| vagrantc | with new firmware and/or protected battery boards? | 00:00 |
|---|---|---|
| noam | New firmware, old (unprotected) battery board | 00:00 |
| noam | also, uh. I think I found the root of the issue | 00:00 |
| noam | One of the cells lists a negative voltage. | 00:01 |
| noam | No, I didn't have the terminals backwards | 00:01 |
| ex-parrot | there is an issue even with the newer LPC firmware where it'll continue to draw current when "off" | 00:01 |
| ex-parrot | but 30mA seems higher than I'd expect for that | 00:01 |
| vagrantc | mine drains about 3-4% per week ... starting at 100% | 00:02 |
| noam | ...oh. If I'd known that, I probably wouldn't have updated the firmware TBH | 00:02 |
| ex-parrot | it's an order of magnitude better on the new firmware | 00:02 |
| noam | And I _definitely_ wouldn't have been so cavalier about leaving it at 0% for a few hours | 00:02 |
| ex-parrot | won't kill the batteries in a week or two like it would previously | 00:02 |
| vagrantc | well, the old firmware drained about 5-10% per day, so ... | 00:02 |
| ex-parrot | the protected battery boards are a good upgrade imo | 00:03 |
| ex-parrot | I just did mine ~ a week ago | 00:03 |
| noam | but yeah, pretty sure the cell with a negative voltage is the issue here | 00:03 |
| ex-parrot | yeah that won't be helping :( | 00:03 |
| noam | It's actually getting less negative as I leave it on the multimeter O_o | 00:03 |
| ex-parrot | I'm also on my second set of cells if it makes you feel better :P | 00:03 |
| noam | it does not :/ | 00:04 |
| ex-parrot | fair | 00:04 |
| ex-parrot | nobody sells the right cells here, I ended up having to import some protected cells from China and de-protecting them | 00:04 |
| noam | I, uh | 00:05 |
| noam | Is there a way to try to fix this cell? O_o | 00:05 |
| ex-parrot | you can try and slowly charge with e.g. a bench PSU | 00:05 |
| ex-parrot | but the harder over discharged they've been the less likely they are to come back | 00:05 |
| noam | At what, 3V? | 00:05 |
| noam | Only one of them is *extremely* bad | 00:06 |
| ex-parrot | and they will tend to be less reliable after that and may even not be entirely safe to use :/ | 00:06 |
| noam | oof | 00:06 |
| noam | Would the protected board have prevented this? :/ | 00:06 |
| ex-parrot | I'd just try doing a constant current trickle charge at hundred milliamps or so | 00:06 |
| ex-parrot | yes the protected board prevents this | 00:06 |
| ex-parrot | I had one cell go right down to zero and it was a dead short more or less after that | 00:06 |
| ex-parrot | the protected board prevents any given cell from under or overcharging | 00:07 |
| noam | okay yeah buying that. | 00:07 |
| ex-parrot | _b it's worth doing | 00:07 |
| noam | I don't know if I _can_ apply a constant current charge with anything I have lying around :/ | 00:07 |
| vagrantc | lifepo4 are abuse resistant... but have limits and probably some quality controll issues now and then | 00:07 |
| ex-parrot | you could probably use a basic PSU with a resistor if you just experiment with the resistor values | 00:08 |
| ex-parrot | this is all kind of sketchy and may not be entirely safe tho, as I feel I should repeat | 00:08 |
| ex-parrot | so keep a close eye on it | 00:08 |
| noam | Yeah | 00:08 |
| noam | Can I run the Reform with no batteries, safely? >_> | 00:09 |
| noam | Might be easiest to just _only_ power it off the external battery until new cells arrive | 00:09 |
| vagrantc | there are instructions in the manual | 00:09 |
| noam | thanks | 00:09 |
| ex-parrot | ^ yeah if you follow the instructions you should be OK, the main thing to avoid is running it with some cells dead or missing | 00:09 |
| vagrantc | lifepo4 have a pretty bizarre charge curve ... basically it hovers at 3.2v for the vast majority of it's capacity | 00:10 |
| noam | Not a single one of mine is above 1V right now :( | 00:10 |
| vagrantc | once you are below 3.2v ... it is nearly dead kinda sorta | 00:10 |
| noam | yeah | 00:10 |
| ex-parrot | below 1v is pretttty dead territory | 00:11 |
| noam | Is there any where to buy the board from in the US? | 00:11 |
