| jfred | josch: Ooh, if you get Guix running on the Reform that'll be exciting! | 00:09 |
|---|---|---|
| josch | jfred: according to the forum you can have guix on the reform today :) | 00:09 |
| jfred | :O | 00:09 |
| jfred | I'll have to try this out | 00:10 |
| jfred | I looked at the board support stuff in Guix before but my u-boot-fu is lacking | 00:10 |
| josch | jfred: https://codeberg.org/lykso/mnt-reform-nonguix | 00:14 |
| + eukarya (~clay@173.17.198.111) | 00:43 | |
| - Gooberpatrol66 (QUIT: Ping timeout: 268 seconds) (~Gooberpat@user/gooberpatrol66) | 01:19 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 01:23 | |
| - natalie (QUIT: Remote host closed the connection) (~natalie@user/natalie) | 01:59 | |
| + natalie (~natalie@user/natalie) | 02:01 | |
| - cobra (QUIT: Ping timeout: 268 seconds) (~cobra@user/Cobra) | 02:20 | |
| + vagrantc (~vagrant@2600:3c01:e000:21:7:77:0:50) | 02:24 | |
| violet | minute: ok i ordered the exact part number from the schematics BOM ( Würth Elektronik | 03:38 |
| violet | oops | 03:38 |
| violet | Würth Elektronik 687733050002 | 03:38 |
| violet | and this is indeed the wrong cable. it's 3 pins less than what i actually need | 03:38 |
| violet | so apparently the cable is actually 33 pin, and this part number / the parts of the docs that says "30 pin" are incorrect | 03:38 |
| violet | (im referencing https://mntre.com/reform2/handbook/schematics.html ) | 03:39 |
| violet | 687733100002 and 687733152002 seem to be the 33pin ones (100mm and 152mm length respectively) | 03:42 |
| violet | <- could have counted the pins in the original cable before buying, but decided to just check a new cable against the old cable. lol | 03:45 |
| + cobra (~cobra@user/Cobra) | 03:51 | |
| violet | i ordered the part that i think is correct. if it works i can PR a fix to the handbook | 04:03 |
| - vagrantc (QUIT: Quit: leaving) (~vagrant@2600:3c01:e000:21:7:77:0:50) | 04:55 | |
| josch | oh nooo i missed vagrantc XD | 07:39 |
| josch | curse these timezones! | 07:39 |
| + mgd (~mgd@host-89-240-113-218.as13285.net) | 07:59 | |
| - mgd (QUIT: Remote host closed the connection) (~mgd@host-89-240-113-218.as13285.net) | 08:01 | |
| - cobra (QUIT: Ping timeout: 272 seconds) (~cobra@user/Cobra) | 08:39 | |
| Twodisbetter | bummer, is he ahead or behind us? | 08:45 |
| + cobra (~cobra@user/Cobra) | 09:17 | |
| + gustav28 (~gustav@c-0c35524e.019-141-67626730.bbcust.telenor.se) | 10:02 | |
| - Gooberpatrol66 (QUIT: Ping timeout: 256 seconds) (~Gooberpat@user/gooberpatrol66) | 10:29 | |
| - amk (QUIT: Remote host closed the connection) (~amk@user/amk) | 10:30 | |
| + amk (~amk@user/amk) | 10:30 | |
| vkoskiv | josch: I'll do some poking around with a scope to try and understand the backlight thing a bit more, thanks for the reminder! | 10:43 |
| josch | vkoskiv: thank you! <3 | 10:55 |
| josch | it's a super, super tiny issue but that's why i recorded it -- it's so easy to get lost because it really is not a bother unless it's pitch black darkness :) | 10:55 |
| - nybble (QUIT: Ping timeout: 260 seconds) (~sprang@user/csprng) | 10:58 | |
| + Gooberpatrol66 (~Gooberpat@user/gooberpatrol66) | 11:22 | |
| + murph__ (~murph@ool-457bb02e.dyn.optonline.net) | 11:55 | |
| - murph_nj (QUIT: Ping timeout: 252 seconds) (~murph@ool-457bb02e.dyn.optonline.net) | 11:57 | |
| - gustav28 (QUIT: Quit: Quit) (~gustav@c-0c35524e.019-141-67626730.bbcust.telenor.se) | 12:32 | |
| + chomwitt (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 13:12 | |
| - chomwitt (QUIT: Ping timeout: 255 seconds) (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 13:35 | |
| + chomwitt (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 14:09 | |
| - Gooberpatrol66 (QUIT: Quit: Konversation terminated!) (~Gooberpat@user/gooberpatrol66) | 14:23 | |
