The Compact Model Workshop Minutes

Compact Model Council Agenda
September 26th, 1997

7:45am	Continental breakfast and registration
8:15am	Welcome and Call to order	Bhaskar Gadepally, Motorola Britt Brooks, Texas Instruments
8:30am	CMC Membership review	Bhaskar Gadepally, Motorola Britt Brooks, Texas Instruments
8:45am	BSIM3v3 Updates	Chenming Hu, UC Berkeley Weidong Liu, UC Berkeley Xiaodong Jin, UC Berkeley
9:35am	Break
9:50am	BSIM3v3 Updates continued	Keith Green, Texas Instruments Bernd Lemaitre, Siemens Joe Watts, IBM Paul Humphries, Analog Devices
11:15am	Lunch
12:30am	Membership matters	Bhaskar Gadepally, Motorola
1:00pm	Member Company Issues	Keith Green, Texas Instruments Dick Klaassen, Philips
1:30pm	BSIM3v3 Standardization	Britt Brooks, Texas Instruments
2:00pm	Break
2:15pm	Verification Suite	All
2:45pm	Future Work Suggestions Funding Issues	All
3:45pm	Next Meeting planning	All
4:30pm	Adjourn

Meeting opened: 8:20am on September 29th, 1997

27 people attended the Compact Model Council meeting. This was the second meeting of the members of the Compact Model Council under the EIA umbrella. After welcoming the attendees, Bhaskar Gadepally (Motorola) and Britt Brooks (TI), reviewed the current membership. The list included prospective members as well as activities that will be monitoring the CMC effort. The current members of the CMC are:

Advanced Micro Devices	Analog Devices	Avant!
Cadence	Digital Equipment Corp.	Hewlett Packard
IBM	Intel	Lucent Technologies
Motorola	NEC	Philips
Rockwell	Siemens	Texas Instruments

Companies that are not yet members, but are pursuing membership are:

Analogy

National Semiconductor

TSMC

Groups monitoring the CMC activities are:

FSA

NIST

SEMATECH

The review ended with the listing of several Web sites and email repeater/exploder information. Here is the information:

CMC homepage - http://www.eia.org/eig/CMC
NIST Model Validation homepage - http://ray.eeel.nist.gov/modval.htm
BSIM3v3 Homepage - http://www-device.EECS.Berkeley.EDU/~bsim3
CMC membership exploder - cmclist@eia.org
The SEMATECH repeater - compact-model-wg@sematech.org

BSIM3v3 Updates

The session began with Chenming Hu (UCB) presenting the rationale for updating the Capacitance model. He also showed a list of companies and representatives that have helped in the BSIM3v3 development thus far.

Weidong Liu (UCB) presented a proposal for correcting the scalability of the substrate current model and the new capacitance model (CAPMOD=3). The review went well with several questions from the audience concerning the updates and new documentation. Two new model parameters, MOIN and ACDE were added to the capacitance model.

Xiaodong Jin (UCB) presented the current list of future work that UCB would be investigating. Velocity overshoot, a modification to the RSCE model, and a better NQS model are all topics that UCB will be investigating in the coming year. Most of the CMC members agreed that the current RSCE model meets their satisfaction and that the NQS model needs to be added into the production versions of BSIM3v3 from the software vendors. The velocity overshoot is seen as an interesting topic, yet, at this time, BSIM3v3 did not need to be modified to show this affect.

The next few presentations showed BSIM3v3 evaluation data from the new capacitance model and modifications required for the Isub model. Joe Watts (IBM) showed encouraging results were the model improved the fit error from the CAPMOD=2 model. He also suggested several additional improvements for the model, including inner fringe capacitance, a separate overdrive voltage for capacitance (with an additional parameter to adjust the fit), and Vth0, K1, and K2 explicitly as functions of Tox.

Keith Green (TI) followed with a presentation of CAPMOD=3 plots that showed improved fitting with the new model without changing the existing model parameter set. Some adjustment of the new parameters may have yielded better results, but the results were, once again, encouraging. Keith also presented a proposal to eliminate the diode limiting equation from the BSIM3v3 model. He showed an example circuit where the equation causes erroneous results. Based upon CMC membership inputs, the diode model will be changed. See the action item list below for more information.

Paul Humphries (AD) showed the modifications Analog Devices used to enable the substrate current model to be scalable. The change added 3 parameters and removed the 1/L dependence from the equation. The results showed marked improvement in the fit to data. He also noted that all of the changes may not be necessary, but the first order change was critical.

NOTE: The Agenda was modified at this point due to a missed communication between the hotel and the meeting coordinator. At this point, the Membership matters section was moved before lunch.

Membership Matters

During this session, a proposal from Bhaskar Gadepally was made to allow participation of consultants (companies with less than 2 employees). After much discussion, consensus was reached that the fee for such a member should be $1K. After further consideration, the proposal was dropped for a different proposal. A membership called "Sponsored" was agreed to, where a member company may choose to invite interested parties to attend the meetings. This party would have no vote, unless the member company chose to relinquish their vote to the sponsored party.

Other member matters discussed:

Meeting format will be one day
Meeting frequency will be held at 4 per year
One of the meetings to be in Europe/Asia every other year.

