ChanServ changed the topic of #glasgow to: glasgow interface explorer · code https://github.com/GlasgowEmbedded/glasgow · logs https://freenode.irclog.whitequark.org/glasgow · discord https://1bitsquared.com/pages/chat · production https://www.crowdsupply.com/1bitsquared/glasgow · CrowdSupply campaign is LIVE!
ali_as has quit [Remote host closed the connection]
Stormwind_mobile has quit [Ping timeout: 256 seconds]
happycube has quit [Ping timeout: 260 seconds]
happycube has joined #glasgow
GNUmoon2 has quit [Remote host closed the connection]
GNUmoon has joined #glasgow
electronic_eel_ has joined #glasgow
electronic_eel has quit [Ping timeout: 264 seconds]
GNUmoon has quit [Remote host closed the connection]
GNUmoon has joined #glasgow
PyroPeter_ has joined #glasgow
PyroPeter_ is now known as PyroPeter
PyroPeter has quit [Ping timeout: 260 seconds]
<marcan> whitequark: we don't have any TI parts, do we?
<marcan> because
<marcan> 13:37:14 < davidrysk[m]> note, TI has been dmca-happy lately
<marcan> (apprently DMCAing datasheets)
<marcan> that alone DQ's them from any design I ever make going forward
<d1b2> <you_snus_you_lose> glasgow uses a ton of TI parts last time i checked
<d1b2> <you_snus_you_lose> well it's still better than manufacturers that just don't publish any datasheets in the first place
<d1b2> <Darius> if they are DMCAing them from the low quality datasheet.com places which make downloading them a PITA fine by me 😕
<d1b2> <you_snus_you_lose> you should try to choose parts that have a substitute
<d1b2> <you_snus_you_lose> TPS73101 is fine because i can find plenty of pin compatible substitutes
<d1b2> <you_snus_you_lose> but INA233 is less ok because it's single sourced
<d1b2> <you_snus_you_lose> i really try to avoid these unicorn TI parts
<d1b2> <Darius> it's hardly unique to Ti
<d1b2> <Darius> lots of mfgs have very convenient/compact/featureful devices only they make
<d1b2> <you_snus_you_lose> i don't use those
<d1b2> <you_snus_you_lose> should take a look at how chinese consumer electronics are designed
<d1b2> <Darius> well sure if you want your device to have fewer features, be more expensive and larger..
<d1b2> <you_snus_you_lose> even when there is a "nice" single chip solution available they opt for circuits based on discrete transistors
<d1b2> <you_snus_you_lose> i'm sure it's cheaper is the main reason they do it this way
<d1b2> <you_snus_you_lose> i looked at a LCD driver board that drives those 50pin rgb interface lcd panels
<d1b2> <Darius> probably also easier to design...
<d1b2> <you_snus_you_lose> those lcd panels require a bunch bias voltages, 10V, 19V, -6V, etc,
<d1b2> <you_snus_you_lose> but instead of using a chip that does all that for you, the circuit on that driver board is some ordinary buck or boost IC, with a bunch of diode and capacitor voltage multipliers
<d1b2> <you_snus_you_lose> it's really clever
<d1b2> <you_snus_you_lose> they really don't want to pay big bucks to IC manufacturers
<d1b2> <you_snus_you_lose> that entire driver board costed me like 40cny, or like $7
<d1b2> <you_snus_you_lose> not gonna achieve that if you are using $5 chips
<d1b2> <you_snus_you_lose> what i'm saying is they really go out of their way and put in extra engineering just to get the bom cost down a little or make the design use only generic parts
<d1b2> <you_snus_you_lose> i don't think "easy to design" ever comes across the minds of these engineers
<d1b2> <you_snus_you_lose> i mean if you want it "easy to design" why did you become an engineer LOL
<d1b2> <Hardkrash> Time, fast, money.... pick 2. Different engineers have different requirements. I've built systems using hand wound bi-filiar inductors individually tuned, I've built systems that use expensive drop on power converters, and I've built systems were every fraction of a cent cost was scrutinized. There is no one "true" engineer.
