FIRST FORMAL MEETING OF THE

POWER ELECTRONIC MODULE INTERFACE WORKING GROUP

P1461

WHEN: 7:00pm to 10:00pm Tuesday September 10, 1996

WHERE: BALLY's Resort Hotel, Las Vegas Nevada, in conjunction with PCIM 96

CHAIR: Dennis Darcy, SatCon

SPONSOR: Jason Lai, VPI

VICE CHAIR and SECRETARY: Mark Harris, EMPF/NAWCADI

ATENDANCE:

Sunil Chaya Carlo Venditti
Ken Plant Tim Hickey
Pete Savagian Todd Nakanishi
L.(Vish) Viswanathen Daryl Prince
Norman Wong Donald Humbert
Paul Davis Howard Imhof
Ian Cuelland Tom Saunders
Jim Sarjeant John Miller
Angelo Ferraro Stan Krauthahmer
Jeff Fishbein Fenton Rees
   

ACTIVITY: (Roughly followed the agenda)

7:00 Started the sign up sheet around. At around 7:10pm Mr. Dennis Darcy (Chair) and Dr. Jason Lai(Group Sponsor) called the meeting to order, and thanked everyone for attending so late a meeting in a city with many other attractions. Dennis also said that we would be setting the stage for future work together. Went around the room for people to introduce themselves. We then reviewed the agenda and asked for any additions, deletions or comments, of which there were none. Since there had been no previous meetings we did not need to approve the minutes.

Dennis Darcy Presented PAR & IEEE Standards setting process after which we introduced the PAR for our group, reviewed the and discussed the purpose of the Working Group. Several People asked for copies of the PAR. Act.Itm-Make sure that people got copies of the PAR.

Some discussions and opinions expressed, about standards vs. Recommended Practices. It was the opinion of one or more that the Working Group should target the development of Recommended Practice rather than a standard. An RP could evolve into a standard after a period of industry validation, comment and acceptance. The Current PAR document actually is for a Recommended Practice and the commentary from the floor all through the meeting supported this.

Dennis Darcy, Discussed the Working Group Operations. Started with the IEEE Imperative Principles of Due Process, Consensus, Openness, Balance and Right of Appeal. For more information on this discussion refer to the IEEE Standards Companion pp4-6. Then went on to discussing Voting rights:

One of the members had strong opinions and asked to put up a viewgraph. emphasized that members should do work, attend the meetings and bring expertise, then they could vote.

Several people argued that it would be hard for some corporations to always send the same person, that a corporate designee or something similar would be needed so that groups could send different people but still vote. The counter argument was that this is a working group, if attendees rotate through then it will be hard to maintain continuity and working momentum.

It was also pointed out that some groups had people who put in a lot of work and voted on issues even though they never physically attended the meetings. This person interacted by e-mail and telephone and was very successful.

Overall tone of the discussion was that continuity and input are the important criteria for voting rights within the Working groups. Obviously modulated by the IEEE Imperative Principles as necessary. It was pointed out at the end of the discussion that voting in the working group needs to be personal, with a minimum of corporate bias. We do need to make sure that one company, one industry or segment does not control the proceedings. The more formal balloting process will be controlled by the IEEE committees not the WG. Act.Itm.-Try and recruit a broader representation of module manufacturers and also systems and end user representation.

Mark Harris the Secretary reviewed plans to use the WWW for Work Group Communications. Plan on using the WWW e-mail and electronic document transmission to the maximum extent possible. Membership lists, minutes, agendas and working documents would be available over the WWW. Paper versions would be a fall back for those without Internet access. Passed around Group Questionnairs, asking for everyone to sign up and give us their e-mail information. Act.Itm. Get the site set up within the next two months.

Dennis Darcy Reviewed our Tentative Future Meeting Schedule

Dennis Darcy Reviewed what we felt needs to be done, when we need to do it and how we will organize. Our proposed timetable is to have a strawman ready for discussion by APEC '97 and a draft by PCIM '97 although no one expects either to be close to final form. Our proposed organization is into Task Forces that would be organized to work on sections of the Strawman/Draft.

Dr. Sarjeant volunteered to act as coordinator between the passives group and to help with commentary in the PELS Transactions of which he is a Technical Editor.

Dennis Darcy Review PAR and its background, why we had started the process to get the group going and the direction that we were taking initially. The WG is an outgrowth of discussions between the Sponsor the Chair and the Secretary about the need for some standardization in the interfaces on Power Electronics Modules, that would make the design and fabrication of systems quicker and would provide component manufacturers targets towards which they could aim other than over optimizing the performance of their own component. They asked that people think about this and reply.

Dr. Arthur Kelley had been invited to speak for a while and he put up a few viewgraphs and discussed standardization concepts, why they might me good and what they could look like. This set of a series off wide ranging discussions on the nature of the industry, technology and applications. His viewgraphs showed what could be done by a very simple standardization of input and output of pad positions.

The opinion was put forth that there could never be 'standards' in this area because it limited the ability of the component vendor to maximize the performance of his product.

There was also mention of the fact that standardization would turn this into a commodity market (i.e. suppliers would lose control of captive 'sockets') and that would allow the low cost suppliers to take away the market.

Standardization at too high a level of integration would push the silicon vendors into the position of providing systems since they could/would integrate to the maximum extent possible and be able to undersell everyone else.

Standardization of the 'socket' i.e. physical outline and placement of buss contacts was not an IEEE function but one for an organization such as JEDEC.

Some of the participants were looking for a highly standardized product that would allow them to shop around for the best price/performance mixture

Many of the participants agreed that the way modules (and other components) are spec'd can hide as much as they reveal. This lead to a general agreement that standardization of at least how and what you specify would be of value.

Currently at the low power end of the spectrum everything is highly integrated, in some markets, defacto or actual standards have come into being. In the high power end, everything is to a degree custom, most 'catalogue products' are in fact modules built for a specific application that were then put in the catalogue with the hope that more customers would buy it (the profit margin being much greater in general after the bugs are worked out with the first customer(s)).

Although everyone builds custom and perhaps its impossible to standardize physically, you may be able to standardize how you specify performance.

In the end it seemed to meet general agreement that we would indeed be writing something which was a set of suggested practices rather than a standard in the sense of a product definition.

That some minimum set of mechanical standards (options) for the interfaces would make sense.

That what we would try and address was the minimum set of issues that would imporove time to market.

8:30 Break

8:45 Working Group Kickoff

Had Dr. Sunil Chaya presente the strawman module spec from the Partnership for the Next Generation Vehicle

Open Discussion of strawman led to a rehash of the discussions held earlier.

Then went on to discuss the structure of the Task ahead and the Task Forces that would carry out the majority of the work.

Topics for Task Forces, after some discussion Dennis Darcy Listed the following:

Generally accepted as being a starting point

Asked the attendees who would be willing to head each task group

Discussed Schedules for Task Forces

9:55 Review of Action Items

NEXT MEETING WILL BE AT APEC 97 Late Feb in Atlanta GA. Draft Agenda, More details at a later date..

BACK