| noam | Or is the only option to order it from the MNT shop? | 00:11 |
| ex-parrot | I ended up just leaving my reform plugged in for a year while I waited for the protected boards | 00:11 |
| vagrantc | my solar battery got as low as 0.9v ... seems to have partially recovered ... but it has been the first to drain and the first to reach max voltage for some time now... | 00:12 |
| abortretryfail | noam: if you have a stand-alone LFP charger you can charge them in that just so the cells don't get damaged | 00:12 |
| noam | I don't have one, I'll order one rn :/ It'll come in handy anyways probably | 00:12 |
| ex-parrot | def a nice thing to have | 00:13 |
| ex-parrot | just make sure it can definitely handle LiFePO4 | 00:13 |
| abortretryfail | they're good to have. I got an old Tenergy one that charges 4 cells and will do Li-ion, NiMH, and NiCd too | 00:13 |
| vagrantc | if you can, find one that *only* does lifepo4 ... many do other li-ion chemistries and require you to manually change it when you plug the battery in | 00:13 |
| abortretryfail | yeah mine has a switch, it's not a big deal | 00:13 |
| vagrantc | and those other chemistries are dangerous for the lifepo4 | 00:13 |
| noam | Any suggestions on which one to get? :/ | 00:14 |
| vagrantc | i've got an xtar vp2, but apparently it is discontinued :( | 00:15 |
| vagrantc | (er, rather, the voltages used to charge other chemistries are too high for lifepo4) | 00:16 |
| noam | trying to find a decent one *in the US* so it doesn't take months to ship :/ | 00:17 |
| ex-parrot | oh noam also, upgrading the LPC firmware gets you working battery monitoring in the OS :) so it's also worth doing for that too | 00:17 |
| ex-parrot | josch: are you around today? I meant to follow up with you on the emmc updater script | 00:18 |
| abortretryfail | noam: this is the one I have. https://www.amazon.com/Tenergy-Intelligent-Universal-Rechargeable-Batteries/dp/B00L4HNGQS/ | 00:19 |
| noam | ex-parrot: I did do that already | 00:19 |
| ex-parrot | nice | 00:19 |
| noam | abortretryfail: found this from home depot https://www.homedepot.com/p/Dantona-ULGLION-4-Battery-Charger-ULGLION-4/314247174 , but I'm leaving the state on Monday and it can't deliver before tuesday :/ | 00:20 |
| noam | So I'll probably order that, and figure out a hack for today | 00:20 |
| noam | Need to figure out how to get constant-current to feed to them | 00:20 |
| noam | ...if I had a way to look at current, I could hack up a program for a microcontroller to adjust the intensity of a transistor or something... | 00:21 |
| ex-parrot | got a microcontroller with analog input? | 00:22 |
| noam | yes | 00:22 |
| ex-parrot | could read the voltage across a shunt maybe | 00:22 |
| ex-parrot | but probably a constant voltage + resistor would do, you just want to charge gently mainly | 00:22 |
| noam | how would that help for keeping the _current_ constant? | 00:22 |
| noam | ah, k | 00:22 |
| noam | So, what if I just used a buck converter to drop voltage to ~1V or some such? :P | 00:22 |
| ex-parrot | so the voltage drop across the shunt is a function of the current | 00:23 |
| ex-parrot | if you can't measure current you can make an ammeter with a shunt + voltmeter | 00:23 |
| noam | Shunt? | 00:23 |
| ex-parrot | low value resistor | 00:23 |
| ex-parrot | say 30 ohms or so | 00:23 |
| noam | [I mostly do software stuff, I've forgotten most of what I knew about the hardware side :/] | 00:23 |
| noam | ahhh, right, thanks! | 00:23 |
| ex-parrot | ohms law says that current = voltage / resistance | 00:24 |
| noam | actually, here's an idea | 00:24 |
| ex-parrot | you need a charging voltage for the cell that is at least a bit higher than what the cell voltage is currently to get the chemical reaction going | 00:24 |
| noam | What if I grabbed a 1.2V battery and discharged that into the cell? | 00:24 |
| noam | Maybe through a resistor to lower the voltage even further | 00:24 |
| ex-parrot | yeah might work, tho I'd suggest if you can you really want to measure the current flow in to the cell and try and fiddle with the resistance to keep the current flow low, below say 100ma or so | 00:25 |