| - chomwitt (QUIT: Ping timeout: 272 seconds) (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 14:29 | |
| + chomwitt (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 15:05 | |
| josch | checking build system type... x86_64-unknown-linux-gnu | 15:22 |
| josch | checking host system type... Invalid configuration `aarch64-linux-gnu': machine `aarch64' not recognized | 15:22 |
| josch | minute: seems like trying this on the arm64 runner instead would make sense :) | 15:22 |
| josch | minute: https://source.mnt.re/josch/mnt-reform-nonguix/ was the project | 15:23 |
| josch | the guix scripts seem to require to be run on arm64 :) | 15:24 |
| josch | minute: if you could switch/add the arm64 runner to that project, that would be nice -- thank you! | 15:24 |
| - cobra (QUIT: Ping timeout: 268 seconds) (~cobra@user/Cobra) | 15:37 | |
| - murph__ (QUIT: Quit: Leaving) (~murph@ool-457bb02e.dyn.optonline.net) | 15:41 | |
| - chomwitt (QUIT: Ping timeout: 256 seconds) (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 15:41 | |
| + cobra (~cobra@user/Cobra) | 16:02 | |
| - cobra (QUIT: Ping timeout: 256 seconds) (~cobra@user/Cobra) | 16:29 | |
| - eukarya (PART: !!unknown attribute: msg!!) (~clay@173.17.198.111) | 16:33 | |
| truck | ACTION waves to _rm | 16:43 |
| truck | (prototype #8) | 16:45 |
| vkoskiv | Very likely time for a clean. | 16:45 |
| vkoskiv | Two screws release the retaining ring, ball can then be removed and you can check to see if there is dust on the sensor | 16:46 |
| truck | vkoskiv: did you give rm where you got the new battery cells (and the battery fix thing) | 16:46 |
| truck | I... um... didn't touch it for too long and "oh look it's very not in good shape" ): ) | 16:46 |
| vkoskiv | I just got the protected battery boards from mnt directly https://shop.mntre.com/products/protected-battery-board | 16:46 |
| vkoskiv | And I sourced the cells from a Finnish supplier, let me check which one | 16:47 |
| truck | ACTION remembers 'akku' is not 'duck' but 'battery' | 16:47 |
| vkoskiv | Yeah, it was akkula, this particular one: https://akkula.fi/tuote/18650-lifepo4-1800-mah-32v-3c-akku-flat-top/ | 16:48 |
| vkoskiv | _rm: Yeah! I remember :] | 16:48 |
| truck | rm wanted this and since, you know, since mine was in "I want to use it but I need to learn Finnish and get employed again" I thought it best that it be USED, so... he's 'borrowing' it (which seems to be more "fixing") | 16:48 |
| vkoskiv | I don't know how much the prototypes differ from final production units, but the protected battery boards will be a great upgrade | 16:49 |
| truck | yeah, that needs to be done. and possibly a cpu upgrade. | 16:49 |
| vkoskiv | I've considered getting the rockchip SoM, which would then leave me with a spare A311D module, which I'd be happy to sell to anyone for a good price | 16:50 |
| vkoskiv | A311D was a really good upgrade already | 16:50 |
| truck | and possibly the changing of the lid to 'backlit' ... and also move the antennae to there, like what ... darn it, who are the other people who go to Revision who have reforms (: | 16:51 |
| vkoskiv | I'm actually debugging the keyboard firmware at the moment, I'm trying to figure out some weirdness with the keyboard backlight PWM control | 16:51 |
| truck | not venty, hmm | 16:51 |
| vkoskiv | truck: I believe they are vimja and gaston | 16:51 |
| truck | yes. vimja. | 16:51 |
| truck | vimja has a really cool modification for his original (he has 2 (: ) where he made the lid backlit, printed his name in it, and moved the antennae to it, improving the radio reception | 16:52 |
| truck | yeah it seems there are a lot of sceners who picked this up (: | 16:53 |
| truck | yeah it should (: | 16:55 |
| josch | vkoskiv: i'd be interested in your a311d -- which continent are you on? :) | 17:02 |
