| whitequark | roh: hey, do you recall our discussion about smartphones? I've finally determined that for myself | 00:27 |
|---|---|---|
| whitequark | if, and only if, you use your smartphone like a dumbphone, it will easily live for a week. I just confirmed that. | 00:27 |
| whitequark | the problem is, why the hell would you want to do that if you have a smartphone. | 00:27 |
| Action: kristianpaul is using dumbphones since 2000 | 00:31 | |
| kristianpaul | and at that time a palm m100 ( the PDA thing) | 00:33 |
| kristianpaul | and my 1.2M camera of course :) | 00:34 |
| whitequark | kristianpaul: it's a nice thing to have one gadget which works as a music player + gps navigation device + a camera (SGS2 has a camera which is on par with cheap "discrete" ones) + a dumbphone. | 00:34 |
| whitequark | not to mention occasional web browsing, through it's non-vital. | 00:35 |
| whitequark | the point is, ARMs _are_ energy-hungry, and they can convert electricity to heat very fast. they just sleep very well. | 00:36 |
| kristianpaul | same reason i dont buy swiss knife... | 00:36 |
| kristianpaul | so yes you could leave with it, just need to take the desition | 00:37 |
| kristianpaul | same reason you dont reply i guess al your mail in your smartphone | 00:37 |
| kristianpaul | i hope you dont :) | 00:37 |
| whitequark | well, do you keep a knife, screwdriver set, pliers, etc with yourself at all times? | 00:38 |
| whitequark | I need gps once in a month, when I'm lost, and camera maybe once a week, and so on | 00:38 |
| whitequark | I simply don't have the brainpower to remember if I will need a particular function at a particular day | 00:39 |
| whitequark | heck, it was hard enough to remember not to forget the phone itself at home. | 00:39 |
| whitequark | kristianpaul: I would reply to a sizable part of my mail on my smartphone, if not for the virtual keyboard. Unfortunately no one makes good physical keyboards now | 00:40 |
| whitequark | which is kinda disappointing, especially given the fact that Android has _very_ good support for keyboard and mice | 00:40 |
| wpwrak | the world clearly needs new input devices. how about position-of-the-tongue for picking a key ? swallow to enter, burp to delete. | 01:41 |
| wpwrak | kristianpaul: here, dumbphones since around 1996. they get the job done, what more can you ask for ? :) | 01:43 |
| whitequark | wpwrak: touchscreens are great for lots of things | 02:30 |
| whitequark | I'd say for more things you would like to do with a smartphone, except writing text. | 02:30 |
| Action: xiangfu try to use the oven for soldering the chip now | 03:32 | |
| xiangfu | the 4 steps will be: | 03:32 |
| xiangfu | Set to 90C for 3 minutes to preheat | 03:32 |
| xiangfu | Set to 150C for 1 minute for thermal soak. | 03:32 |
| xiangfu | Set to 210C for 1 minute for reflow. This results in a Tal of about 15 seconds. | 03:32 |
| xiangfu | Turn off oven, open door, and cool to ambient with room air | 03:32 |
| xiangfu | base on azonenberg's blog | 03:33 |
| wolfspraul | wow, oven! | 03:33 |
| wolfspraul | you got a random cheap chinese mini-oven? how much did you pay? | 03:34 |
| xiangfu | no. I buy a ACA oven. 270RMB. | 03:35 |
| xiangfu | it have 4 heating pipes. | 03:35 |
| xiangfu | <250RMB. all is 2 heating pipes. | 03:36 |
| xiangfu | this one: http://www.360buy.com/product/150841.html | 03:37 |
| wolfspraul | ACA? | 03:38 |
| xiangfu | wolfspraul, I already burned a ftg256 chip by using hot air :( | 03:38 |
| wolfspraul | great | 03:38 |
| wolfspraul | speed up your discoveries :-) | 03:38 |
| wolfspraul | buy another 10 of them | 03:38 |
| wolfspraul | that's why we work with a 7 USD chip now, not a 70 USD chip... | 03:39 |
| wolfspraul | you should just try in series right away, different durations etc. | 03:39 |
| wolfspraul | unless the results are extremely obvious (smelly), you probably need 2 or 3 of the better attempts to understand yield issues | 03:40 |