| ex-parrot | to prevent heating the lifepo cell and damaging it further, and it supposedly helps with the crystal formation in the battery etc to charge at a lower rate | 00:25 |
| ex-parrot | if you don't have any external resistor then you're going to be limiting the current based on the internal resistance of the power supply + your lifepo4 cell | 00:26 |
| ex-parrot | which if the lifepo4 is very dead or shorted could be quite low, causing large current to flow and heating etc | 00:26 |
| noam | right... | 00:26 |
| noam | ...what if I just use a reallllllllllly long copper wire lol | 00:26 |
| noam | that might actually work as a makeshift resistor | 00:26 |
| ex-parrot | have you got a multimeter handy? | 00:26 |
| noam | I've done that before with a USB charger to reduce the voltage | 00:26 |
| noam | ex-parrot: yep, that's how I know the cell's at a negative voltage :P | 00:26 |
| ex-parrot | it hasn't got a current measurement mode? | 00:27 |
| noam | it does | 00:27 |
| ex-parrot | ok cool | 00:27 |
| ex-parrot | and you're confident in using it, e.g. you know you need to move the leads usually to a dedicated current measurement terminal and put the meter in series with the load rather than in parallel like you would for measuring voltage? | 00:27 |
| noam | as ok last week when I blew the 10A fuse by putting it in parallel? Yes | 00:28 |
| noam | :P | 00:28 |
| noam | s/ok/of/ | 00:28 |
| ex-parrot | yeah nice :D | 00:28 |
| ex-parrot | so this is getting in to definitely don't try this at home territory | 00:28 |
| ex-parrot | but if you can find a graphite pencil | 00:28 |
| ex-parrot | you can use the graphite lead as a resistor | 00:28 |
| noam | okay but | 00:28 |
| ex-parrot | and adjust the length of the piece of graphite to contorl the resistance | 00:28 |
| noam | I have a spool of copper wire | 00:28 |
| noam | Wouldn't that be easier? :P | 00:28 |
| ex-parrot | you could measure out a few metres of wire I guess, but it'll only be a few ohms most likely | 00:29 |
| noam | or heck, a long strand of solder | 00:29 |
| ex-parrot | you will prob want a few hundred ohms at least | 00:29 |
| ex-parrot | to gently charge the lifepo4 | 00:29 |
| ex-parrot | it sounds like you have all the things you need to macguyver a trickle charger anyway :P | 00:29 |
| noam | cept the knowledge :P | 00:30 |
| ex-parrot | wire up a circuit with power supply + multimeter in ammeter mode + lifepo4 cell, and just try different resistances until you get around 100ma flowing in to the battery | 00:30 |
| ex-parrot | then keep an eye on it for half an hour or so and see if the lifepo4 cell voltage increases t all | 00:31 |
| noam | dunno that I have a good way to keep the multimeter in the circuit :/ | 00:32 |
| noam | ...actually | 00:33 |
| + S0rin (~S0rin@user/s0rin) | 00:36 | |
| josch | ex-parrot: in my timezone, "today" started half an hour ago, so the answer is "yes" after having slept for the next six hours :) | 00:39 |
| ex-parrot | perfect :) | 00:39 |
| noam | ex-parrot: problem is, the only terminal I have is fused at 200mA; if I'm accidentally running 205mA through, I lose my voltmeter too :( | 00:40 |
| ex-parrot | josch: ping me if you want to talk about the rescue flash script at some point | 00:40 |
| ex-parrot | noam: :( | 00:40 |
| josch | ex-parrot: what was the issue with it? | 00:41 |
| ex-parrot | currently it doesn't work because it downloads a system image with the modules etc for the -9- kernel and the repo is serving up a -10- kernel package | 00:42 |
| noam | ooh | 00:42 |
| ex-parrot | (the reform.debian repo) | 00:42 |
| noam | ex-parrot: I'm giving it brief pulses of power from a NiMH AAA battery (brief so as to avoid overheating it - like, two or three seconds at a time), and it's up from -0.3 to -0.1V :) | 00:43 |
| ex-parrot | noam: good luck :) | 00:43 |
| noam | I can at least use the voltmeter in between pulses | 00:43 |
| josch | ex-parrot: ah you are using the reform-flash-rescue script with the image from reform.debian.net? | 00:44 |