<d1b2> <Darius> yeah, exactly
<d1b2> <esden> @you_snus_you_lose I think you are focusing on the wrong aspects. I understand that you see it from the chinese market perspective but the hardware we build is not that.
<d1b2> <esden> Also the ad hominem "why did you become an engineer LOL" is not a way we talk to each other here.
<d1b2> <esden> Please stop that.
<d1b2> <Darius> does anyone know if the probe kit will be available separately?
<d1b2> <Darius> I already have a Glasgow but A Real Probe Kit (tm) would be nice
<d1b2> <esden> @Hardkrash indeed, there is always a compromise, and everyone has different priorities. It is important to understand that. 🙂
<d1b2> <esden> @Darius do you mean like a flywire harness and chip clips?
<d1b2> <Darius> the flywire clips
<d1b2> <edbordin> I had also been meaning to ask that. I was planning to build up a board and was considering at least buying the nice case for it, would I be missing out without the wire harness etc?
<d1b2> <Darius> the USB C cable wont be much good for my micro USB version 😄
<d1b2> <esden> So the harnesses we made for the glasgow will be available after the campaign as replacement and for those that already have Glasgow.
<d1b2> <Darius> OK great
<d1b2> <edbordin> sweet 🙂
<d1b2> <Darius> the case looks awesome too, looking to grab one 🙂
<d1b2> <esden> That said, there are chip clips you can get and flywires.
<d1b2> <Darius> I have some junky flywires
<d1b2> <esden> Let me look them up for you.
<d1b2> <Darius> danke
<d1b2> <esden> First option is EZ-Hook XKM https://e-z-hook.com/test-hooks/micro-hook/xkm-micro-test-hook-double-gripper-with-0-025-in-square-pins/ the 10 assorted color set for $29 is really nice, this is what you see in the campaign video. 😄 They are very good quality and pretty colors. 🙂
<d1b2> <esden> the second option is a set of Hantek 18pcs/lot Mini Test Hook HT321 of which 2 are black and the rest is gray. They are also pretty decent quality. You can find them on several websites. We also resell them through 1BitSquared https://1bitsquared.com/products/bitmagic-mini-test-hooks
<d1b2> <esden> As for flywires, I don't have alternatives that I can point to from the top of my head. The biggest issue is to find some that have proper copper wire, not copper coated aluminium ... (yes that is what the really cheap flywire sets are made from ... they are pretty bad)
<d1b2> <esden> @Darius I hope this is useful. 🙂
<d1b2> <Darius> cool thanks
<d1b2> <Darius> lol amazing
bvernoux has quit [Quit: Leaving]
bvernoux has joined #glasgow
GNUmoon has quit [Ping timeout: 240 seconds]
<d1b2> <Hardkrash> I wonder if there are any out there with the "wet noodle" silicone wires.
<d1b2> <esden> heh... I have not seen any yet... I did think of making a product like that a while ago... but never came around to actually do it. Maybe I should 😄
<d1b2> <esden> ohh adafruit seems to make them? https://www.adafruit.com/product/4635
<d1b2> <esden> Yeah it seems like it. Also all the recepticle and pin combinations.
<d1b2> <Darius> "wet noodle"?
<jpa-> i have some similar silicone wires from ebay - they are good, more flexible and thicker wire than regular jumper wires
<kc8apf> Silicone insulated wire tends to be so flexible that the wire flops around like a cooked spaghetti noodle
<d1b2> <Hardkrash> It is the name of W S Deans, 12 AWG with 1660 strand count. Popular in the RC car hobby world as very flexible for 12 AWG.
<d1b2> <j4cbo> it’s not just silicone; the “wet noodle” ones are made of a ridiculously high number of strands
<d1b2> <Darius> ahh cool
<smkz> my favorite flavor of wire is spec 44 (PVDF) / spec 55 (ETFE), it's far more pleasant to work with (it's easy to bend but not floppy like silicone) than either PVC or silicone, and the mechanical/electrical properties of the insulation means they can make it quite thin :3
<d1b2> <Hardkrash> I also believe the wet noodle would be a straight lay bundles vs twisted bundle.
