**** BEGIN LOGGING AT Sat Jan 12 02:59:59 2013 Jan 12 04:04:05 if i have a beaglebone that boots fine, but isn't recognizable on Windows/osx/linux, any suggestions on next steps? Jan 12 04:04:20 tail of dmesg shows buffer I/O errors Jan 12 04:29:22 next steps? whoa you are skipping exact steps first Jan 12 04:35:40 can you point me at exact steps. i'm obviously missing something dumb. Jan 12 04:36:11 it's an A6 board, so (from reading "getting started page") i don't think i need the drivers (at least for osx/linux) Jan 12 04:37:43 under windows, it sees an unformatted removable drive. under osx, it sees nothing. under linux, i can see the connection in dmesg Jan 12 04:38:00 then drops when it gets the i/o error Jan 12 04:40:21 pastebin the messages Jan 12 04:42:30 http://pastebin.com/0hfXk9Gk Jan 12 04:46:10 fwiw, i've got a DVI cape on it, and it boots ok if i just plug it into a monitor Jan 12 05:12:42 metalbot: buy a more expensive SD card :-D Jan 12 05:14:01 oh actually thinking of the wrong thing! Jan 12 05:14:36 it just seems weird that it would boot ok, but not operate in storage mode Jan 12 05:15:04 previous owner never hooked it to another machine - he was just using directly on a monitor Jan 12 05:57:07 how do you configure the serial parameters (baud rate, etc.) under linux for the serial console of a beagleboard XM? Jan 12 05:57:37 i've tried "setserial -G /dev/ttyUSB0" but i get "Cannot get serial info: Invalid argument" Jan 12 05:57:51 correction: Jan 12 05:58:02 not. no, that's correct. Jan 12 06:08:17 Bus 004 Device 003: ID 06cd:0121 Keyspan USA-19hs serial adapter Jan 12 06:08:32 is this device incompatible with the default linux usb-serial driver? Jan 12 06:08:42 3.4.11-2.16-desktop Jan 12 06:08:47 (openSUSE 12.2) Jan 12 06:36:53 HI every 1 Jan 12 06:37:38 is any body there Jan 12 06:38:05 nope. we're all gone Jan 12 06:38:18 where dear Jan 12 06:38:43 then y that list seems to be fill Jan 12 06:39:00 i need some help Jan 12 06:39:23 emeb_mac: in some sense we are always here :P Jan 12 06:39:38 PLZ HELP MEEE Jan 12 06:40:52 helo Jan 12 06:41:17 ask your question. stop begging. Jan 12 06:42:18 where can i get the accessories of BeagleBoard-xM Jan 12 06:42:18 i'm getting a brew Jan 12 06:42:30 er have a credit card? Jan 12 06:42:44 nope Jan 12 06:43:14 give me the farnell code of that Jan 12 06:44:44 Could any one give me the links where i can buy accessories of BeagleBoard-xM Jan 12 06:44:52 what kind of accessories do you want? Jan 12 06:45:47 USB Mini-A to USB A Female OTG Cable Jan 12 06:46:14 what's the default serial port parameters for the beagle xm? Jan 12 06:46:24 CABLE ASSEM HDMI-DVI 19POS Jan 12 06:46:31 https://specialcomp.com/beagleboard/index.htm Jan 12 06:46:40 i've searched and searched the manual but no joy. Jan 12 06:47:57 emeb i already have it but my company need only farnell code for that Jan 12 06:48:43 like for Beagle XM the code is 1823269 Jan 12 06:49:07 also in that no accessories included Jan 12 06:49:09 emeb_mac: have fun :) Jan 12 06:49:16 heh Jan 12 06:49:59 mranostay: you don't want to stick around and help fill the empty aching void? Jan 12 06:51:27 got it Jan 12 06:59:25 it's working now with minicom. Jan 12 06:59:27 cool. Jan 12 07:14:12 is there a how-to showing how to rebuild the angstrom/the kernel from a standard x86_64 desktop machine (openSUSE 12.2)? Jan 12 07:14:38 and reflashing the miniSD? Jan 12 11:49:50 Hello every one! i have a Beagle bone A6 . The tutorial http://beagleboard.org/static/beaglebone/latest/README.htm explain that i don't have to install drivers but when i try to put the LED USR3 ON with Cloud9 that doesn't work... (I work on Ubuntu 12.10 and just begin with this card^^) Jan 12 12:06:12 hey! My computer shutdown. As I said, I have a probleme on cloud 9. I'm trying the test code to make the led bliking,but the led d'ont blink on the card. (I am on Ubuntu 12.10 and I just begin with this card) Jan 12 12:43:19 Yann_: make sure you aren't running it in debug mode... it runs just fine Jan 12 13:14:51 lambert: i made it but that doesn't work... no error , the console (in cloud9IDE) said "true" Jan 12 14:12:56 hi someone can help me on