| ex-parrot | yeah | 00:44 |
| ex-parrot | I'm also happy for "that's not supported" :P | 00:44 |
| josch | ex-parrot: but you are running reform-flash-rescue from an sd-card flashed with an image from reform.debian.net? | 00:44 |
| ex-parrot | ah right, no I haven't switched to using your SD images yet | 00:45 |
| ex-parrot | this is the latest v3 sysimage where I switched over to your repo | 00:45 |
| josch | then that issue is independent of reform.d.n | 00:45 |
| ex-parrot | got it, didn't realise there were images on reform.d.n already | 00:45 |
| ex-parrot | I'll switch over to that :) | 00:45 |
| josch | it will happen in the same way when the kernel from the mnt.re CI and mnt.re repos differ | 00:45 |
| ex-parrot | cool | 00:45 |
| ex-parrot | neat to have images and a branded page and everything | 00:46 |
| josch | you mean the new MNT research page? | 00:46 |
| ex-parrot | I just mean https://reform.debian.net/ | 00:47 |
| josch | ah | 00:47 |
| ex-parrot | last time I looked at it it was just your plaintext page with the repo instructions | 00:47 |
| ex-parrot | that might have been ~ 2 weeks ago though :D | 00:47 |
| josch | it was easy because https://raspi.debian.net/ exists :) | 00:47 |
| ex-parrot | :D | 00:47 |
| ex-parrot | excellent work josch | 00:47 |
| ex-parrot | I know I say this a lot but I really appreciate the hours everyone is pouring in to this project, it's so good | 00:47 |
| josch | :) | 00:48 |
| ex-parrot | I need to decide if I want to swap out my LFS install on the nvme for Debian | 00:49 |
| ex-parrot | less fun but maybe more useful overall. not that I really need more "useful" computers | 00:49 |
| ex-parrot | oh uh josch | 00:53 |
| ex-parrot | https://reform.debian.net/images/reform2-imx8mq-system.img.xz is 404 currently | 00:53 |
| ex-parrot | as is https://reform.debian.net/images/reform2-imx8mq-rescue-system.img.xz | 00:53 |
| josch | oh whoops | 00:55 |
| noam | apparently fully drained the AAA just getting back up to -0.02V O_o | 00:56 |
| noam | ...it was probably low before, actually | 00:56 |
| noam | that's why I picked it, its voltage was lower :P | 00:57 |
| ex-parrot | I hate to say it noam but I suspect your lifepo4 cell isn't coming back from this | 00:57 |
| noam | We'll see! | 00:57 |
| noam | It's almost back to 0V :P | 00:57 |
| josch | ex-parrot: i fixed the rescue system image location | 00:58 |
| ex-parrot | ACTION downloads | 00:58 |
| josch | ex-parrot: the full image cannot be created right now because of https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1039472 | 00:58 |
| ex-parrot | aha | 00:58 |
| ex-parrot | v good | 00:59 |
| ex-parrot | I'll flash my rescue system in the meantime | 00:59 |
| noam | it's up to ~0.5V :) | 01:08 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 01:13 | |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 01:19 | |
| noam | ex-parrot: I probably shouldn't put them back in the Reform until and unless they're _all_ at ~2V, right? | 01:45 |
| ex-parrot | uhhh it can rebalance them to some extent | 01:45 |
| noam | But not if three cells are at near zero volts, presumably | 01:46 |
| noam | I'd expect that to go Hissssss | 01:46 |
| noam | no? | 01:46 |
| ex-parrot | yeah you don't want that | 01:47 |
| ex-parrot | you'll need them to be approx ready to go | 01:47 |
| noam | yeah, I figured :/ | 01:48 |
| noam | I need to order new cells for sure | 01:49 |
| noam | Any recommendations on where to look, other than BatterySpace? | 01:49 |
| noam | [for within the US] | 01:49 |
| ex-parrot | no idea, I've never been to the US | 01:49 |
| ex-parrot | I got my most recent set from AliExpress but had to clip the protection circiutry off each cell | 01:50 |
| ex-parrot | they seem OK otherwise tho | 01:50 |
| noam | That's all shipping from China, no? | 01:52 |
| ex-parrot | yeah | 01:52 |
| noam | That'll take until September to arrive O_o | 01:52 |
| noam | ...which might be the right way to go. | 01:53 |
| ex-parrot | it's almost september :P | 01:53 |
| noam | I won't be here when they'd arrive lol | 01:53 |
| ex-parrot | :( | 01:53 |