| wolfspraul | anyway, have fun! :-) | 03:40 |
| wolfspraul | a little Sunday cooking | 03:40 |
| wolfspraul | chip brunch | 03:40 |
| wolfspraul | watch the fumes btw | 03:40 |
| wolfspraul | if possible do this near a big and open window, ideally outside but maybe that's difficult | 03:41 |
| xiangfu | I will try to use the broken chip first. | 03:41 |
| wolfspraul | especially when you burn a lot you may create a lot of fumes | 03:41 |
| xiangfu | today is cold. I goto one small room. close the door. open the window. :) | 03:41 |
| wolfspraul | yes good | 03:42 |
| wolfspraul | those fumes can be really bad | 03:42 |
| xiangfu | yes. | 03:42 |
| wolfspraul | especially if you do one experiment after another, and day after day | 03:42 |
| xiangfu | yes. | 03:42 |
| wolfspraul | I just looked out my window... made me thinking whether OPENING the window is such a good idea :-) | 03:43 |
| wolfspraul | but ok... | 03:43 |
| wolfspraul | putting the typical Beijing air aside, it's still true :-) | 03:43 |
| xiangfu | :-) | 03:43 |
| xiangfu | be back in 10 mins. try my first chip-cooking. :) | 03:43 |
| wolfspraul | good luck | 03:43 |
| xiangfu | totally failed. http://downloads.openmobilefree.net/tmp/IMG_1393.JPG | 04:08 |
| xiangfu | I use hot air on left one. | 04:09 |
| xiangfu | use oven on right one. | 04:09 |
| wolfspraul | he | 04:12 |
| wolfspraul | failed = the soldering didn't happen? | 04:13 |
| wolfspraul | the right one looks nicely toasted though :-) | 04:13 |
| wolfspraul | buy more chips and boards, maybe 20 or so | 04:13 |
| xiangfu | (failed = the soldering didn't happen?) yes. | 04:13 |
| xiangfu | I guess I miss up the temperature. | 04:14 |
| wolfspraul | yes | 04:14 |
| wolfspraul | not sure how accurate the setting is, or to reach/hold it etc. | 04:15 |
| wolfspraul | difficult | 04:15 |
| wolfspraul | a precise measuring is also difficult | 04:15 |
| wolfspraul | too bad you can't just program the whole world | 04:15 |
| wolfspraul | room.set_temp(25.00) | 04:15 |
| wolfspraul | maybe too warm | 04:16 |
| wolfspraul | room.set_temp(19.50) | 04:16 |
| xiangfu | :) | 04:16 |
| wolfspraul | well | 04:16 |
| wolfspraul | of course a chinese heating computer would immediately and happily return with "19.5000 exactly reached, have a nice day" | 04:17 |
| xiangfu | exactly. | 04:22 |
| xiangfu | I made 10 of those PCB. in parallel. I try to improve the pcb design. I found it's really not easy route more wires in 2 layers. | 04:22 |
| xiangfu | the youtube or other example videos/documents are all simple chips. simple design. | 04:23 |
| xiangfu | the azonenberg | 04:23 |
| xiangfu | the azonenberg's board is good reference for learning. | 04:24 |
| xiangfu | bbl, lunch time. | 04:24 |
| wolfspraul | enjoy | 04:52 |
| wolfspraul | imagine those would be sample chips from our own chip design | 04:52 |
| wolfspraul | let's say we got 10, after waiting for 2 months | 04:53 |
| wolfspraul | that would be painful to burn! :-) | 04:53 |
| wolfspraul | by then we better have figured this out... :-) | 04:53 |
| wpwrak | hmm, that pcb's problems still start a long way before the soldering ... | 08:19 |
| wpwrak | at least with a pcb (as opposed to a ball of wires) it's easier to see what's wrong :) | 08:20 |
| wpwrak | xiangfu: nicely cooked pcb :) | 09:00 |
| wpwrak | the silk screen varies quite a bit with respect to the copper. you'll probably have to find a more precise fab when you get to doing more serious things. | 09:02 |
| wpwrak | about the layout: you can make your life easier by routing ground outside the chip, not under it. e.g., if you make a U-shaped ground trace (include connectors and oscillator), then you need only relatively short traces that go under the chip. this lowers the risk of make unintended contacts. | 09:06 |
| wpwrak | then, you only have one silo cap at the power connector, but no bypass caps at the fpga. the longest distance between ground and cap is about 50 mm, which is a bit excessive. | 09:07 |
