4740-udc
<DocScrutinizer05> 0x55 the other one, besides of course 0xff and 0x00
<larsc> both seems to be built-in
<DocScrutinizer05> basically all memtesters use 0xaa and 0x55 for test patterns
<larsc> I think the kernel also uses it to poison memory
<wpwrak> hmm. disconnecting RESET makes the programmer enumerate. SWD still fails later on, though.
<wpwrak> .. because there is no reset. yes, that makes sense :)
<wpwrak> very strange
<wpwrak> resistance of RESET is normal, too. some 13 MOhm to GND (if driven high), about 50 kOhm to VDD (if driven low)
<apelete> mth larsc: any idea on how I should proceed to solve this ?
<apelete> I tried to step into the musb glue layer code with gdb but it didn't break
<apelete> don't know why
<apelete> I mean, I can see the kernel booting inside gdb, but putting a breakpoint in the probe() function does not break the kernel
<larsc> comment out parts until it is working again
<larsc> this looks like a real bad memory corruption
rz2k has joined #qi-hardware
<wpwrak> *hmm* could it be ... freescale being assholes ? there's a thread suggesting that the "open"sda subsystem simply doesn't enumerate if the chip isn't the one on the board. https://community.freescale.com/thread/303031
<apelete> larsc: ok thanks, will try that
<wpwrak> what's odd with this: 1) if i break reset such that it can't even talk to any chip, it does enumerate. 2) the chip is not identical but quite similar. (same core, same memory, different package)
<wpwrak> IDCODE should be the same for all M0+. hmm.
<larsc> so you think the firmware finds the chip and then shuts itself down?
<wpwrak> i think i found it :) there are two slightly different debug connectors on the board, J6 and J8. i attached to the wrong one. let's try the other one.
porchao has quit [Quit: Leaving...]
<wpwrak> yeah ! talks :)
kilae has quit [Quit: ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258]]
<larsc> now it just need to walk
<wpwrak> let's try something innovative and completely unexpected ... blink a led :)
viric has quit [Ping timeout: 240 seconds]
pcercuei has quit [Quit: brb]
pcercuei has joined #qi-hardware
viric has joined #qi-hardware
porchao has joined #qi-hardware
<wpwrak> kewl. the led blinks on the first try. that was too easy :)
<eintopf> :D
wej has quit [Ping timeout: 245 seconds]
<DocScrutinizer05> lol
<DocScrutinizer05> on a sidenote, I felt a little puzzled about the LED's series R
<DocScrutinizer05> a tad low for a lowpower device
<DocScrutinizer05> you usually use high-efficiency LEDs in this case, which don't need >1mA to properly shine
<DocScrutinizer05> I didn't check the math, but 120R sounds a tad low, thus current probably higher than I'd consider OK
<DocScrutinizer05> sounds more like 10..20mA than 1..2mA
<wpwrak> nice. MCU comes up, SWD works, LED works. also tested that the transceiver has a clock. good enough for today. there's still a lot more to test. the boost converter may still cause a few surprises. that's the last completely new subsystem remaining. well, and USB will be its usual messy self.
<DocScrutinizer05> I still wonder how exactly this thing is menat to get used
<DocScrutinizer05> the description on git doesn't sell it to me
<DocScrutinizer05> do you plug this in to PC USB and it emulates a (2nd) kbd?
<DocScrutinizer05> and what for is the RF?
<DocScrutinizer05> btw wtf SWD?
hellekin has joined #qi-hardware
<wpwrak> you don't like SWD ?
<wpwrak> the basic usage concept is that it stores PWs. then you have a number of ways to use them: 1) display and let the user do something with them. 2) send over USB (as HID). 3) send over RF (as HID). 4) use for challenge-response (with USB/RF). etc.
<wpwrak> RF is just to avoid the cable.
<whitequark> using RF for a password safe... huh
<wpwrak> whitequark: well, encrypted of course
<whitequark> for a typical user would make no difference, I guess
<whitequark> wpwrak: still much more attack surface
<wpwrak> true. you can choose to not use RF :)
<whitequark> but, if someone wants your credentials that much, I guess there are worse problems
<whitequark> so it's probably OK actually
<wpwrak> there's always rubberhose cryptography ...
apelete has quit [Ping timeout: 264 seconds]
<whitequark> I was thinking more along of "looking across the shoulder"
<whitequark> since that's likely much easier to pull than RF snooping
<whitequark> (isn't it rubberhose cryptoanalysis? :)
<whitequark> DocScrutinizer05: wait wait, there actually are LEDs which need <1mA to shine?
<wpwrak> RF or USB prevent this, though. well, unless your login form echoes things back
<wpwrak> (cryptoanalysis) yeah, that's better :)
<DocScrutinizer05> ~wtf swd
<DocScrutinizer05> ooh
<whitequark> DocScrutinizer05: swd is jtag's little two-wire brother
<DocScrutinizer05> whitequark: sure
<DocScrutinizer05> and what the TLA?
<whitequark> same idea, slightly different on the protocol level, works only for arm cores
<whitequark> serial wire debug
<DocScrutinizer05> whitequark: a whole class of LEDs is specified at Ifwd=2mA instead of the usual 20mA
<whitequark> by "slightly" I of course mean "bears no resemblance to jtag whatsoever"
<DocScrutinizer05> serial wire debig \o/ Thanks!
<whitequark> DocScrutinizer05: that's interesting, what's a keyword for them?
<DocScrutinizer05> errr, high efficiency or whatever
<DocScrutinizer05> lemme check
<whitequark> that brings up high-power stuff for me
<whitequark> found them, kingbright lc (low current) leds
<whitequark> wonder how they achieved it. some kind of better, more expensive semiconductor?
<whitequark> as in, what's the catch
<whitequark> wpwrak: neat!
<DocScrutinizer05> low-current actually
<DocScrutinizer05> ugh, couldn't post THAT URL here, but seems SMD are less common on 2mA, but I found a blue and a white one @ 5mA IF
<DocScrutinizer05> and now I finally read backscroll ;-P and notice you already found all this as well
<DocScrutinizer05> (catch) I don't know exactly. something with different dotation(?), transparent chips, reflectors, maybe even different material all together (like GaAsP)
<wpwrak> whitequark: thanks :)
<whitequark> do..tation?
<whitequark> is that even a word?
<DocScrutinizer05> no probably not a word
<whitequark> oh
<DocScrutinizer05> in german it's "Dotierung"
<whitequark> different dopants?
<DocScrutinizer05> my english sucks here
<DocScrutinizer05> it's the traces of other chemical element in semiconductors making it N or P
<DocScrutinizer05> N-dotation P-dotation
<DocScrutinizer05> sorry no idea what's the english term
<DocScrutinizer05> doping
<DocScrutinizer05> hehe
<DocScrutinizer05> so yeah, dopants
<wpwrak> sporty ;-)