| *** tpb has joined #photonsdi | 00:00 | |
| *** Bertl_oO is now known as Bertl_zZ | 06:09 | |
| Kjetil | How doe sthe cap. placement affect the panelizablity of the PCBs ? | 10:38 |
|---|---|---|
| felix_ | well, if the ceramic capacitors are too cloese to a v-cut or mousebite and the boards get separated after being populated, the ceramic capacitors get quite a bit of mechanical stress which may cause failure and those ofter fail with a short | 11:10 |
| felix_ | *often | 11:10 |
| Kjetil | yeah. But will typically have frame around the board for the pick and place to grab onto anyway even if it is single PCBs per panel | 11:11 |
| felix_ | the axiom pcbs i saw at the office didn't have a frame around them and also weren't in a panel | 11:15 |
| felix_ | i usually don't place anything too close to the board edge, but this board was rather space-constrained... | 11:18 |
| *** Bertl_zZ is now known as Bertl | 13:24 | |
| Bertl | it's not critical, even if we do mouse bites (for a panel) we do it in such way that it can be depanelized without much stress for the boards | 13:26 |
| felix_ | ok | 13:28 |
| felix_ | just wanted to make sure before the boards get manufactured | 13:28 |
| felix_ | now the gtp power supplies are routed and the footprints of the linear regulator is fixed | 14:22 |
| felix_ | the footprint of the switch mode regulator still needs some patching and then the 75 ohm parts need to be routed | 14:22 |
| felix_ | now only the 75 ohm parts are still unrouted | 15:46 |
| felix_ | https://i.imgur.com/Ufjuoeb.jpg | 15:47 |
| Kjetil | remember to void the ground plane underneath the 75ohm part | 16:47 |
| felix_ | huh, no ground plane? i'm still undecided if i try to fit a coplanar waveguide configuration in there or just route the few millimeters as microstrip line. both heve a ground plane beneath | 17:09 |
| Kjetil | There should be a ground plane but probably not in the next layer | 17:10 |
| felix_ | sure, add a cutout benaeth the capacitor and pin of the hdbnc connector, to avoid too big impedance mismatch | 17:10 |
| Kjetil | yep | 17:10 |
| felix_ | i did the calculations for layer 2 as ground plane and got sane values for the geometry | 17:11 |
| Kjetil | sure you won't get too high capacitance? | 17:12 |
| felix_ | i'm not very used to route non-differentail high speed traces though | 17:12 |
| felix_ | hmm, good question | 17:12 |
| felix_ | i put the values into the saturn pcb calculator and changed the geometry until i got about the impedance i wanted | 17:13 |
| Kjetil | I guess the only way to find out is to route the board and measure it | 17:13 |
| Kjetil | but antipads for the centerpin and capacitor I guess is a given anyway | 17:14 |
| felix_ | yep | 17:14 |
| Kjetil | And it's a minor modification to the board for the next batch anyways. As long as there are no traces in that area | 17:15 |
| felix_ | there are no near traces on the top layer | 17:16 |
| Kjetil | I was thinking in the layers underneath the HDBNC connectors | 17:16 |
| felix_ | there are no components or traces | 17:17 |
| *** se6astian|away is now known as se6astian | 17:43 | |
| felix_ | board is fully routed \o/ | 18:10 |
| felix_ | now some food ;) | 18:10 |
| se6astian | hurray! | 18:30 |
| Kjetil | \o/ | 18:36 |
| felix_ | is there some logo that i should add to the slkscreen? | 19:15 |
| felix_ | *silkscreen | 19:15 |
| felix_ | http://gerblook.org/pcb/s4L8MhjxyP6RkgcEZbgxGT | 19:35 |
| tpb | Title: GerbLook (at gerblook.org) | 19:35 |
| felix_ | Bertl: http://sigsegv.notmysegfault.de/intern/AXIOM-photonSDI-hw.zip the gerber export. i hope i checked the right boxes for the drill files | 19:38 |
| felix_ | i did some slight changes on one of the power supply polygions after the gerber export, but just some minor details | 20:15 |
| felix_ | uploaded a new version at the same url | 20:22 |
| felix_ | Bertl: do you know if there's some recommended geometry for the antipads in the planes below the capacitor between the hdbnc connector and the sdi chip? didn't find anything really useful with a few minutes of research :/ | 20:26 |
| felix_ | enough for me for today; i'd say the board is already looking quite nice | 20:33 |
| *** se6astian is now known as se6astian|away | 21:13 | |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!