| noam | I mean the _end_ of september | 01:53 |
| noam | It's probably fine | 01:53 |
| noam | As long as I can power it off of my external battery pack, this isn't the end of the world | 01:53 |
| noam | I mean, it's not regardless | 01:53 |
| noam | :P | 01:53 |
| noam | and I'm going to *try* to resuscitate these cells | 01:53 |
| noam | but holy heck, the protected battery board should be _mandatory_ | 01:54 |
| ex-parrot | sounds good to me | 01:54 |
| ex-parrot | yeah I think the protected board is shipping in all new units? someone may correct me on that | 01:54 |
| noam | i think so, but only from the MNT shop | 01:54 |
| noam | I ordered mine through Crowd Supply, the US distributor | 01:54 |
| ex-parrot | it took a while to understand the problem with the power drain too | 01:54 |
| noam | huh, there's 2Ah cells on AliExpress - what are the odds that those are legit? | 01:54 |
| noam | ex-parrot: wait, do we know where that's coming from now? | 01:55 |
| ex-parrot | yes | 01:55 |
| noam | And? :) | 01:55 |
| ex-parrot | the charge circuitry on the motherboard does have a low voltage cutoff but there's one connection "above" it which doesn't get cut | 01:55 |
| ex-parrot | which is why the new battery boards also protect the individual cells | 01:55 |
| ex-parrot | let me link you the cells I have | 01:56 |
| noam | Thanks :) | 01:56 |
| ex-parrot | hmm they don't seem to exist anymore | 01:56 |
| ex-parrot | they were So | 01:56 |
| ex-parrot | SoShine brand 1800mah | 01:57 |
| ex-parrot | removing the protection is fairly simple | 01:57 |
| noam | what are the odds I can assume that e.g. https://www.aliexpress.us/item/3256805594707990.html are legit? I haven't used aliexpress, don't know how trustworthy it/the sellers are | 01:58 |
| - yewscion (QUIT: *.net *.split) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 02:00 | |
| - klardotsh (QUIT: *.net *.split) (~klardotsh@98.97.112.82) | 02:00 | |
| - mjw (QUIT: *.net *.split) (~mjw@gnu.wildebeest.org) | 02:00 | |
| - mtm (QUIT: *.net *.split) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 02:00 | |
| - XYZ (QUIT: *.net *.split) (~XYZ@37-48-48-59.nat.epc.tmcz.cz) | 02:00 | |
| - inhji (QUIT: *.net *.split) (~inhji@2a01:4f8:1c0c:42e4::1) | 02:00 | |
| - Ar|stote|is (QUIT: *.net *.split) (~linx@149-210-5-215.mobile.nym.cosmote.net) | 02:00 | |
| ex-parrot | yeah not sure tbh for batteries, I only have that one data point | 02:00 |
| + mjw (~mjw@gnu.wildebeest.org) | 02:00 | |
| + inhji (~inhji@2a01:4f8:1c0c:42e4::1) | 02:00 | |
| ex-parrot | "laptop batteries" on aliexpress for e.g. thinkpads tend to be pretty low quality, don't know about individual cells | 02:00 |
| + XYZ (~XYZ@37-48-48-59.nat.epc.tmcz.cz) | 02:00 | |
| noam | Is there money back policy legit? | 02:00 |
| noam | their* | 02:00 |
| ex-parrot | yep | 02:00 |
| noam | jeez | 02:00 |
| + klardotsh (~klardotsh@98.97.112.82) | 02:00 | |
| noam | Okay | 02:00 |
| ex-parrot | the buyer protections are generally fine | 02:00 |
| ex-parrot | the buyer protections are generally fine | 02:01 |
| noam | I've e.g. bought a 55Wh thinkpad battery on ebay because I trusted that if it *wasn't* what it claimed to be, I'd be fine | 02:01 |
| ex-parrot | "1 sold" is a bit of a red flag maybe | 02:01 |
| noam | if aliexpress' protections are as good, then it's fine | 02:01 |
| noam | Sure | 02:01 |
| ex-parrot | yeah should be | 02:01 |
| ex-parrot | also 0.0% Positive Feedback | 02:01 |
| ex-parrot | I'd maybe look for a store with more feedback / sales | 02:01 |
| + yewscion (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 02:01 | |
| noam | https://www.aliexpress.us/item/3256805020586897.html?spm=a2g0o.productlist.main.5.71bf20c71bHMlI&algo_pvid=4670ad9c-5601-46b8-9d43-4adcd49a0f67&algo_exp_id=4670ad9c-5601-46b8-9d43-4adcd49a0f67-2&pdp_npi=3%40dis%21USD%2144.56%2131.19%21%21%2144.56%21%21%40211bc71916898970830858392d0777%2112000032163081980%21sea%21US%210&curPageLogUid=FJScmSIK4uNx | 02:01 |
| noam | Eugh | 02:01 |
| noam | their tracking is worse than ebay's lol | 02:01 |
| noam | that one looks better, at least... | 02:01 |
