Greetings, Before my recent trip to San Diego for the IPC CONFERENCE and APEX/EXPO fades to a dim memory, I thought I would share my experience with those of you who couldn't attend. (since we have had shrinking travel budgets, I always appreciate hearing from YOU if I can't make it to something) So, most of my involvement was in the IPC Standards Development Committee Meetings, and here's a synopsis of the ones I attended: (apologies in advance if I misunderstood or misrepresented anything. Sometimes I still feel like "the new guy" during these discussions) J-STD-001/IPC-A-610 SYNERGY MEETING (Saturday 8:00-5:00) This meeting works to resolve discrepancies between the Requirements for Soldered Electrical and Electronic Assemblies (J-STD-001) and the Acceptability of Electronic Assemblies standard (IPC-A-610). - The committee discussed the "void criteria" in Table 7-16 for components with non-collapsible solder balls. There is not enough data available to support a decision either way, so the committee was proposing to add the text "void criteria will be established between user and supplier". There were arguments against this, because it could create "action items" for many companies to establish acceptability criteria for voids with its suppliers. The committee voted to accept the text “void criteria is not established.” - Solder “Vertical Fill” Requirement (IPC-A-610 Table 7-4) One of the longest discussions was determining the solder filling requirement for plated holes per . The committee was considering relaxing the requirement for designs with copper planes (or for holes connected to thermal planes), because of the difficulty of getting solder to fill the hole. Maybe this example will illuminate the important points of the discussion: Let's say we specify a 2-pin through-hole connector in a design as a "power" connector, one pin connected to a power plane and the other to a ground plane. Now suppose we specify another 2-pin connector in the same design as a "sensor" connector, with signal connections (neither pin tied to a plane). Now, if the soldering requirements are relaxed for holes connected to thermal planes, the question will immediately be asked, "Why does my Sensor Connector have to be soldered BETTER than my Power Connector?" It doesn't make sense. If the relaxed requirement is acceptable for one, it should be acceptable for the other, and the very next proposal will be to relax the requirement for ALL soldered holes. The committee is not prepared to accept that yet without reliability data to support the change. On the other side of the discussion, suppose we are specifying a 70-pin Through-Hole connector and we decide to connect all of the unused pins to the ground plane. As long as the mechanical strength of the connection is not compromised and the electrical connection is maintained through-out our worst-case scenarios (extreme thermal cycling, for example) do we really care if those holes are completely filled with solder? I don't think this issue has been fully resolved by IPC, but for now, the proposal to change the requirements for designs with thermal planes was REJECTED. - Board Edge Delamination (J-STD-001 9.1.3, IPC-A-610 10.2.5) The committee resolved a discrepancy between the way edge delamination criteria was stated in the two documents. If I understood correctly, one document declares a reject if the delamination is greater than 50% of the distance to the nearest conductor or GREATER THAN 0.1 inches from the board edge. The other document declares a reject if the delamination is greater than 50% of the distance to the nearest conductor or WITHIN 0.1 inches of the conductor. The committee voted for the second option. IPC-6012 / IPC-A-600 BARE BOARD ACCEPTABILITY (Sunday 8:00-5:00) Copper Wrap Requirement (IPC-6012 Table 3-4) The committee had proposed changing the wrap requirement from 500 micro-inches to 200 microinches. The document is falling behind current manufacturing practices (it has been reported that 80% of high density board designs are being fabricated at 200 with written waivers or exceptions to the requirement), but several negative votes have been submitted against the proposed standard for ballot. I think Don Dupriest's point was that many of his products had specifically been designed for enough clearance to maintain the 500 requirement, and by changing the standard, he wouldn’t be able to prevent fabricators from building to 200 without revising the procurement documentation. Others expressed reluctance based on the lack of reliability data, so the committee voted to retain the 500 specification for now. This brought up another discussion related to legacy data, below: - “Cut-In” dates to preserve Legacy Product Requirements: The traditional method of calling out IPC standards on