06:08 <ThunderChicken> So in a wondrous lapse of luck, happenstance, and awareness a tiny drip of solder found its way onto the PMIC chip across pins 3 to 6 though mainly across the battery and bat sense pins(BBB from like....2013?). Having discovered this after plugging a battery into the header (populated minimally but properly from what I read (ie with resistor and

06:08 <ThunderChicken> such)) and the wires/pins/pcb attempting to go nuclear, I further discovered the board would not respond at all when powered from USB or barrel plug. These attempts were made before discovering the shorted pins. I have taken solder braid and flux to the side of the chip in hopes of clearing any bridging from solder that may have flowed under the

06:08 <ThunderChicken> package though I cant say for sure I got it all. My next step is to remove the LAN and barrel plugs to get a better look at it, then just remove the chip all together and either reflow it back onto the board after some pad cleanup, or order a replacement. There arent any obvious signs of high currents down stream (no blown components or discolored

06:08 <ThunderChicken> joints/packages), and the datasheet for that chip says it has some pretty spiffy safety features (which I realize would not prevent 4.2v from a fully charged lipo from energizing whatever rail it touched, though it was dead shorting to itself/gnd, so I imagine there wasnt much preference for flowing down stream). Finally to the question: Anyone

06:08 <ThunderChicken> care to give-me-some/crush-all hope that the rest of the board beat the odds and could be okay once any bridge fault is removed or PMIC is replaced? Is there any way I could test the rest of the circuits for 'not being blown' without having a PMIC on the board?

06:49 <zmatt> ThunderChicken: well a short between bat and bat-sense wouldn't cause any problems, they're *supposed* to be shorted together (albeit preferably as close to the battery as possible)

06:49 <zmatt> while a short from bat to ground will mostly be unpleasant for the battery, not for the bbb

06:50 <zmatt> ohh you shorted it to VLDO1 ?

06:52 <zmatt> if you managed to put 4.2V on VLDO1, you can be pretty sure the AM335x is dead

06:52 <zmatt> since that's a 1.8V supply

06:53 <zmatt> (pin 3 is VLDO1)

07:00 <ThunderChicken> (*reply to initial reply about shorts: so a good chance there is an uncleared bridge is still under it maybe and the IC protections are doing their jobs?) I cant say absolutely, It happened a few years ago and its been in a 'random stuff to mess with' box, finally getting around to actually gearing up to lift the chip. I did the solder removal

07:00 <ThunderChicken> before boxing it, so the pins/pads are all stripped, but IIRC I believe the short was initially over the batt batt sense and possibly pin 7 ('sys' on the datasheet). At least at the time the battery was on it. The solder did flow about some in initially trying to clean it up. There is a solid hope against hope that no dangerous voltage found its

07:00 <ThunderChicken> way into something bad, but the only chance it would have (supposing only the batt pins initially) would be when It would have been powered either through USB or barrel. Which Im hoping means the PMIC detected some f$%kery and is just keeping itself in a deactivated fault state

07:00 <ThunderChicken> Like, I wouldnt be surprised if its fried. But if it were couldnt hurt it trying some stuff before binning it LOL

07:01 <ThunderChicken> its not like i could make it more broken if that were the case

07:02 <zmatt> I can't remember what happens if bat is shorted to batsense but not to an actual battery, I guess it's possible it confuses the pmic

07:03 <zmatt> I don't think that's the case, but I don't remember for sure

07:04 <zmatt> shorting to sys could certainly cause weird things though

07:04 <zmatt> that sounds plausible

07:04 <ThunderChicken> well at the header, in my case, it sparked some and quickly heated the resistor I added (per suggested addition). Nothing else gets hot or anything on 5v and power seems to stop at the PMIC from what ive probed so far and what I can remember

07:05 <ThunderChicken> So, it acted like a damn near dead short between + and - battery pins

07:05 <zmatt> wait it heater the resistor? what?