<d1b2> <j4cbo> the stuff I got recently was definitely lightly twisted
<d1b2> <j4cbo> but that was off-brand high strand count silicone, not name brand Wet Noodle
<d1b2> <Hardkrash> The PVFD is intersting, very thin.
<smkz> i've made wire harnesses with spec44/spec55 and they look really nice, i've never actually used it for aerospace applications though >_> https://twitter.com/alt_kia/status/1228151716366827520
<d1b2> <j4cbo> there is dispute among experimental aircraft builders as to whether it’s worth the $, aiui
<d1b2> <j4cbo> but it sure looks good
<d1b2> <Hardkrash> I've been using mPPE hookup wire and cables. recyclable and thin. e.g. Alpha EcoWire+
<d1b2> <Hardkrash> Those do look nice.
GNUmoon has joined #glasgow
<marcan> ugh, I hadn't realized the INA233 was TI
<marcan> I knew about the shifters / LDOs
<marcan> but those are less unicorn
<marcan> Darius: they are DMCAing them from archival places, not unlike the goals of GlasgowEmbedded/archive
<cyborg_ar> lame
<gruetzkopf> i found them DMCAing them from distrelec (a fairly well-known german distributor) "interesting"
<TD-Linux> I got some to play with, it is indeed very nice and very hard to abrade
<TD-Linux> the abrasion and crush resistance is pretty unique, a lot of the other exotic wire choices (ptfe, silicone) are really no better or even worse at it
<TD-Linux> they also make a high strand count variant called flexline
DarthCloud has quit [Ping timeout: 240 seconds]
DarthCloud has joined #glasgow
egg|anbo|egg_ has joined #glasgow
egg|anbo|egg has quit [Ping timeout: 260 seconds]
electronic_eel_ has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
electronic_eel has joined #glasgow
<d1b2> <Attie> parts are here for the CAN addon!
<gruetzkopf> neat
<apo> TD-Linux: Want me to give it a try? I can be pretty abrasive
jstein has joined #glasgow
Stephie has quit [Ping timeout: 260 seconds]
<_whitenotifier> [glasgow] whitequark closed pull request #261: gateware.uart: fixup parity handling - https://git.io/JLxYL
<_whitenotifier> [GlasgowEmbedded/glasgow] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/JLpaz
<_whitenotifier> [GlasgowEmbedded/glasgow] attie-argentum 2c5295e - gateware.uart: fixup parity handling
<smkz> TD-Linux: Spec 44 / Spec 55? yeah abrasion resistance is important there because vibration (also they make special dual-layer insulated versions where the inner insulation layer has a contrasting color to make inspections easier)
<smkz> also if you want to learn more about exotic wire choices here's a neat pdf https://docs.rs-online.com/96af/A700000007028901.pdf
<whitequark> marcan: we have a fuckton of TI parts and we'll keep using them because unlike Maxim parts they actually work
dominik[m]1 has joined #glasgow
<sorear> do we have secure copies of the documentation for all TI parts used on glasgow?
<whitequark> not yet, but electronic_eel has opened an issue just to that extent two days ago
<whitequark> someone whose life isn't shit should spend time collecting them
<whitequark> (i.e. not me)
<kmc> what needs to be done?
<kmc> link t othe issue?
dominik[m]1 has left #glasgow ["User left"]
nicoo has quit [Remote host closed the connection]
<kmc> what are the criteria for including applets in the glasgow source tree? if I write little applets to talk to various sensors that I have, or whatever, should I do so with an eye towards merging them?
