**** BEGIN LOGGING AT Sat May 08 02:59:57 2021 May 08 08:31:37 Hello May 08 08:31:57 Can anyone tell me how to find out gpio numbers for beaglebone ai May 08 08:32:10 say for example gpio3_14 May 08 08:33:00 I tried 3*32 +14 May 08 08:33:22 gpio110, but i cant address gpio3_14 with that May 08 08:37:32 I failed to address onboard led D5 May 08 08:37:53 I was trying to do blinking with that led May 08 11:05:54 Good day, I got a question, I tried to flash the debian eMMC image to my new beaglebone black wirless, but it seems i have done something wrong as when powering it on the power led flashes shortly and then nothing happens May 08 11:08:12 Can anybody help me recover from this? May 08 11:19:37 philmacfly: what happened exactly between the last time your bbbw worked and it just showing the power led flash? May 08 11:20:28 since what you're describing is caused by a damaged SoC, most commonly due to exposing I/O to overvoltage May 08 11:20:49 not really something that can be caused attempting to reflash May 08 11:22:07 zmatt: I flashed the bone-eMMC-flasher-debian-10.4-console-armhf-2020-05-18-1gb.img.xz image to a microsd card put it in the microsd card slot, attached the usb wire into my usb hub, powered on the bbb while holding the s2 button, then it started lighting all 4 leds. after that it started blinking until it lit all 4 leds again, then i powercycled the bbb by disconnecting the usb cable and May 08 11:22:09 reconnecting it May 08 11:22:13 since then I have those symptombs May 08 11:26:21 oh never mind that actually sounds more like your usb hub is failing to supply sufficient power May 08 11:26:31 and is cutting power as result May 08 11:27:10 I felt the same way an now connected a 5v 3A powerbrick but still similar results May 08 11:27:52 indyenthusiast: the on-board leds can't be controlled via the sysfs gpio interface anyway, they're controlled via the sysfs led interface May 08 11:29:38 indyenthusiast: but in general a whole bunch of things are obnoxiously off-by-one on the AM572x SoC, so gpio3_14 is actually 2*32+14 rather than 3*32+14 (because the first gpio controller on the AM572x is gpio1 rather than gpio0) May 08 11:29:47 philmacfly: okay that's really odd May 08 11:30:52 its gets even more interesting: I connected a usb to ttl to the bbb (without connecting 5V or 3V) and then it started doing the flashing routine again May 08 11:32:00 uhhhhhhhhh that sounds really bad May 08 11:32:01 ok now I used yet another power brick and it seems to be booting again now? May 08 11:33:31 maybe the 5v 3a brick is shitty aswell May 08 11:33:50 like, you were talking about a serial cable to the debug console? that makes no sense, it has a buffer to isolate it at power-off May 08 11:34:07 there should be no way for power to backflow onto supply rails via that May 08 11:34:22 (which would otherwise be a great way to destroy hardware) May 08 11:35:14 (mind you, the serial console is pretty much the only pins that have such protection, in general driving signals into an unpowered beaglebone is a recipe for murdering your beaglebone) May 08 11:35:28 did you check the polarity of the power brick's plug? May 08 11:35:53 center-negative (rather than center-positive) can unfortunately sometimes be found in the wild May 08 11:36:13 I think the observation with the serial console was random May 08 11:36:26 sorry wiht power brick I meant a usb power brick May 08 11:36:33 one of my old raspi ones May 08 11:38:24 ah May 08 11:38:53 a bad usb cable can also cause all sorts of funky problems May 08 11:38:59 so I have no clue anymore what happens here, now I power cycled it again and now the power blinks constantly May 08 11:39:46 did your bbbw maybe get shipped with ghosts? maybe try an exorcism May 08 11:40:38 yeah I think rn thats the only possibility May 08 11:42:11 I hate it when that happens May 08 11:43:22 "Dear farnell please send a new bbb unit as this seems to habe ghosts" May 08 11:43:56 but yeah I think it was the power supply, afte reflashing the microsd it seems to boot a ok again May 08 11:46:08 inconsistent results from a power supply