TECHNET Archives

1996

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Sid Tryzbiak <[log in to unmask]>
Date:
Fri, 18 Oct 1996 17:36:57 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (112 lines)
>Return-path: <[log in to unmask]>
>X-Sender: [log in to unmask]
>Date: Fri, 18 Oct 1996 13:35:58 -0400
>To: [log in to unmask],lspotts.com
>From: Brian McDermott <[log in to unmask]>
>Subject: Proposed response to Technet
>
>Folks,
>	This discussion appears to be focused on laser -v- plasma etching with
>little input from the Photo-Via developers. Here it is ! 
>
>CCI has been developing a photo defined micro-via process using
>photoimageable dielectric for the last 3 years (called PhotoLink). We began
>production in October 1995. We have various products in pre-production and
>qualification. 
>
>We believe this process is the way forward in terms of cost, throughput and
>via quality/reliability. Unfortunately, the diversity of micro-via
>technologies, the confusion on the pro's/con's and the absence of a low cost
>process with multiple established suppliers, has prevented large scale
>implementation of micro-vias. 
>
>PhotoLink uses a curtain coated photoimageable dielectric, applied to the
>substrate, dried, exposed with an artwork and developed, forming 4 mil vias
>in the dielelectric. Copper is then plated onto the surface of the
>dielectric and into the vias. Vias can be resolved down to 1 mil. Our
>standard via is 4 mils in a 12 mil pad. 
>
>The photo-dielectric is a product from Ciba Geigy which has been
>specifically designed for this purpose with excellent processing and plating
>characteristics. Boards have passed UL and various reliability testing at
>CCI and our customers. 
>
>This process is considerably less costly than IBM's SLC process, with none
>of the non-value added process steps in the SLC process. This process does
>not have the plating adhesion concerns of SLC. Peel strengths average 8lbs/in. 
>
>THROUGH-PUT / EXTRA PROCESS STEPS
>
>There is no lamination process (material is curtain coated). Through put for
>curtain coating is 120 18x24 panels/hour in a continuous flow process. I
>believe the current plasma etching equipment takes 6 panels at a time (30 to
>60 mins processing time ??). Laser vias may be effective for localized small
>numbers of vias (< 2000 per panel). For photo-vias the number of vias has no
>effect on the cost (the first one costs $10 and the rest are free !!). If
>available at no extra cost, designers will use the vias in their designs to
>reduce size and layer count.  Our designs have a minimum of 10,000 vias per
>side on each 18x24 panel.
>
>Laser drilling and plasma etching require the additional steps of selective
>etching of a dot in the outer layer copper to define the subsequent via. Plasma
>etched vias also require an etching process after via formation to remove
>the copper overhang formed by the undercut of the plasma etch process.
>Photo-defined vias by their nature have a clean straight side-wall with no
>undercut.  
>
>COSTS
>
>As an example of the some of the cost benefits of this technology, one
>customer redesigned an 8 layer double blind via board to 4 layers (d/sided core
>with photo dielectric on both sides), resulting in a 40% cost reduction.
>Another 6 layer board is redesigned to 3 layers, going from d/sided SMT to
>single sided SMT with both PCB and assembly cost savings. Assembly savings
>were more than the PCB cost.
>
>I agree that manufacturers of microvia PCBs need to stanardize the physical
design rules. However, will our customers purchase PCBs for a single
product, manufactured with different technologies, even if our physical
design rules are the same ?  If the electrical function of the different
technologies is the same, then maybe so. However, recent testing with the
different technologies on an RF application suggests that each
>technology will require a different circuit design because of dielectric
constants and other differences. Thus, in this case, a single technology
will be chosen by the customer based on cost and reliability. 
>
>I welcome an open discussion on the pro's and con's of each technology. I
>think the conclusion will be that photo defined vias are the lowest cost. We
>need to stanardize our design rules, and put a cost estimate against each
>incremental improvement in the design rules. We need multiple suppliers with
>a low cost volume process in order to get this market moving in the US. 
>We would be glad to share more information on PhotoLink and Ciba Geigy's
>photo-dielectric with both PCB manufacturers and end users. We are holding a
>series of technology workshops in the coming months, the first of which is
>in the Boston area on Nov 14th. Please contact me for more details.
>
>Brian.
>
>
>________________________________________________________________
>Brian McDermott
>
>New Product Manager
>
>Continental Circuits
>1150, Belle Avenue,
>Winter Springs,
>Fl 32708, USA
>Tel: 407 6995000	  
>Fax: 407 6996871
>email: [log in to unmask]
>
>

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************



ATOM RSS1 RSS2