g_printer module? Jan 12 19:21:55 * _av500_ does not print Jan 12 21:58:51 mranostay, happy Sat Jan 12 22:20:46 koen: i heard you are involved in gathering patches for the angstrom bbxm builds - true? Jan 12 22:21:11 what's the chances of getting a 3.5.x kernel up so we can use the bluez 5.1 stack? Jan 12 22:21:50 s/what's/what are/ Jan 12 22:22:48 koen: ? Jan 12 23:18:03 howdy Jan 12 23:51:16 how accurate are the ADC pins on the beaglebone, using a 10k thermistor I'm getting a temperature that's about 17F too high Jan 12 23:51:25 thermistor gives a valid reading on a teensy Jan 12 23:52:06 trelane: accuracy is going to based on the range and the bit depth of the adc Jan 12 23:52:34 I'm wondering if there might be some resistance that I'm not accounting for? Jan 12 23:53:31 prpplague: hey purple Jan 12 23:53:33 my question therefore isn't about bit depth, which shouldn't matter in this formula much, it's giving me an answer that appears to be wrong Jan 12 23:53:56 mranostay: greetings Jan 12 23:54:09 trelane: bit depth always counts Jan 12 23:54:13 the AIN pins are floating Jan 12 23:54:47 travalas: if the teensy is using 10bit and the beagle is configure for 8-bit, each values represents a bigger "block" Jan 12 23:54:54 travalas: or "range" Jan 12 23:55:34 prpplague: was apartment shopping in MV today... fun prices :) Jan 12 23:55:44 mranostay: ugh, don't tell me Jan 12 23:55:54 mranostay: i am already nervous Jan 12 23:56:40 right and I'm doing simple math on a voltage divider and the answer I'm getting is wrong. the 10k thermistor using a voltage divider and some math is giving me a measurement of 7731 ohms, if I stick it on a multimeter I'm getting 9963 ohms. I've checked the 10k thermistor it's accurate to 1% Jan 12 23:56:59 prpplague: working on my poker face when the one was $3k for a 1 bedroom :) Jan 12 23:57:43 travalas: sounds like you are missing something then.... and/or leaving out some information Jan 12 23:57:54 mranostay: holy cow Jan 12 23:58:47 prpplague: yeah needless to say that one is a little of my price range :) Jan 12 23:59:39 mranostay: i hear the dumpsters behind most chillies are nice and they have free continetal breakfast Jan 12 23:59:48 are you sure you have the right voltage level? Jan 13 00:01:24 Russ: that was my next question Jan 13 00:01:32 * prpplague hands the reins over to Russ Jan 13 00:01:39 1.8V, 3.3V, 5V Jan 13 00:02:10 2.5V Jan 13 00:02:20 Russ I'm running 1.8 from the header to the thermistor, which shouldn't care about voltage Jan 13 00:02:45 have you measured and make sure it's 1.8V? Jan 13 00:02:58 it shouldn't matter if all I'm doing is comparing two resistors Jan 13 00:03:34 there is a 1.8v from the header? Jan 13 00:03:41 mranostay, yes :) Jan 13 00:03:49 1.75ish actually but it's there Jan 13 00:04:11 and what's vdd_adc? Jan 13 00:04:17 (connected to vrefp, vrefn) Jan 13 00:04:54 (pretty sure it's also 1.8V) Jan 13 00:05:16 although, really, I'd use vdc_adc on the expansion header rather than just some 1.8 digital source Jan 13 00:05:24 Russ: all I'm doing is (40960000/ain2)-10000 Jan 13 00:05:55 its just if you are connecting your thermistor to 1.8V digital, you could be getting noise Jan 13 00:06:06 the actual number seems stable Jan 13 00:06:44 well it'd be noise of a higher frequency than the sampling rate Jan 13 00:06:46 which is why I'm running the 10k thermistor to ground and a 10k resistor from vdc_adc Jan 13 00:07:03 ok, so you are connecting to vdc_adc rather than just 1.8v? Jan 13 00:08:24 right Jan 13 00:09:44 vddadc ---- 10k resistor --|-- 10k thermistor ---- ground Jan 13 00:10:03 with the ain2 pin at the | Jan 13 00:13:13 * Russ gets confused looking at the trm trying to figure out how ain2 is actually read Jan 13 00:13:38 ah, there it is FIFO0DATA, FIFO1DATA Jan 13 00:14:35 ok, so lets see, 12 bits of precision, with 0x000 being vrefn and 0xfff being vrefp? Jan 13 00:15:08 I'm not getting anywhere near that advanced but yes it's 12 bits, all I'm doing is directly reading the file in /proc associated with that pin which gives me a number between 0 and 4096 Jan 13 00:15:18 behaves basically like an arduino but more precise Jan 13 00:16:43 seeing