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 · no ETAs at the moment
Stormwind_mobile has quit [Ping timeout: 264 seconds]
Stormwind_mobile has joined #glasgow
Foxyloxy has quit [Ping timeout: 256 seconds]
Foxyloxy has joined #glasgow
Getorix_ has quit [Ping timeout: 264 seconds]
nmz787 has quit [Ping timeout: 256 seconds]
nmz787 has joined #glasgow
electronic_eel has quit [Ping timeout: 265 seconds]
electronic_eel has joined #glasgow
PyroPeter_ has joined #glasgow
PyroPeter has quit [Ping timeout: 264 seconds]
PyroPeter_ is now known as PyroPeter
_whitelogger has joined #glasgow
Stormwind_mobile has quit [Remote host closed the connection]
Stormwind_mobile has joined #glasgow
Getorix has joined #glasgow
Getorix_ has joined #glasgow
Getorix has quit [Ping timeout: 240 seconds]
Exec1N has joined #glasgow
Exec1N has quit [Ping timeout: 265 seconds]
Exec1N has joined #glasgow
Exec1N has quit [Ping timeout: 256 seconds]
Exec1N has joined #glasgow
Exec1N has quit [Ping timeout: 246 seconds]
<d1b2> <EmilJ> man, the coverage this is getting now is cool to see
Exec1N has joined #glasgow
Exec1N has quit [Ping timeout: 240 seconds]
electronic_eel has quit [Ping timeout: 256 seconds]
electronic_eel has joined #glasgow
Stormwind_mobile has quit [Remote host closed the connection]
Stormwind_mobile has joined #glasgow
Sellerie has quit [Quit: The Lounge - https://thelounge.chat]
Sellerie has joined #glasgow
Sellerie has quit [Client Quit]
Sellerie0 has joined #glasgow
Stormwind_mobile has quit [Ping timeout: 264 seconds]
m4ssi has joined #glasgow
<d1b2> <Crobisaur> hello
<d1b2> <Crobisaur> I'm looking to build one myself but oshpark is pricey for 3 boards ~$60 US has anyone been making these in batches per chance?
<d1b2> <Crobisaur> if not no worries, I just thought I should ask before I send an order in
<d1b2> <TiltMeSenpai> there's a few people who have run batches, I think the problem you'll run into is those people might not be geographically close you you
<d1b2> <TiltMeSenpai> if you wanna buy from someone who has bare pcb's you have to deal with shipping and stuff
<d1b2> <atx> @crobi (she/they) Someone bought PCBs from JLC and got working boards
<d1b2> <TiltMeSenpai> wasn't the JLC run modified? can't remember if that was for part availability or to meet manufacturing tolerances
<d1b2> <atx> @TiltMeSenpai @noopwafel ordered unmodified and @Twix ordered modified
<d1b2> <atx> Not sure why
<d1b2> <TiltMeSenpai> oh, there were 2 runs
m4ssi has quit [Remote host closed the connection]
<d1b2> <atx> I think it just barely meets the specs
<d1b2> <TiltMeSenpai> I know someone modified for parts available on LCSC
<d1b2> <TiltMeSenpai> that was probably Twix
<noopwafel> yes, I just bought bare PCBs
<d1b2> <TiltMeSenpai> it's 4 layer right?
<noopwafel> if anyone's in Europe, I do have spare revC1 boards
<noopwafel> I mean, PCBs :p
<d1b2> <atx> Note that revC2 is in progress
<d1b2> <atx> So maybe it's worth waiting for that
<noopwafel> yes, 2-3 weeks said electronic_eel
<Twix> <d1b2> <TiltMeSenpai> that was probably Twix <-- nope, i just modified the vias and changed the micro usb to usb-c.
<d1b2> <gimbas> I have some slightly modified revC1, Portugal - Europe
<Twix> i used the original parts and placed them by hand :)
<d1b2> <TiltMeSenpai> ah, ok
<d1b2> <atx> The JLC PCBA can only do one sided assembly anyway
<noopwafel> what did you change on the vias?
<noopwafel> I put the jlcpcb ones under microscope and e.g. BGA looks fine
<Twix> according to the diff, i made the drill size smaller. I don't remember why o_Ö
<d1b2> <atx> iirc the via to pad spacing on the BGA is sort of close
<d1b2> <TiltMeSenpai> JLC annular ring limits maybe?