| ex-parrot | report back how it goes :) | 02:02 |
| noam | There's a review, even - someone claims that the batteries tested to only be 1.7Ah _when discharged to 2V_, which... uh. People are discharging HOW far?? | 02:03 |
| noam | I"m actually more annoyed if the listed capacity is by increasing the rated depth-of-discharge lol | 02:03 |
| ex-parrot | you def take a bit more of a chance on aliexpress with this stuff | 02:04 |
| ex-parrot | I didn't have much choice here | 02:04 |
| + Ar|stote|is (~linx@149-210-5-215.mobile.nym.cosmote.net) | 02:05 | |
| noam | I'm probably better off ordering from BatterySpace. That'll be a tad faster anyways | 02:06 |
| noam | On the one hand, I want to grab two sets, just in case - on the other hand, they only really HAVE two sets, so that's probably unethical | 02:06 |
| ex-parrot | ehh they'll have more on order | 02:07 |
| ex-parrot | no ethical consumption under capitalism anyway | 02:07 |
| noam | Ehhh | 02:09 |
| noam | Buying up the entire stock because I *MIGHT* need it *in a while* seems dubious. | 02:10 |
| ex-parrot | maybe :P | 02:10 |
| ex-parrot | just order another set later? | 02:10 |
| noam | And nah, they literally only have 17 cells right now | 02:10 |
| noam | Or so they claim anyways :P | 02:10 |
| noam | Yeah, looks valid | 02:12 |
| noam | They now claim to have only nine | 02:12 |
| noam | Given that I bought eight a few mintues ago... yeah | 02:12 |
| - mjw (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org) | 02:13 | |
| noam | ex-parrot: thanks for all the help :) | 02:15 |
| ex-parrot | no worries | 02:16 |
| noam | I really need a good resistor... otherwise, my only real options are 0.2V, 5V, or to try using _another battery_ which, upon reflection, is an extremely ridiculous idea | 02:17 |
| noam | [or 12V or 20V, but, uh, that's obviously NOT helpful] | 02:18 |
| sevan | :q | 02:31 |
| sevan | derp :) | 02:31 |
| noam | ex-parrot: is 500mA a bit too high? :/ | 02:34 |
| noam | I can get ~500mA at 3.3V stably through the voltage regulator on a small microcontroller I've got lying around :P | 02:34 |
| ex-parrot | ehh might be OK? if nothing is getting too hot | 02:37 |
| noam | seems to be working :) | 02:37 |
| noam | The uController _should_ only be giving like 100mA, but apparently it's actually _better_ than the spec sheet claims | 02:39 |
| noam | Which in this case makes it *worse* :P | 02:39 |
| noam | How concerned should I be, safety wise, if I *do* get all eight cells back to nominal voltage? | 02:40 |
| + jacobk (~quassel@47-186-110-219.dlls.tx.frontiernet.net) | 02:45 | |
| noam | shoot. ONE of the cells seems dead-dead :( | 02:54 |
| noam | I was so proud of myself, too; I got *seven* of the eight nearly up to the nominal (I was rotating between the cells), but the last one's just... gone :( | 03:07 |
| - nsc (QUIT: Ping timeout: 246 seconds) (~nicolas@94-98-142-46.pool.kielnet.net) | 03:56 | |
| + nsc (~nicolas@3-49-142-46.pool.kielnet.net) | 03:58 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 04:10 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 05:50 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 06:12 | |
| - XYZ (QUIT: Ping timeout: 246 seconds) (~XYZ@37-48-48-59.nat.epc.tmcz.cz) | 06:41 | |
| - minute (QUIT: Ping timeout: 252 seconds) (~mntirc@softboy.mntmn.com) | 08:20 | |
| + reformer (~reformer@softboy.mntmn.com) | 09:11 | |
| vkoskiv | minute: any lion sightings yet? :D | 09:20 |
| - ajr (QUIT: Quit: Connection closed for inactivity) (uid609314@user/ajr) | 09:22 | |
| nsc | vkoskiv: i can confirm that the north of Berlin is still free from lions | 09:31 |
| + mjw (~mjw@gnu.wildebeest.org) | 11:31 | |
| minute | vkoskiv: we believe it's a boar | 11:59 |
| minute | this is good i guess https://www.arm.com/company/news/2023/07/arm-expands-open-source-partnerships | 12:00 |
| - mjw (QUIT: Ping timeout: 246 seconds) (~mjw@gnu.wildebeest.org) | 12:27 | |
| * mark_ -> mjw | 12:35 | |
| minute | https://www.collabora.com/news-and-blog/news-and-events/a-helping-arm-for-panfrost.html | 12:50 |