gnd_adc Jan 13 00:17:21 isn't there some natural log or something involved in the calculation? Jan 13 00:18:45 or are you just concerned with getting R right now? Jan 13 00:19:39 I'd like to get R but yes I'll be taking natural log later Jan 13 00:19:47 I'm trying to get a number that approximates what I'm getting on my multimeter Jan 13 00:20:08 I have the same formula working on a teensy2 Jan 13 00:20:14 sitting next to the bone Jan 13 00:21:48 once I get a validly measured resistance I just throw that in Steinhart Hart and I get a temperature in Kelvin Jan 13 00:22:05 it bothers me that this is off but only just a little (making me think that there is additional resistance on that pin Jan 13 00:23:06 so the number you are reading is? Jan 13 00:23:28 2210 presently Jan 13 00:24:07 what are the chances of getting a 3.5.x kernel up so we can use the bluez 5.1 stack? Jan 13 00:25:49 Russ: that gets me a reading of 28C or 83F, I have a termometer from Oregon Scientific sitting less than an inch from the termistor giving me 71.6, and trust me it's not 83 in here Jan 13 00:28:34 yates: if you're using open embedded and have a development tree building a kernel isn't too horribly difficult Jan 13 00:28:47 open embedded also murders innocent baby penguins in their sleep Jan 13 00:28:55 and do you have a multimeter you are reading ain2 with? Jan 13 00:29:13 no but I can measure htat Jan 13 00:30:12 could you? Jan 13 00:30:36 .961v between Jan 13 00:31:12 1.8V*2210/4096 = .971V Jan 13 00:31:29 right Jan 13 00:32:18 which means that the measurements I'm getting from the ADC pin are wrong Jan 13 00:32:41 I would bet that if I reverse that and solve for .961V Jan 13 00:32:42 .961v vs 0.971v? Jan 13 00:32:45 I would get the correct temperature Jan 13 00:33:04 right Jan 13 00:33:24 that's off by less than 1% Jan 13 00:33:47 I'm only off by about 5 degrees celsius Jan 13 00:35:19 I'm calculating 11.7k for Rt Jan 13 00:36:05 I don't think (40960000/ain2)-10000 is right Jan 13 00:36:27 10k/(1-ain2/4096)? Jan 13 00:36:44 that is working on the teensy Jan 13 00:37:03 er, 10k/(1-ain2/4096) - 10k Jan 13 00:37:06 hold on Jan 13 00:37:09 http://playground.arduino.cc/ComponentLib/Thermistor2 Jan 13 00:38:47 I = (vadc-ain2)/10k Jan 13 00:39:16 Rt=ain2*10k/(vadc-ain2) Jan 13 00:39:48 Rt=10k/(ain2/vadc - 1) Jan 13 00:40:09 er, those should be Vain2 Jan 13 00:40:32 Vain2=vadc*ain2/4096 Jan 13 00:40:53 Rt=10k/(ain2/4096-1) Jan 13 00:41:20 gah, I suck at math on irc Jan 13 00:41:29 yeah it's hard to represent math on one line Jan 13 00:43:40 try Jan 13 00:43:45 I just measured it at 11.5kohm Jan 13 00:43:47 Rt=10k/(4096/ain2-1) Jan 13 00:44:27 11717 Jan 13 00:44:58 their circuit is not the same as your circuit Jan 13 00:45:24 vddadc ---- 10k resistor --|-- 10k thermistor ---- ground Jan 13 00:45:29 [Ground] ---- [10k-Resister] -------|------- [Thermistor] ---- [+5v] Jan 13 00:46:25 and with that, I think I'll take the dogs for a walk Jan 13 00:46:57 and we have 70.55F Jan 13 00:46:59 you win! Jan 13 00:47:00 thanks! Jan 13 00:50:31 Umm, quick question: Is there a "common" diode/level translator/something else way to protect the GPIO pins on the BeagleBoard? My friend has already fried a BeagleBoard and we really don't want to do it again :( Jan 13 00:53:46 depends on the circuit all pins are 3.3v except the adc pins which are 1.8v Jan 13 00:54:05 I always test anything I'm about to plug into the board on a multimeter first Jan 13 00:54:10 haven't fried a board yet Jan 13 00:57:09 Yeah, I was reading up here: https://groups.google.com/forum/?fromgroups=#!topic/beagleboard/aOm8_AMjDbc but there wasn't a single consensus Jan 13 00:57:28 I'm worried about the 3.3 V GPIO pins, last time we shorted a 5 V's output with one of GPIO pins Jan 13 00:57:52 10 minutes before our project was to be demoed... Jan 13 01:11:19 WebcamWonder, don't do that and it won't break :) Jan 13 01:12:19 I try to avoid having the 3.3V pins exposed to the project voltage by having them switch a FET isntead of whatever the real switch is Jan 13 01:12:24 that also allows 5v switching logic Jan 13 01:12:40 I would just buy some level translators off sparkfun Jan 13 01:12:53 