<Twix> maybe
<d1b2> <crobi (she/they)> I guess is it a challenge to assemble this thing by hand? I figured it would be somewhat do-able
<d1b2> <crobi (she/they)> also I'm in the states :c
<d1b2> <TiltMeSenpai> I know the ecp5 designs I've done have had to use the minimum drill size to get vias between pads while meeting annular ring limits
<Twix> no, it wasn't a challenge to assemble it, just time consuming
<Twix> i took 8 hours or something for the first one
<d1b2> <TiltMeSenpai> @crobi (she/they) if the glasgow pcb's are 4 layer, JLC's 4 layer service got a whole lot cheaper recently
<d1b2> <crobi (she/they)> oh really? good to know ty
<d1b2> <atx> iirc it's like 7$ for 5 boards or something
<d1b2> <TiltMeSenpai> it would probably be around $30-40ish for 5 pcb's + a stencil (remember to set a custom size on the stencil)
<noopwafel> it was my first attempt to do a board with reflow oven, and my first attempt worked ok, it was just super time consuming, and I would want the stencil
<Twix> The hardest parts were the resistor arrays. Maybe i took too much solderpaste
<d1b2> <crobi (she/they)> oooh yeah a stencil would be super handy, idk if the oshpark 4layer offers a stencil or not
<d1b2> <TiltMeSenpai> they ship you a pretty big stencil by default, but you can set a custom size, which helps the shipping costs a lot (this is JLC)
<Twix> yes, i also have stencils
<d1b2> <crobi (she/they)> I have a reflow station and a hotplate/stovetop lol
<d1b2> <TiltMeSenpai> plus they round the corners and deburr the edges, so you won't cut yourself on them as easily
<Twix> <-- cologne/germany if someone wants one of mine :)
<d1b2> <crobi (she/they)> I'm still putting the BoM together atm, probably won't have budget to order for another month or so. Ty for the help on manufacturers ^-^ I'm still very new to hardware building
<d1b2> <TiltMeSenpai> glasgow is great hardware, and I love mine. Kinda a lot of parts though, honestly assembling one myself would be a bit of an ambitious project. Good luck!
<d1b2> <crobi (she/they)> I figure it would be a learning experience for sure, I have some decent soldering experience, smt is still new to me atm but I've had a good bit of practice from work recently
<d1b2> <crobi (she/they)> just trying to do it in my home lab is gonna be a challenge haha
<d1b2> <atx> @crobi (she/they) It does have a BGA, which might be tricky if you never soldered one before
<d1b2> <crobi (she/they)> oh really? 😬 gosh
<d1b2> <TiltMeSenpai> it's like .8mm pitch though, it's not too bad
<d1b2> <atx> @TiltMeSenpai Wait, isn't it 0.4mm?
<Twix> nope, 0.8 mm
<whitequark> yeah, i've actually tried to not use a BGA, this was a hard requirement in the revA/revB times
<d1b2> <TiltMeSenpai> I'm pretty sure it's .8, <.5mm is hard to route on JLC tolerances, and doesn't work for more than like 4x4
<d1b2> <atx> Oh, 0.8 actually
<whitequark> and it ended really badly
<whitequark> so for revC i just went with a single coarse-pitch BGA *shrug*
<d1b2> <crobi (she/they)> I guess the challenge would be getting the balls to sit still long enough flow them haha
<d1b2> <atx> @whitequark Not many non-BGA FPGAs around
<whitequark> atx: well, iCE40 has quite a few
<d1b2> <TiltMeSenpai> you can't escape <.5ish mm without via in pad I think (or it's super hard and requires substantial creativity to do so)
<Twix> oh yes, i had one project with the tqfp-144 variant of the ice40 4k. The PCB manufacturer could not have solder resist between the pins. So the chip had a lot of bridges and rotated during reflow. It was horrible
<d1b2> <noopwafel> jlcpcb glasgow BGA
<d1b2> <crobi> oh dear
<whitequark> Twix: that's *exactly* why i'm using a BGA
<whitequark> it's also gigantic
<whitequark> but... large pin count TQFPs are even harder to solder than coarse-pitch BGAs
<whitequark> like physical size notwithstanding
<d1b2> <TiltMeSenpai> meh, .8mm is just about as much in your favor as you can get for a bga
<d1b2> <crobi> .8mm does seem doable but I'm probably gonna have to think of a clever magnification solution