<gruetzkopf> i have some spec55 wiring in railway gear because that was made by the same department of philips that also built RADARs and fire control systems
<russss> if TI sends a takedown, I have a few contacts within TI. Bit surprised that they care at all tbh
<electronic_eel> TI seems to be a bit strange. on the one hand they want to be the only ones hosting the datasheets, on the other hand they'll be deleting their processors wiki soon. i guess the latter was a wakeup-call for many to not rely on vendors keeping docs around
<electronic_eel> kmc: re merging applets - this is not really fleshed out yet. on the one hand it would be nice to have everything in one common source tree, on the other hand adding a lot of applets there will create substantial maintenance burden
<electronic_eel> so the current plan is to create infrastructure in the glasgow software to allow multiple source trees. that would allow creating a contrib-repo with not as strict merge rules (and lesser usability promises to the users)
nicoo has joined #glasgow
<electronic_eel> but of course it would still make sense to merge mature applets that appeal to a bigger crowd into the main repo
<electronic_eel> so for now i would just plan to eventually get your applet merged and write it as such. then send a PR and see what happens.
<kmc> okay, thanks
<electronic_eel> kmc: what kind of applet do you have in mind to write?
Stephie has joined #glasgow
<kmc> well for example i have a lot of sensors (DHT11 temp/humidity sensor, DS18B20 1-Wire temp sensor, BME680 temp/humidity/pressure/gas, PMSA003I particulate sensor, accelerometers, IR rangefinder, etc)
* sorear remembers back when version control was less common and a lot of projects had a contrib/ directory in the main distribution with explicit "if it breaks you get to keep both pieces" rules
<kmc> the BME680 is a relative of the BME280 which is already in tree, perhaps the driver could be expanded to support both, I haven't looked into how similar they are actually
<electronic_eel> kmc: afaik there is no 1-wire protocol in glasgow yet. i think getting 1-wire capability (at least host) would be interesting and definitely something for the main repo
<electronic_eel> the BME-sensors are quite similar iirc, so they could be merged into one applet
<electronic_eel> so i think all you mentioned has the potential to go into mainline
<electronic_eel> whitequark has quite a high bar for code quality though, also she is a bit overworked so it can take quite a while for getting stuff reviewed
<electronic_eel> also she might ask you to send her demo boards, so that she is able to test the applet after some changes a few years down the road
GNUmoon has quit [Remote host closed the connection]
GNUmoon has joined #glasgow
<whitequark> kmc: all of those are in-scope for having in-tree
<whitequark> re 1-wire, i started doing it already in a branch
bvernoux has quit [Quit: Leaving]
<d1b2> <Attie> kmc: depending on what you're trying to achieve, sensors that use a supported bus (e.g: I2C) could be implemented as a simple script too
<_whitenotifier> [glasgow] attie opened pull request #262: examples: fixup PCF8574 usage - https://git.io/JLpyD
<kmc> cool
<tnt> Is there an easy way for an applet to control the power sequencing / initial pin state ? One thing I've been hitting "recently" is for an ice40-flash programming, I need the reset line to be asserted low before power is applied. If at any point the fpga is given the opportunity to "boot", it will execute the bootloader code which locks the flash until power down.
<sorear> possible to power the fpga off an IO or too much current?
<tnt> Well, this was more about how that kind of requirement can be cleanly expressed in an applet. Because as work around so far I've just tied reset to GND pin ...
<d1b2> <Attie> that might be tricky - because the level shifters run off the same supply as the port's Vout
<d1b2> <Attie> you might get a little glitch oun power up
<d1b2> <Attie> but otherwise, i think an applet has control over power... let me dig quickly
<tnt> Well glitch is not too bad, the fpga needs quite a while to boot ...
<tnt> I mean "a while" for the logic, it's insanely short from a human PoV ...
<d1b2> <Attie> yeah...
<d1b2> <Attie> you could try splicing that into the applet's run() and probably interact() phases, to power down, and re-power when you're ready
<d1b2> <Attie> (i.e: await device.set_voltage(port, 0) ... setup / get ready ... await device.set_voltage(port, 3.3))
<tnt> Yup I should give that a try.
trh has quit [Quit: weg]
trh has joined #glasgow
ender| has quit [Quit: “I could sulk,” he said. “I have a sulking subroutine.” — Brandon Sanderson: Starsight]