or trelane's idea Jan 13 01:12:58 https://www.sparkfun.com/products/10213 Jan 13 01:13:12 that will switch at 3.3v, and protect your GPIO from other voltage Jan 13 01:13:13 if you are really paranoid, you can buy opto-isolators Jan 13 01:13:35 if you're that paranoid you're probably already on that doomsday bunker show :) Jan 13 01:14:27 WebcamWonder, if you want to use mosfets, this is fairly useful though it's for an arduino everything is pretty similar: http://bildr.org/2012/03/rfp30n06le-arduino/ Jan 13 01:15:10 also the 3.3v pins are "weak" compared to ardino or other, it's really not a bad idea to use the fets anyway as they don't require much current Jan 13 01:15:38 please make careful note of the diode on the motor :){ Jan 13 01:15:40 er, :) Jan 13 01:18:58 indeed! or just throw a HUGE CHUNKY RELAY after the mosfet :) Jan 13 01:19:20 * mranostay yawns Jan 13 01:19:23 morning Jan 13 01:21:48 Russ: ages ago a friend of mine pointed me to these http://www.mpja.com/10A-480VAC-Solid-State-Relay/productinfo/17154+RL/. If I absolutely do not trust something I stick that after the mosfet Jan 13 01:24:52 Sweet, thanks a lot for the choices guys! :) Jan 13 01:26:00 Would a simple zener also work? (hardware noob here:( ), was looking at this: http://www.thebox.myzen.co.uk/Raspberry/Breakout.html Jan 13 01:26:18 And it seems to use a zener and a resistor Jan 13 01:27:14 WebcamWonder, no worries. Jan 13 01:27:14 fets are cheap in bulk and if you need current to push something VCC will give you much more than the 3.3v GPIO Jan 13 01:28:04 probably but that requires math and my brain is broken now :) Jan 13 01:28:13 also diodes require current/voltage to work Jan 13 01:28:16 so you have to factor that in Jan 13 01:30:30 Given that we don't have anything really complicated connected to our beagleboard, I think I'll go pick up some fets then :) Jan 13 01:51:04 trelane: i can build _a_ linux kernel, but can i build a >=3.5.x linux kernel? Jan 13 01:52:21 and would you suggest using the openembedded tools directly instead of the angstrom build? Jan 13 01:53:05 i'm using debian Jan 13 01:53:24 and its worked out to be pretty easy.. Jan 13 01:53:55 mrpackethead_: what kernel version have you gotten to work? Jan 13 01:54:22 i'm just not sure what the advantage of the angstrom / oe thing is. Jan 13 01:54:44 just a sec Jan 13 01:54:59 can you customize debian so that, e.g., you don't build in the x11 or desktop environment stuff? Jan 13 01:55:11 x11 server, thatis. Jan 13 01:55:18 3.2.33-psp26 Jan 13 01:55:24 pretty decent. Jan 13 01:55:41 still short of the 3.5.x that bluez 5.x requires, though... Jan 13 01:55:56 yes, i don't have any X / desktop stuff installed Jan 13 01:56:00 mrpackethead_: is that a rev 3.1.0 board? Jan 13 01:56:14 i have some v5 and v6 boards Jan 13 01:56:19 they both work without issue Jan 13 01:56:53 I went down the debian path, mainly because it was what i was familar with Jan 13 01:57:25 i seem to get the impression though that some people think that the angstrom system is better, but i've not got any idea why Jan 13 01:58:31 yeah, me too. Jan 13 02:01:28 mrpackethead_: does your build include the apt package manager so that you can install stuff from their repos? Jan 13 02:03:25 mrpackethead_: do you have a good pointer on how to build debian? i'm getting too much variety in my google hits Jan 13 02:14:01 yates: yes, apt- works Jan 13 02:14:30 i use a net-installer Jan 13 02:14:57 do you native build or cross-build? Jan 13 02:15:52 native Jan 13 02:16:06 but thats jsut so far Jan 13 02:16:53 https://github.com/RobertCNelson/netinstall Jan 13 02:17:00 grab this netinstall Jan 13 02:17:31 run on a *nix machine of some kind, and create the bootloader on the sdcard Jan 13 02:17:43 pop your sdcard into the beagle, connect to the network Jan 13 02:17:48 and follow your nose Jan 13 02:17:51 its pretty simple. Jan 13 02:26:57 holy spamming batman Jan 13 02:42:43 mranostay: ok, thanks for the pointers! Jan 13 02:44:24 what did i do now? Jan 13 02:44:38 doh! Jan 13 02:44:49 i mean mrpackethead_: thanks for the pointers! Jan 13 02:44:57 * yates bangs head on wall **** ENDING LOGGING AT Sun Jan 13 02:59:58 2013