| wpwrak | the number of connections to power is also suspicious. there should be more of them. | 09:11 |
| wpwrak | and another one: P1 ground should connect to P2 ground via a trace, not through the chip. | 09:12 |
| wpwrak | the aspect ratio (hole to ring diameter) of the holes for the connectors seems quite small, especially if we consider that your fab is not very precise. maybe make it a bit bigger. | 09:14 |
| LunaVorax | Hello! | 10:44 |
| Intro | hi | 10:48 |
| Intro | hi gargom | 10:49 |
| gargon | intro: hi | 10:50 |
| gargon | if someone interested, I found an interesting article about chip/pcb cooking with a toaster like appliance: http://avr.tavir.hu/modules.php?name=Content&pa=showpage&pid=118 | 10:51 |
| wpwrak | i like the triac :) the rest of the circuit seems over-engineered. | 12:31 |
| wpwrak | why not use a cheap AVR with V-USB. that solves: 1) power supply, 2) controls, 3) indicators, 4) communication with modern hosts. | 12:32 |
| wpwrak | of course, you need some confidence in having gotten the galvanic separation right. the price of failure is a PC :) | 12:34 |
| roh | wpwrak: the moc series is nice.. optoisolated hv-switches | 14:24 |
| wpwrak | just integration of the "large" triac is missing | 14:26 |
| kristianpaul | triac are live saving, even more if you add an optocoupler | 14:35 |
| roh | wpwrak: the moc is the driver as far as i can see.. but there are solid-state-switches which have all that integrated | 14:36 |
| roh | zero-crossing, optoisolated line switches | 14:36 |
| wpwrak | do you know one ? | 14:36 |
| wpwrak | the problem with external components is that anything that's rated for mains voltage is bulky, often just to obtain a large enough air gap. when integrated, such issues disappear. | 14:38 |
| roh | http://www.pollin.de/shop/dt/OTI1OTU2OTk-/Bauelemente_Bauteile/Mechanische_Bauelemente/Relais_Zugmagnete/Solid_State_Relais_XSSR_DA2420.html | 14:38 |
| roh | just an example. there are more with different ratings | 14:38 |
| roh | you can directly drive those from the avr, and connect to ac ;) | 14:39 |
| wpwrak | ah, a module. i was hoping for a chip :) | 14:39 |
| roh | nope... there are some chips, maybe even in the moc series with integrated triac | 14:39 |
| roh | i guess these sil solid state relays are more what you want | 14:40 |
| wpwrak | at that size, the module should also supply power to the outside :) | 14:40 |
| roh | its the same case size for 40A | 14:41 |
| roh | mechanics matter when it comes to power rating ;) | 14:41 |
| roh | https://www.sparkfun.com/products/10636 | 14:41 |
| roh | well.. there are readymade kits for driving a toasteroven or similar appliances to defined temperatures ;) | 14:43 |
| wpwrak | the one at sparkfun looks much nicer :) let's see if digi-key have something like that .. | 14:45 |
| wpwrak | hmm, all quite expensive | 14:51 |
| roh | digikey is expensive | 14:52 |
| wpwrak | but factor of 6 ? the cheapest that can handle >= 400 V and >= 8 A are USD 24.76 @ 100. (these are actually 530 V 25 A, but the 8 A models are even more expensive) | 14:54 |
| wpwrak | well, this one is still reasonable, though also more expensive (800V 5A, USD 6.05 @ 100): http://www.clare.com/home/pdfs.nsf/www/CPC1998.pdf/$file/CPC1998.pdf | 14:57 |
| roh | well.. i dont buy at dk.. dunno whats their issue | 15:03 |
| wpwrak | ah. they also have the sharp part. but listed as "120 V". price similar/better to/than sparkfun: 5.10 @ 1 vs. 4.95, 4.08 @ 10 vs. 4.46, 3.37 @ 100 vs. 3.96 | 15:13 |
| wpwrak | now that makes a lot of sense. alas, the "240 V" part (S208T02) is "call for price" | 15:14 |
| roh | wpwrak: yeah. i guess there are some dozend of similar devices | 17:56 |
| wpwrak | yeah, seems that i'm not the only one who likes their circuits to be compact :) | 18:24 |
| qi-bot | The build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.minimal-20121020-1228 | 21:04 |
| --- Mon Oct 22 2012 | 00:00 | |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!