07:05 <zmatt> no significant current should ever flow through that resistor

07:06 <zmatt> (it's supposed to be an NTC for measuring the battery temperature, using a plain resistor is a trick to say "yeah yeah temperature is fine don't worry")

07:07 <zmatt> nothing you've described explains how you'd get a short to ground either

07:07 <zmatt> so whatever went wrong, it went way horribly more wrong than you could get by shorting bat/batsense/sys

07:07 <ThunderChicken> I believe so. my first hint something very hot under my finger. Maybe it was the pad/traces/vias and im remembering wrong. Yeah....I went with the 'its a reputable battery, I trust it, it has safeties, just work dang it' pain resistor heh

07:08 <ThunderChicken> Im under the assumption there was a path to the gnd pad under the ic. but that is an assumption

07:10 <zmatt> that pad is relatively far away, neighbouring pins are closer

07:11 <ThunderChicken> It was a relativly tiny blob that flew off my iron as some point. the tips were toast, was waiting for new ones, and just kind of beat them along til then. at some point a drip found that exact spot. And it was tiny.....until you look at the pins of that chip LoL  SMD stuff man.....crazy small some times

07:11 <zmatt> so I'm not sure how you'd get enough solder under there to short to ground without shorting a whoooole bunch of pins

07:11 <ThunderChicken> at some point I basically did, jsut not with the battery

07:12 <ThunderChicken> Tried to flow it away, or suck it up with stranded wire, before having braid and flux on hand

07:12 <zmatt> and destroyed the pmic in the process by overheating it? :P

07:13 <ThunderChicken> I didnt power it at all with obvious shorted pins, so there is that.

07:13 <ThunderChicken> haha yep, thats the other option

07:13 <ThunderChicken> XD

07:14 <zmatt> are you sure a single bbb is worth all this effort?

07:14 <ThunderChicken> I didnt pay for it (got it as a gift for designing some haunted house controllers), so Im only really out having to buy another one if I wanted to replace it.

07:15 <ThunderChicken> At ~$80 from what Ive seen, its worth messing with

07:15 <zmatt> depends on how much time you spend on trying to fix it, and how much you value your time

07:16 <ThunderChicken> Like, I dont expect it to come back. It would be a total fists popping up over monitor with a 'yissssssssssss' moment if it did. But, I figure I can learn a few things ruining it more if its toast anyways. So....[shrug]

07:17 <zmatt> anyway, my main worry in all this would be battery voltage making its way onto VLDO1 or some other output supply rail (whether due to a short or due to damaging the pmic), if that happened then it is all over for the board

07:17 <zmatt> if that didn't happen then it sounds plausible enough that replacing the pmic might be eonugh

07:17 <ThunderChicken> Currently, I have time to burn. Going through the random tinkery box is what is keeping the stir crazy at bay

07:18 <ThunderChicken> at the very least itd make a heck of a reddit post, rezzing a 'dead' BBB lol

07:18 <zmatt> some info about the bbb's power supply stuff here: https://elinux.org/BeagleBone_Power_Management

07:19 <zmatt> (I wrote that page a long time ago)

07:19 <zmatt> I don't recommend attempting to use the battery terminals at all, see warning on that page and the thread it links to

07:29 <ThunderChicken> sweeeet! I like digging into technical stuff, and it seems a lot of posts end at 'meh toss it and get another'. Well, the components exist, these are dev boards, some with the express purpose of introducing people to electronics, why not peel it apart find the dead thing and fix it? ya know? idk, maybe Im just that obsessive LoL Seems like fun to

07:29 <ThunderChicken> me though. Then again, the vast majority of people do need to consider time v lost cost, etc.... Still, while the bar has moved there is still some of that wall between the weekend warriors of plugging stuff in together and running tutorial code (and there is NOTHING wrong with that. At all. Ever. Learning is awesome, it doesnt matter how its dont

07:29 <ThunderChicken> or to what length), and the people who are aware of, know how to set efuses in fresh out the box SoC CPUs(per se), and what they affect, at the finest level of granularity. Like, someone could have had built most of these SBC designs (practically all if you have a reflow oven) so its a logical next step to troubleshooting, find the broken thing,

07:29 <ThunderChicken> replace it with a new one, initialize and fist pump XD

07:29 <ThunderChicken> oh?! there is a known issue with the battery stuff I take it?