<d1b2> <atx> @crobi you can't see under the BGA anyway
<d1b2> <crobi> fair point
<d1b2> <TiltMeSenpai> nah, you see pads around the edges by the time you're off by a whole row
<d1b2> <TiltMeSenpai> and all the pads are surrounded by solder mask, so surface tension works in your favor
<d1b2> <crobi> is there some sort of jig to get the ball grid to line up? I should probably look into how to do one of these by hand
<d1b2> <atx> @crobi Some people stick kapton tape around the footprint
<d1b2> <atx> Which makes it less likely to glide around
<d1b2> <crobi> ooh that's a good idea
<d1b2> <TiltMeSenpai> I use loctite gc10 solder paste, just drop the bga as close as I can and gently push it around until it lines up
<Twix> i placed it by hand. I placed it at first so i could clean up and start again if i would have missed the footprint. But it was really no problem to place it free handed
<d1b2> <atx> Worst thing is, if you fuck up you have to use a new chip
<d1b2> <atx> You can technically reball, but that is more effort than it's worth for 7$ part
<d1b2> <crobi> I've reflowed bga chips in the past just never assembled one
<whitequark> it's not that hard to reball a BGA if you have a stencil
<whitequark> if you *don't* have a stencil though
<sorear> how close are we to someone distributing assembled revC or edinburgh boards?
<d1b2> <atx> @whitequark I used one of those reballing jigs in the past
<d1b2> <crobi> yeah a stencil would be a good idea
<d1b2> <atx> And I just could not get it to work
<d1b2> <atx> Heating it up made the stencil flex
<d1b2> <atx> Which fucked up the alignment
<whitequark> atx: yeah, same, actually
<whitequark> it seems that you have to remove the stencil before heating
<whitequark> but this requires having a flux with very specific properties
<d1b2> <atx> Tried that too, my hands are way too shaky
<d1b2> <atx> Maybe there are fancier reballing jigs which can be lowered by a lever or something
<d1b2> <TiltMeSenpai> hmm I wonder if you could temper the stencil or something to stop it from flexing when you heat it
<d1b2> <TiltMeSenpai> this is heading rapidly outside the realm of practicality but
<d1b2> <crobi> I figure having a hotplate to keep things warm might help let the stencil settle
<d1b2> <TiltMeSenpai> the JLC stencils have some stress in them or something, I've tried baking moisture off them a couple times and they warp a bit every time I've tried that
<d1b2> <TiltMeSenpai> this is at ~100-120C, so not even close to reflow temps. I assume they'd warp even worse at those temps
<whitequark> TiltMeSenpai: i think that's just stainless
<whitequark> horrible material to heat
<d1b2> <TiltMeSenpai> yeah that wouldn't surprise me
<d1b2> <TiltMeSenpai> would you have better luck with a polyimide stencil?
<whitequark> unsure
Stormwind_mobile has joined #glasgow
TiltMeSenpai has joined #glasgow
TiltMeSenpai has quit [Quit: ZNC 1.7.5+deb4 - https://znc.in]
TiltMeSenpai has joined #glasgow
TiltMeSenpai has quit [Client Quit]
TiltMeSenpai has joined #glasgow
TiltMeSenpai has quit [Client Quit]
TiltMeSenpai has joined #glasgow
m4ssi has joined #glasgow
m4ssi has quit [Remote host closed the connection]
Stormwind_mobile has quit [Ping timeout: 265 seconds]
<awygle> polyimide should be fairly dimensionally stable under heat
<tnt> The reballing stencils I have don't warp (well not much, not enough to be an issue) so I can reflow the balls with the stencils on it. I tried making my own with JLC or aisler but those warp insanely and not usable. They must be using some different alloy than good reballing stencils.
<tnt> They're also much thinnner than "dedicated purpose" reballing stencils.
<tnt> I tried using them to just apply paste to form balls, but the balls were definitely undersized. Still managed to solder the reballed ECP5 with them but it's nearly stuck to the pcb rather than having a bit of space under it.
Stormwind_mobile has joined #glasgow
Sellerie0 is now known as Sellerie
PyroPeter has quit [Ping timeout: 264 seconds]
PyroPeter has joined #glasgow