Closed gwright83 closed 4 years ago
I'm not sure about the file version.
What actually got fabricated was what I was trying to explain in my previous email ... nothing was "plugged with epoxy". That is, the Thru (unplugged) vias in the first rev were replaced, I believe, by stacked vias. When you have stacked vias in the "thin copper foil process", the L1-L2 via automatically gets plated ... i.e. filled with copper (not plugged with epoxy and plated over). In the "thin copper foil process" required for the fine-feature etching, the dielectric between L1 and L2 is required to be the thinness required for blind vias so that the plating step by itself will fill the L1-L2 via with copper.
On Fri, Jun 5, 2020 at 1:07 PM Gregory Wright notifications@github.com wrote:
@kostatro https://github.com/kostatro @aflynn https://github.com/aflynn Does the Altium database from the BWRC reflect the final state of the board as manufacturer by Sierra, or did Sierra make changes that were not backported to the BWRC?
The question came up because it appears as if there are thru vias under the IC even in the file fader2_blind_vias_pcb.PcbDoc. My understanding is that these were plugged so that the under-chip bump attach points would be flat.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/FaderOrg/Fader25_board/issues/21, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA5YEEGLCA3JIBHDCC2336TRVFGBFANCNFSM4NVC6EIA .
You could ask Sierra if you like. The contact is justinb@protoexpress.com.
On Fri, Jun 5, 2020 at 1:07 PM Gregory Wright notifications@github.com wrote:
@kostatro https://github.com/kostatro @aflynn https://github.com/aflynn Does the Altium database from the BWRC reflect the final state of the board as manufacturer by Sierra, or did Sierra make changes that were not backported to the BWRC?
The question came up because it appears as if there are thru vias under the IC even in the file fader2_blind_vias_pcb.PcbDoc. My understanding is that these were plugged so that the under-chip bump attach points would be flat.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/FaderOrg/Fader25_board/issues/21, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA5YEEGLCA3JIBHDCC2336TRVFGBFANCNFSM4NVC6EIA .
@aflynn Got it. I will ask them just to be sure.
The updated Altium file should reflect the changes required by Sierra. I don't have access to Altium to check it now... It's a good idea to ask Justin.
On Fri, Jun 5, 2020 at 23:21, Gregory Wrightnotifications@github.com wrote:
@aflynn Got it. I will ask them just to be sure.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.
Yes, I have no idea about Kosta's board, but on my last EagleX board, Sierra did make a change that I approved during the time that their CAM Shop asks questions, so my file in the repo is not exactly what they manufactured ... so it happens...
(Their change did not change the functionality of my board, but it streamlined their process. The change was to change my Thru vias to a set of stacked vias, as I mentioned in my previous email.)
On Fri, Jun 5, 2020 at 9:40 PM Kosta Trotskovsky notifications@github.com wrote:
The updated Altium file should reflect the changes required by Sierra. I don't have access to Altium to check it now... It's a good idea to ask Justin.
On Fri, Jun 5, 2020 at 23:21, Gregory Wrightnotifications@github.com wrote:
@aflynn Got it. I will ask them just to be sure.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/FaderOrg/Fader25_board/issues/21#issuecomment-639980975, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA5YEEGQTGZKCHYVDDSFGADRVHCCHANCNFSM4NVC6EIA .
Picture of the identifying portion of the previous FADER board:
Justin Blakely at Sierra Proto Express sent me a zip of the files used to make the existing board (Sierra job number 151023). Here it is: job_151023.zip
This should tell us the solder mask geometry.
@caolen Hi Carl, you might want to take a look at the files from Sierra, including in the previous comment on this issue.
Greg, thanks.
I have also forwarded to Pekka and Joe. Carl
From: Gregory Wright notifications@github.com Sent: Thursday, June 11, 2020 1:08 PM To: FaderOrg/Fader25_board Fader25_board@noreply.github.com Cc: Olen, Carl (Nokia - US) carl.olen@nokia-bell-labs.com; Mention mention@noreply.github.com Subject: Re: [FaderOrg/Fader25_board] Does the Altium database correspond to the board Sierra built? (#21)
@caolenhttps://github.com/caolen Hi Carl, you might want to take a look at the files from Sierra, including in the previous comment on this issue.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/FaderOrg/Fader25_board/issues/21#issuecomment-642815715, or unsubscribehttps://github.com/notifications/unsubscribe-auth/APMOSPPQSR7CBPCP2BWLQM3RWEFPLANCNFSM4NVC6EIA.
On today's phone call we agreed that the Gerber files provided by Sierra Proto Express show that there were changes made at Sierra that were not back-ported to the Altium database at the BWRC. Rather than reverse engineering what was done at Sierra, I will close this issue and we can concentrate on a design that can be built by Nokia's preferred vendors.
There is anecdotal evidence that the Eagle chip, a RISC-V cpu plus specialized radar signal processor, may have had trouble with reliable connections to the circuit board. Eagle was an identical size chip with the same ball pitch and its board was also made by Sierra. The main difference was that Eagle had many more signal connections than FADER. The problems with the Eagle board have never been explained; it is best that we concentrate on our own approach.
@kostatro @aflynn Does the Altium database from the BWRC reflect the final state of the board as manufacturer by Sierra, or did Sierra make changes that were not backported to the BWRC?
The question came up because it appears as if there are thru vias under the IC even in the file fader2_blind_vias_pcb.PcbDoc. My understanding is that these were plugged so that the under-chip bump attach points would be flat.