| - cwebber (QUIT: Remote host closed the connection) (~user@user/cwebber) | 13:44 | |
| - Boostisbetter (QUIT: Ping timeout: 264 seconds) (4a410829d7@irc.cheogram.com) | 14:01 | |
| - mtm (QUIT: Ping timeout: 244 seconds) (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 14:03 | |
| vkoskiv | minute: https://cdn.discordapp.com/attachments/1012807179075600535/1131927169887649824/20230721_143414.jpg | 14:43 |
| minute | cute | 14:57 |
| abortretryfail | noam: *one* cell failing in an entire pack of them is about on par for what I find on laptop packs I've disassembled. Usually one will end up reverse-biased somehow and be properly dead and the others can be brought back by something like what you did. | 15:09 |
| abortretryfail | The bms on the pack will do the safe thing and just refuse to charge all of them because the one dead one might be dangerous. | 15:10 |
| + [tj] (sid609767@id-609767.lymington.irccloud.com) | 15:20 | |
| - yewscion (QUIT: Ping timeout: 246 seconds) (~yewscion@2601:547:a01:2900:5c15:78d2:b913:fc63) | 15:28 | |
| + bkeys (~Thunderbi@199.226-24.cm.ptn.tftn.dynamic.friendlycity.net) | 15:38 | |
| + yewscion (~yewscion@c-73-236-134-241.hsd1.pa.comcast.net) | 16:06 | |
| + mtm (~mtm@c-71-228-84-213.hsd1.fl.comcast.net) | 16:10 | |
| josch | so... i never got zoom via browser working on the reform -- the imx8mq is just too slow | 16:13 |
| josch | but today i had a video chat (via browser) on an online authentication platform | 16:13 |
| josch | which went smooth like butter | 16:13 |
| josch | no visible frame drops or audio jitters | 16:13 |
| josch | cpu didn't even go beyond 25% utilization | 16:13 |
| josch | i wonder what zoom is doing wrong... | 16:14 |
| minute | huh! | 16:19 |
| minute | maybe it requests a different webcam resolution? | 16:20 |
| minute | i'm struggling getting the LPC SPI driver to work on a311d... i'm using spi-gpio | 16:20 |
| minute | the SCLK and MOSI look good on the scope, but there's nothing on MISO | 16:21 |
| josch | and another thing: after having swapped out some cells, the reform doesn't anymore die at 35% but at 9%. Does the percentage at which the reform will die go up the older my cells get? | 16:27 |
| sknebel | some browser-video things caused performance issues in general because they requested multiple resolutions | 16:34 |
| minute | hmm, it doesn't look like spi-gpio is honoring the speed limit | 16:38 |
| minute | josch: is this the percentage as reported on LPC oled? | 16:38 |
| josch | minute: yes, the oled percentage | 16:45 |
| josch | (it matches the one reported via the lpc dkms kernel module) | 16:46 |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 16:46 | |
| minute | sigh > #define spidelay(nsecs)do {} while (0) | 16:47 |
| minute | (linux/drivers/spi/spi-gpio.c) | 16:47 |
| minute | josch: normally it works like that: when the voltage of a cell goes critical, lpc defines this at the new "empty" point (0%). so the % is flexible. it tries to remember the capacity in mAh at this zero point and also tries to find a new 100% point when charging. | 16:51 |
| minute | josch: when you reset lpc, it forgets all of this | 16:51 |
| josch | that sounds reasonable | 16:51 |
| minute | josch: so when you exchange the cell, you reset lpc, and the process begins anew | 16:51 |
| josch | then i'll go back down to zero again | 16:51 |
| josch | which should now work | 16:51 |
| minute | hopefully | 16:52 |
| josch | funny that the last time it died at 35%, meaning i haven't been close to 0 for a while, no? XD | 16:52 |
| minute | it is also possible that the algorithm is not yet perfect :3 | 16:52 |
| minute | it tries to measure mA in/outflow every second and adds or subtracts these from the "account". if time is stretched, there's a lot of sleeping and waking etc, it's possible that this timing and accounting is inaccuarate and errors compound | 16:53 |
| minute | hmm, actually there is data coming out of MISO | 17:09 |
| mjw | Wait there are bling, bling, silver mnt reforms? https://mastodon.social/@mntmn/110751841517802277 | 17:24 |
| amospalla[m] | I like a lot the images of the Reforms keyboards. | 17:27 |