drawings is to use the document number WITHOUT the revision letter. A specification that uses a document number without its revision level is always assumed to be “whatever the latest revision is”. That way, the product can be manufactured using the latest industry standards without having to revise the drawings or other procurement documentation. The IPC is entertaining the idea of using “cut-in dates” in certain situations to allow fabricators to “catch up” to the latest publications. For example, the committee that maintains the IPC-4101 “Specification for Base Materials” is implementing this concept to allow laminate manufacturers a certain amount of time to use up their existing materials before the new specification goes into effect on the cut-in date. - Micro-section Preparation There were two ways of interpreting the tolerance for preparing micro-section samples for quality control inspection, related to the acceptable amount that the section can be off-center. The specification allows 10%, but it is not clear how to determine the 10% tolerance. Is it 10% of the distance away from the true center of the hole? or is it that the measurement from side to side can’t be more than 10% less than the true diameter of the hole? After a lengthy discussion, the committee voted to measure the diameter, which is more definitive than trying to measure the depth of the "grinding zone", but this method allows a greater margin of error. DESIGNER DAY (Monday 7:30-1:00) This was a series of short presentations on topics of interest to circuit board designers. - Planning for Design Success - Carl Schattke, Tesla Motors - Component Mounting Issues and Recommendations - Rainer Taube, Taube Electronic GmbH and FED - Surface Mount Design and Land Pattern - Tom Hausherr, CID+ President, PCB Libraries, Inc. - Successful Communication with Cross-Cultural Teams - Stephen Chavez, CID+ UTC Aerospace Systems - Success Through Control of Cost and Quality - Rick Hartley, retired from L-3 Communications IPC-2152 CURRENT CARRYING CAPACITY (Monday 3:00-5:00) The initial release of IPC-2152 has been out for a few years now, and the IPC has not received any negative feedback regarding the data or advice it provides (no news is good news?). A survey was created to collect user feedback, and to help set a direction for the next steps in the evolution of the document. The summary highlighted three areas for further study; 1) study the combined effect of current through parallel traces 2) study the thermal performance of vias 3) study the performance of flexible circuits A test board design was proposed that would be useful for items 1) and 3), and the fact that this design could also be used to test pulses and current spikes was discussed. FTG offered to make the test vehicle circuit boards for the committee. Oh, almost forgot to mention this; The document/committee name is being changed from "Current Carrying Capacity" to "Thermal Management" KEYNOTE – THE XBOX STORY (Tuesday 8:00-10:00) Robbie Bach, former President of Entertainment & Devices at Microsoft, gave an interesting presentation about “The Xbox Story: Lessons in Strategy, Team Management and Entrepreneurship”. He started out talking about goals and the benefit of focusing a group mission into one sentence. For examples, Microsoft’s ”put a computer on every desk in every home running Microsoft software”, and Sonos’ “bring all the music in the world into every room of your house” (paraphrase). But the concept that impressed me the most was that when Microsoft had the “Red Ring of Death” disaster that cost them over a $BILLION to recover from, their Customer Approval rating actually INCREASED during that time because of the way they handled it (extending warranty to 3 years, no-hassle return policy, etc.) Great Talk! IPC-2221/2222 PRINTED BOARD DESIGN STANDARD (Tuesday 10:00-5:00) At one point in time the IPC had considered adding all of the Design Guidelines for HDI into IPC-2222 (Design Standard for Rigid Boards), but it has been decided that there is too much specific information for HDI, and they will continue to retain that information in the sectional document IPC-2226 (Sectional Design Standard for High Density Interconnect). This document was originally released in 2003 and needs a significant update, as the technology has changed so much since then. The committee addressed some of the initial suggestions and assigned sub-groups to review and revise some of the data in the Tables. SHOW FLOOR RECEPTION (Tuesday 5:00-6:00) I was busy in meetings most of the day so I didn't invest much time on the Exhibition floor, but one of the more interesting technologies I saw was eSurface, which is a new process for creating very thin traces with strong peel strength and tight tolerance. One mil trace/space geometries