07:30 <ThunderChicken> I was so excited when I discovered the BBB had onboard self contained power! Heck, I still complain the RPi and most variants dont have battery provisions and/or PoE

07:32 <ThunderChicken> *how its done

07:32 <ThunderChicken> **could have hand built

07:33 <ThunderChicken> Using a k400 wireless kb in my lap, the ergonomics leave much to be desired......

07:43 <zmatt> general-purpose AM335x doesn't have customer-programmable efuses, only some factory efuses (e.g. for the MAC address, which exact AM335x variant the part is, etc)

07:44 <zmatt> typically if people want to battery-power a BBB you just use a 5V usb power pack

07:45 <zmatt> though I'll admit that's not going to win any efficiency-award :P

08:08 <ThunderChicken> Ahh. Im still in the twilight when it comes to some of those bare metal intricacies. I remember back when broadcom wifi chips were avoided like the plague if you wanted to do anything interesting on wifi (cough cough god bless WEP encryption when traveling cough cough). Then finally someone dropped some driver hack or something that effectively

08:08 <ThunderChicken> wedged promiscuous mode ability to the user. Sometimes it worked, sometimes it crashed the system, sometimes it didnt do anything LOL. Now Broadcoms are in dang near everything! Most of them more open than I ever remember them being before.

08:09 <ThunderChicken> true. I mean hats/capes/whatevs exist for UPS self power, and PoE. Itd just be cool to have a board that took those things into consideration with how the whole maker thing took off

08:10 <ThunderChicken> Though that opinion may be colored through wrestling with feeding RPis over long runs on crappy supplies XD

08:13 <ThunderChicken> I still waiting patiently for someone to incorporate QC communications so the board can tell the supply to bump the voltage up on a lossy run. That would be a selling point, IMHO. Of course, simple ideas can get complicated to impliment real quick....still a fun thought

08:14 <zmatt> seems unnecessary complication and cost, the BBB is already pretty crowded with components

08:15 <ThunderChicken> Oh yeah, especially the BBB. Part of the drive to try with my board is how feature packed it is.

08:16 <ThunderChicken> Even these days, its still got some bullet points to throw around

08:16 <zmatt> if you're using a "lossy run" from a "crappy supply", the solution for that isn't to change the board :P

08:16 <ThunderChicken> Shhhhhhhhhh....... ;P

08:16 <ThunderChicken> lol

08:17 <zmatt> besides the BBB tolerates a fair bit of voltage drop as long as you don't need its usb host port

08:17 <zmatt> iirc

08:27 <ThunderChicken> what was the input voltage window on the BBBs? I dont see anything that looks like a normal vreg, but they do come in all sorts of packages and I cant remember what half of these packages do.... heh

08:31 <zmatt> power input is all handled by the pmic

08:32 <ThunderChicken> right, so the Vmax is the PMICs vmax?

08:32 <zmatt> I don't remember the min voltage, but I'm pretty sure I put it on that wiki page I linked earlier, otherwise see pmic documentation

08:32 <ThunderChicken> ah for sure :)

08:32 <zmatt> as far as max voltage you should not exceed 5.25V I think, per USB spec

08:32 <zmatt> protection kicks in a bit higher than that I think

08:33 <zmatt> but in general you should just supply 5V

08:35 <ThunderChicken> itll take 18v and like it! ....say, whats this pretty blue smoke? haha

08:36 <zmatt> in theory the pmic should protect the rest of the bbb from that... I would not be eager to test that theory :P

08:40 <ThunderChicken> well cnsidering how this board ends up could do all sorts of voltage injection 'experiments' hehehe

08:45 <ThunderChicken> naughty electronics will be made an example of XD

19:52 <set_> Hello!

19:52 <set_> I made it back outside of Not-so-Fun Friday.

19:52 <set_> Anyway, presume as you may. Up and away!

19:52 <set_> bbl...