| amospalla[m] | How does it feel writing on a Reform with this mechanical switch? | 17:28 |
| minute | amospalla[m]: it's one of the main features of the device :3 | 17:28 |
| minute | mjw: no, this was just a CNC sample without anodizing :D | 17:28 |
| sknebel | it does work quite well for the industrial look :D | 17:32 |
| mjw | Too bad. Having a bling bling mnt reform (gold!) would be cool. | 17:36 |
| + Boostisbetter (4a410829d7@irc.cheogram.com) | 18:01 | |
| + mark_ (~mjw@gnu.wildebeest.org) | 18:23 | |
| - mjw (QUIT: Killed (NickServ (GHOST command used by mark_!~mjw@gnu.wildebeest.org))) (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:23 | |
| * mark_ -> mjw | 18:23 | |
| + mark_ (~mjw@2001:1c06:2488:1400:4fd:39a7:74ac:7bae) | 18:24 | |
| + XYZ (~XYZ@78-80-99-104.customers.tmcz.cz) | 18:29 | |
| minute | i'm getting "cb cb cb cb cb cb cb cb" (bytes) from the lpc status command in my log. then i was looking at the scope and counting 1s and 0s on falling clock transitions: 0b11001011... | 18:31 |
| minute | which is 0xcb | 18:31 |
| minute | but it's interpreted as "-13.-365 -13.365 203% 203" | 18:31 |
| noam | holy heck, battery board is expected to be here on _Thursday_?? Wow that shipping is _fast_ | 18:51 |
| noam | also, uh: given seven resuscitated cells, what's a good way to test them to be sure they're safe to use? | 18:51 |
| noam | and: is using them with an eighth cell of equivalent effective capacity safe, in the Reform? | 18:52 |
| + klardotsh (~klardotsh@98.97.36.213) | 18:59 | |
| minute | noam: yes, that's theoretically safe | 19:01 |
| noam | If I can do a full charge-discharge cycle (or three!) after resuscitating the cells, that probably indicates that they're safe to use, right? | 19:09 |
| noam | ...I would never be this cavalier with other chemistries :P | 19:09 |
| minute | noam: yep | 19:13 |
| + ajr (uid609314@user/ajr) | 19:16 | |
| minute | oh nice, the "hack the planet" header is useful now for the first time... | 19:34 |
| minute | i made myself another debug UART for the lpc to my external computer | 19:35 |
| minute | if you set LPC_IOCON->PIO0_19 = 0x01; // TXD | 19:35 |
| minute | then you can read the debug uart output on LPC_TXDa | 19:35 |
| + marty_mcfly88 (~marty@2603-6000-8c01-0f3e-5197-f1da-916a-84d6.res6.spectrum.com) | 19:45 | |
| minute | hmm, it works now, how annoying | 20:17 |
| minute | so the only required fix appears to be `gpioget -B pull-up 0 20` before loading spi_gpio | 20:36 |
| minute | to pull up the MISO line a bit higher | 20:36 |
| minute | our battery driver has a missing attribute, "present" | 21:09 |
| minute | cbatticon expects it | 21:09 |
| vagrantc | "the battery which may or may not actually be present has this much capacity remaining" | 21:12 |
| vagrantc | isn't that the sort of thing you can, infer? | 21:13 |
| minute | well... | 21:26 |
| minute | cbatticon doesn't work anyway because apparently it doesn't use SNI or something for the tray | 21:27 |
| vagrantc | spooky batteries at a distance? :) | 21:27 |
| minute | it uses gtk_status_icon_new | 21:27 |
| minute | ah, this is all deprecated stuff. oh well | 21:32 |
| - robin (QUIT: Quit: Leaving) (~robin@user/terpri) | 21:40 | |
| - klardotsh (QUIT: Ping timeout: 244 seconds) (~klardotsh@98.97.36.213) | 21:42 | |
| + sknebel_ (~quassel@v22016013254630973.happysrv.de) | 21:47 | |
| - sknebel (QUIT: *.net *.split) (~quassel@v22016013254630973.happysrv.de) | 21:54 | |
| - jfred (QUIT: *.net *.split) (sid534649@libera/sponsor/jfred) | 21:54 | |
| + jfred (sid534649@libera/sponsor/jfred) | 22:01 | |
| - yewscion (QUIT: Remote host closed the connection) (~yewscion@c-73-236-134-241.hsd1.pa.comcast.net) | 22:39 | |
| - marty_mcfly88 (QUIT: Ping timeout: 246 seconds) (~marty@2603-6000-8c01-0f3e-5197-f1da-916a-84d6.res6.spectrum.com) | 22:56 | |
| + klardotsh (~klardotsh@98.97.36.213) | 22:58 | |
| + yewscion (~yewscion@2601:547:1480:bc60:9da:d549:1885:c96c) | 23:16 | |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 23:36 | |
| * sknebel_ -> sknebel | 23:37 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