are now routinely being manufactured at Murietta Circuits in Anaheim, CA. Here's How: http://esurface.com/the-esurface-manufacturing-process-steps/ Here's a collection of Photos and Videos from the exhibition floor hosted by "RealTime with... IPC" http://realtimewith.com/rtwboot/show.php?id=81 IPC-7351 LAND PATTERN STANDARD and IPC-7070 COMPONENT MOUNTING (Wed 8:00-12:00) The IPC is reviving the old IPC-CM-770 Component Mounting Guideline (1996?) and after a complete rewrite it will be assigned to number IPC-7070, which will conform to the newer document naming convention. Rainer Taube is leading this effort. He will be collaborating with Tom Hausherr who is heading up a significant revision to IPC-7351 (Land Patterns). The committee voted to rename IPC-7351 from a “Land Pattern Standard” to a “Land Pattern Guideline”, since there is so much variance in land pattern dimensions currently being used in the industry. Some of the more drastic changes in IPC-7351 is to move most of the component mounting information to IPC-7070, to recommend “rounded rectangle” shapes over “rectangle” in many families, especially for RoHS compatible designs (paste screening process doesn’t like sharp corners anyway, harder to “release”, less wetting in reflow leaves exposed copper corners, corners don’t contribute to solder joint reliability anyway). Another change is that companies are starting to realize that all the silkscreen ink underneath components is just a waste of ink, not useful to anyone and in many cases contributes to soldering defects because small components can’t settle into molten solder if they are riding on a silkscreen line (think of a “teeter-totter” on a playground). The three-tier system based on design density is being removed from several families of components, in favor of recommending a “best” or “proportional” solder joint dimension. Also, the “rectangular courtyard” system is being replaced by courtyards that more closely match the component body size to allow better placement strategies that won’t look like errors to CAD rule-checking systems. IPC-7093 BOTTOM TERMINATION COMPONENTS (Wednesday 1:30-3:00) The committee discussed a new etching technology that would expose part of the side wall of the termination to the plating process, which would make a partial side fillet, making automated optical inspection more reliable. Also discussed window-pane techniques for the thermal pad solder stencil and negative values in Table 6-1. One interesting point that came up was that Table 6-1 in 7093 is identical to Table 3-1 in 7351. It wasn't clear who copied from who, so now if that table needs to be revised, who owns it? and who prevents them from becoming out of sync with each other? Maybe we need some kind of digital "links" when documents are borrowing from each other? TRIBUTE TO DIETER BERGMAN (Wednesday 6:30-8:30) Dieter Bergman joined the IPC in 1962, became chairman of the IPC Design Committee in 1968, was elected Chairman of the IPC Technical Activities Executive Committee in 1974, and then became the IPC Technical Director. Known as “the man who would take 45 minutes to answer a 10-second question”, my personal experience was that he and Gary Ferrari came to California to present the new IPC-STD-275 Design Standard to us sometime in the late 80’s, planted the seed that we could all benefit from collaboration, and encouraged us to work towards creating a common foundation for everyone to build upon. That Design Standard energized our PCAD User Group, and I’m still involved to this day because of his inspiration. Thanks, Dieter… R.I.P. Oh, I suppose I should mention some CELEBRATIONS! Gary Ferrari is our newest IPC HALL OF FAME member! (He taught me CID+, and I am so happy that he has been recognized for his work) Oh, and eight people received Dieter Bergman Memorial Scholarships: Don Dupriest, Denny Fritz, Dave Hillman, Bernie Kessler, Bob Neves, Ray Prasad, Randy Reed and Doug Sober and my friend Kelly Dack became an IPC certified Trainer at the conference, successfully completing his first CID class, monitored by Gary. Way to go, Kelly! Now we have 17 new CID members!! and we also have 10 new CID+ members, taught by Paul Fleming!!! and I got to eat ice cream with Bernie Kessler!!!! ok, when I get to the quadruple exclamation points, I realize I need to fade away for awhile. In summary, even though I was there for 6 days I still feel like I missed a lot. Hope this helps those of you who could not attend this year. Jack (aka "the new guy") ______________________________________________________________________ This email has been scanned by the Symantec Email Security.cloud service. For more information please contact helpdesk at x2960 or [log in to unmask] ______________________________________________________________________