might be something like an iffy connector or a cable with a broken conductor May 08 11:46:20 that sometimes makes okay contact, and sometimes does not May 08 11:49:41 lesson learned id say May 08 12:19:52 Thank you very much @Zmatt. This one worked. May 08 12:21:23 oh right rcn sabotaged gpio conflict checking so I guess you actually can control the leds via sysfs gpio now instead of using the sysfs led interface... though you probably shouldn't May 08 12:21:41 it's not something that's *supposed* to work, it's more of an accident May 08 12:27:34 Yes .. I am new to elinux and struggling to figure out ways for a week or so.. After a long time some thing worked. But yes.. This didn't work completely. The LED turns off itself after a while (like 5 sec or so). Does this ring something to you... ? I will try it with sysfs Led too now. May 08 12:28:26 most likely the led is configured as some form of activity-indicator May 08 12:28:50 That doesn't even turn ON once otherwise.. May 08 12:29:05 check its the "trigger" property May 08 12:29:13 *its "trigger" property May 08 12:29:17 Ok May 08 12:30:00 like /sys/class/leds/cant-remember-name-something-with-usr3-probably/trigger May 08 12:30:46 That's exactly something I am wondering .. The name says beaglebone:green:usr3 May 08 12:30:52 which obnoxiously doesn't print the trigger but the list of triggers with the selected one enclosed in [] May 08 12:31:11 "green" May 08 12:31:16 I have no clue how that got approved as a kernel-userspace api May 08 12:31:24 yeah lol that's for historical reasons May 08 12:31:30 the very first beaglebone had green leds May 08 12:31:47 and ever since the led devices have kept the original names for backwards compatibility May 08 12:31:53 even though they're blue nowadays May 08 12:32:04 That was confusing with the beaglebone green May 08 12:32:15 which is whole different hardware May 08 12:32:29 yeah that's not the reason May 08 12:33:33 anyway, dunno if the defaults are different on the BBAI, but on the BBB usr0 is configured as mmc1 (i.e. eMMC activity) by default May 08 12:34:51 regardless, for manual control you want to echo "none" into the trigger attribute May 08 12:37:24 Ohh.. ok will look more into that May 08 12:58:57 @Zmatt just to confirm.. is there anything else I have to do other than echo 1 > brightnesss (imagining I am adressing correct usr number)? May 08 13:00:24 iirc with trigger set to none, you can indeed control the led through the brightness property May 08 13:00:39 it might not be 0/1 though, check max_brightness May 08 13:00:43 use that value May 08 13:00:56 ok May 08 13:01:10 (0/1 would be *logical*, but they might in fact not be logical) May 08 13:03:30 Yes.. You are correct . It is more like pwm based brightness control May 08 13:03:34 max value 255 May 08 16:09:15 got one other small question: Whats the difference between the debian console and the iot image May 08 16:35:09 the console is a minimalistic image intended for people comfortable with debian linux and just installing whatever stuff they want themselves May 08 16:35:50 the iot image is the default image that basically includes everything and the kitchen sink (apart from a desktop environment) May 09 02:26:59 Saturday, fun day! May 09 02:39:09 Am I right? May 09 02:39:20 GenTooMan: Wake up! May 09 02:39:46 I am cookin' and makin' funny articles for people to brush the surface! May 09 02:43:01 I will study more when I get a hold of the correct info. I am lookin' to attain. May 09 02:43:14 Until then, surface brushin'. May 09 02:44:19 I just read a good chapter on ESCs and motors from, "Motors for Makers." May 09 02:45:16 It is a neat book. Not too in depth but yields some interesting facts and neat ideas on specific motor usage, 3.3v logic, and 5v boards to use for specific instances. May 09 02:45:39 Actually, there was a section that put my math skills to work. May 09 02:46:47 It was a brief w/ lots of formulas and equations one could receive when working w/ specific hardware. May 09 02:47:20 Anyway, chattin' to myself here. Break time! **** ENDING LOGGING AT Sun May 09 03:00:50 2021