| vkoskiv | Europe, but the very north bit - Helsinki | 17:04 |
| josch | should be managable to Germany! :) | 17:06 |
| josch | sepa transfers work to finland? | 17:06 |
| vkoskiv | Yep! | 17:07 |
| josch | i call dibs then :) | 17:07 |
| vkoskiv | Cool! I'll let you know when I get the new module set up. I think I may still have the original packaging for the A311D | 17:09 |
| josch | vkoskiv: original packaging is not important -- just that everything is there to update an imx8mq to a311d, like heatsink etc | 17:12 |
| vkoskiv | Yeah, all the original stuff would be included, of course. | 17:13 |
| josch | nice, thank you! :) | 17:13 |
| + chomwitt (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 17:15 | |
| + mjw (~mjw@gnu.wildebeest.org) | 17:20 | |
| - chomwitt (QUIT: Ping timeout: 246 seconds) (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 17:24 | |
| minute | we recently shipped 50x a311d upgrades to crowd supply | 17:28 |
| minute | they are stocked and available at a good price | 17:28 |
| minute | josch: i just switched the runner for your nonguix project to arm64 | 17:32 |
| josch | thank you! i'm re-running the pipeline :) | 17:34 |
| vkoskiv | So at max keyboard brightness, my scope indicates the duty cycle is 2.5%. Sounds quite low to me? | 17:59 |
| vkoskiv | The TPS61185 datasheet indicates a minimum time on of 5 microseconds for the PWM signal. With the current setup, max brightness is 6µs, and then it goes down in 1µs increments from there, down to 0. | 18:03 |
| - GNUmoon2 (QUIT: Remote host closed the connection) (~GNUmoon@gateway/tor-sasl/gnumoon) | 18:04 | |
| + GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 18:04 | |
| vkoskiv | So at 0 brighness, the PWM signal looks correct, but the LEDs are still seeing a steady DC voltage of 6.7V. I'm not sure if TI expects the EN pin to be used to fully turn off power to the LEDs. It's surprising to me that a 0% PWM signal would be interpreted as anything but 'turn output off' | 18:19 |
| vkoskiv | I guess the mention that the minimum PWM time is 5µs is the clue that PWM can't be used to fully disable the output. Though our lowest brightness levels *are* below this 5µs limit, and the chip does still respond to that with varying brightness levels, so *shrug* | 18:22 |
| + nybble (~sprang@user/csprng) | 18:52 | |
| vkoskiv | josch: I added my findings to https://source.mnt.re/reform/reform/-/issues/20 | 19:14 |
| josch | thank you for investigating! :) | 19:14 |
| josch | vkoskiv: how was this not a problem before your change? | 19:15 |
| vkoskiv | Though do keep in mind that I am a computer programmer with an oscilloscope that is better than what I need :] | 19:15 |
| josch | classic, no? ;) | 19:16 |
| vkoskiv | Excellent question, I should investigate that next. I changed the PWM frequency to be within spec, so I'm guessing we traded one fault condition (jumpy backlight adjustment) for another one, which is that the LEDs never fully turn off at 0% brightness. | 19:16 |
| vkoskiv | Another regression with my patch is that the PWM frequency went down into an audible range, so the ceramic cap and/or inductor now make noise | 19:19 |
| vkoskiv | Well, noise within the range of human hearing. It always made noise, maybe probably :D | 19:20 |
| + mgd (~mgd@82-132-220-192.dab.02.net) | 20:05 | |
| - mgd (QUIT: Remote host closed the connection) (~mgd@82-132-220-192.dab.02.net) | 20:29 | |
| + chomwitt (~chomwitt@2a02:587:7a10:cf00:1ac0:4dff:fedb:a3f1) | 21:41 | |
| + cobra (~cobra@user/Cobra) | 22:21 | |
| + jacobk (~quassel@2600:1700:9e1e:7800:c40c:5464:17f4:498c) | 23:14 | |
| - GNUmoon2 (QUIT: Ping timeout: 260 seconds) (~GNUmoon@gateway/tor-sasl/gnumoon) | 23:22 | |
| + GNUmoon2 (~GNUmoon@gateway/tor-sasl/gnumoon) | 23:23 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!