<azonenberg>
i make heavy use of filled via-in-pad on most of my multilayer designs
<azonenberg>
among other things it makes decoupling big bgas super nice
Degi has quit [Ping timeout: 258 seconds]
Degi has joined #scopehal
<azonenberg>
v0.2 MEAD enclosure is a no go. The opening in the front for the MMCX's is still too low
<azonenberg>
it looks good otherwise and i think i can machine it to fit. definitely better overall than the last round
<monochroma>
azonenberg should get one of the cheap printers for fast turn around size/fitness testing
<azonenberg>
Milled it out, ready to assemble
<azonenberg>
I need to print another enclosure for the next pcb rev anyway
<azonenberg>
Ok so after assembly... the 3m adhesive transfer tape i used doesnt seem strong enough, i'll have to think a bit about what kind of glue to use to hold the LCDs to the lids in the future
<azonenberg>
The lid also flexes slightly but that will be a non issue in the metal version
<azonenberg>
monochroma, lain: see phone pic
<azonenberg>
Looks like the opening on the front has to move up another 2 mm or so
<lain>
aye
electronic_eel has quit [Ping timeout: 258 seconds]
electronic_eel has joined #scopehal
<azonenberg>
The fit is also a bit tight
<azonenberg>
i'm gonna make the enclosure a mm or so deeper
<monochroma>
have you verified the dimensions are to spec?
<azonenberg>
They look right. I think i screwed up my math
<azonenberg>
everything else is perfect
<azonenberg>
The big problem is i cant fit test the pcb against the enclosure model
<azonenberg>
because kicad can't import IDF models yet
<azonenberg>
solvespace can't*
<azonenberg>
I need to do another shapeways order soon anyway for the final probe shells
<azonenberg>
prototype came in today and i need to fit test against the board
<azonenberg>
if it looks good i need to order another 18 shells
<azonenberg>
Hmm there's some loose nylon powder stuck inside
<azonenberg>
let me try sonicating it
<azonenberg>
welp, now my cleaning bath has a whole bunch of nylon particles floating in it but they're not quite cleared. Wonder what happened with this batch that didn't happen on the previous ones
<azonenberg>
3 mins might not have been enough, throwing 'em back in for longer
<azonenberg>
Well i cleaned them out enough. hope this doesn't happen on the rest of them
<azonenberg>
Anyway, production batch of probe shells ordered along with the hopefully final MEAD enclosure
LeoBodnar has quit [Ping timeout: 256 seconds]
kc8apf has quit [Ping timeout: 256 seconds]
asy has quit [Ping timeout: 256 seconds]
kc8apf_ has joined #scopehal
LeoBodnar has joined #scopehal
asy_ has joined #scopehal
kc8apf_ is now known as kc8apf
_whitelogger has joined #scopehal
<_whitenotifier-b>
[starshipraider] azonenberg pushed 1 commit to master [+0/-0/±2] https://git.io/JJYBk
<_whitenotifier-b>
[starshipraider] azonenberg 0a021ca - More routing of right side of big FPGA
_whitelogger has joined #scopehal
maartenBE has quit [Ping timeout: 256 seconds]
maartenBE has joined #scopehal
Ekho has quit [Ping timeout: 244 seconds]
Ekho has joined #scopehal
<_whitenotifier-b>
[starshipraider] azonenberg pushed 1 commit to master [+0/-0/±19] https://git.io/JJY26
<_whitenotifier-b>
[starshipraider] azonenberg 1d124a9 - Finished preliminary routing of pod FPGA
<azonenberg>
pod FPGA is now fully fanned out, and all of the pod IOs are routed. The long-range routing for prog_b, init_b, done, and the SPI bus still has to be done but that's easy
<pepijndevos>
Dumb Q: can you use a 1:1 probe as an output? Or do I need to get some BNC cables and converters hehe
<azonenberg>
Left 7 pods still have to be routed to the main FPGA, main FPGA boot flash isn't done, some bulk caps for the main FPGA still have to be done
<azonenberg>
then the muxes for the fast inputs
<azonenberg>
pepijndevos: depends on how the probe is built
<azonenberg>
if it's a 1M ohm passive probe, no
<azonenberg>
if it's a transmission line probe, yes
<azonenberg>
if it's an active probe, definitely no :p
<pepijndevos>
lol
<azonenberg>
well ok to be technical a 1M ohm passive probe could be used as an output with extremely weak drive strength, but it wouldn't be useful in 99% of applications
<azonenberg>
a 1x transmission line probe otoh is just a piece of coax with a needle on one end
<azonenberg>
so it works equally well both ways
<pepijndevos>
It's a passive probe, so much I know. Not sure about the impedance. It's whatever came with the scope. Lemme see...
<azonenberg>
Unless it's a specialized high speed probe, 1M || ~10 pF is a reasonable assumption
<azonenberg>
You're not gonna be using that as an output
<pepijndevos>
I figured that would likely be the case hehe
<pepijndevos>
Oh yea, the leaflet mentions it's 1M impedance
<azonenberg>
For reference that "1:1" really means when connected to a scope with 1M ohm input impedance
<monochroma>
azonenberg: what's that pair of traces routing under the SAS connector shells?
<monochroma>
nice progress :D
<pepijndevos>
ERROR: Error: Failed to initialize GLEW
<monochroma>
pepijndevos: are you trying to run inside of a VM?
<azonenberg>
monochroma: that's the 0V5 enable/good signals to the MCU
<azonenberg>
i couldn't find a good way to route it any other way
<pepijndevos>
Ah, I know... I updated and did not reboot yet.
<pepijndevos>
That sometimes does weird things to my graphics
<azonenberg>
pepijndevos: oh so your graphics driver and libs are probably out of sync
<azonenberg>
nvidia?
<pepijndevos>
yah
<monochroma>
ahhhh
<monochroma>
yea
<azonenberg>
Yeah that happens
<monochroma>
azonenberg: open GPU when? ;)
<azonenberg>
monochroma: hey, let's fix one industry at a time. I already have LATENTPACKET back-burnered while i work on this stuff
<monochroma>
hehehe
<pepijndevos>
uh... still increddibly screwy and slow
<pepijndevos>
Basically now it just runs very slow and does not update the screen.
<pepijndevos>
So I press >|| and I see it trigger on the scope, but the app stays empty
<azonenberg>
you might be hitting the same bug whitequark was. Rigol?
<pepijndevos>
yep
<azonenberg>
DS1000Z series?
<pepijndevos>
No, 5000
<azonenberg>
Interesting
<pepijndevos>
It worked earlier, so... seems like something weird with graphics drivers maybe
<azonenberg>
I'm going to be going to sleep soon (yes it's six in the morning, yes my sleep schedule is totally wrecked thanks to not going into the office since march)
<azonenberg>
could also be a recent regression
<azonenberg>
or same binaries/code rev?
<pepijndevos>
Yea, I'm still on the same branch I tested my force trigger patch with
<azonenberg>
Huh
<pepijndevos>
Not really sure how to even start debugging this. There is no error or crash or anything, just... UI does not update
<pepijndevos>
I'll go buy some BNC cables on Farnell
<pepijndevos>
Sleep well ;)
bvernoux has joined #scopehal
Bird|ghosted has quit [Read error: Connection reset by peer]