Membership Issues

This session began with Bernd Lemaitre (Siemens) presenting a circuit that he has seen problems with when using BSIM3v3 in Saber. The circuit worked well with their internal model, but BSIM3v3 showed wide discrepancies. His major points for the presentation were:

Software vendors need to implement the standard model as a Black Box - no changes to the internal code
There must be a clear reference to the standard model release and implementation release
Software vendors need to document clear and comprehensive quaility of the model implementation
Model parameter sets should ONLY include core parameters of the standard model, no simulator specific parameters.

Dick Klaassen (Philips) then went on to ask several very important questions in his presentation. The issues were:

Meeting slides from April 10th, were they distributed?
Meeting dates need to be firm without changes - helps out international travelers
When will the next version of BSIM3v3 be stable/released?
Will the charge model be tested quantitatively?
What is the standardization process? Is it fixed? We are always improving BSIM3v3, when will it be a standard?
What about other model standardization efforts?

The end of the session ran longer than anticipated, but several action items were documented. Keith Green lead this portion of the session and had captured the action items from the BSIM3v3 updates section. Here is a list of the action items:

Action Item	Date
The UC Berkeley BSIM3 team will incorporate fixes for the current bug list in the next release of BSIM3v3	12/97
Paul Humphries (AD), Mishel Matloubian (Rockwell), and Keith Green (TI) will formulate requirements for an updated Isub model and forward to UCB	12/97
Paul Humphries (AD), Shiuh-Wuu Lee (Intel), Keith Green (TI), Richard Taylor (National), Larry Yu (AMD), Toshiyuki Saito (NEC), and Ferenc Masszi (DEC) will formulate requirements for a BSIM3v3-based MOS capacitor model and forward to UCB	12/97
Min-Chie Jeng (Cadence) and Bob Daniels (Avant!) will work with UCB to restructure the NQS model in BSIM3v3 to make it easier to transfer to other simulators	12/97
Weidong Liu (UCB) will document the proposed CAPMOD=3 model. Documentation will be available from Weidong upon request	12/97
Weidong Liu (UCB) will fix the CGSO- and CGDO-related code bug reported by Joe Watts (IBM)	12/97
Ping-Chin Yeh (HP) and Joe Watts (IBM) will formulate requirements for modeling bias-dependent inner fringing capacitance and forward to UCB	12/97
Ping-Chin Yeh (HP) and Joe Watts (IBM) will formulate requirements for adding new parameters to fit C-V "turn on" characteristics and forward to UCB	12/97
Joe Watts (IBM) and John Krick (TI) will formulate a proposal for adding tox dependence in VTH0, K1, and K2	12/97
Bernd Lemaitre (Siemens) will work with UCB to investigate and fix the bug associated with VFB in BSIM3v3	12/97
Keith Green (TI) will work with UCB to remove the source/body and drain/body diode forward-bias current limiting function and add RWELL to BSIM3v3	12/97
UCB will add code to BSIM3v3 which will cause a warning to be generated if PSCBE2<=0	12/97
UCB will add boundaries to parameters MOIN and ACDE in the developmental CAPMOD=3 code	12/97
Bernd Lemaitre (Siemens) will investigate the necessity to add HSPICE-specific parameters, e.g., LDIF and HDIF, to the standard BSIM3v3 model and report back to the CMC via e-mail	12/97

BSIM3v3 standardization

Britt Brooks presented a quick synopsis of where BSIM3v3.1 stands as far as standardization is concerned. The outstanding action items from this section are:

Obtain permission from UCB regents to use BSIM3v3.1
- Britt and Patti will work with Chenming to achieve this action
Date: End of October
- Britt will supply PDF version of the BSIM3v3 manual to UCB
Date: ASAP
Document verification plan
- IEEE Recommended Practice is finished
Date: 5/21/97
- Test suite from Univ. of Waterloo will be evaluated
Date: December, 1997
Who: Intel - Shiuh-Wuu Lee, Motorola - David Newmark, Avant! - Bob Daniels, Siemens - Bernd Lemaitre, Cadence - Min-Chie Jeng
Apply for EIA standard
- Patti will get the paperwork started
Date: ASAP
Promote BSIM3v3 as a standard MOSFET model
- All members should be doing this currently
Date: Ongoing

Verification Suite

Britt Brooks informed the CMC that the University of Waterloo was planning to release a test suite of circuits that could be used in HSPICE, Spectre, and SPICE3. The circuits range in complexity from a single transistor to a 15,000 transistor RAM. The circuits came from a conference hosted by MCNC called CircuitSim 90. The original intent was to benchmark simulators against each other, but these circuits may be a good starting point for the verification suite we need. The cicuits are not currently available, but Prof. James Barby will make them available shortly. Britt will pass on the information as soon as the circuits are ready for use.

Note: The extraction tools should undergo the same tests that the simulators use. This way, the implementation of BSIM3v3 can be guaranteed for both the simulator and the extractor. Usually, the two tools will have different implementations and therefore must be subjected to the same kind of testing.

The final session of the day was a review of where the CMC would be spending funds. Three items of note, BSIM3v3 development, Standardization documentation of BSIM3v3, and EIA administration fees were discussed. Future modeling issues (Bipolar, SOI, etc.) were visited shortly, but no decisions were made concerning the direction of the Council in dealing with these efforts. The topic of officers was revisited, but no change in the current Council structure was necessary. This issue will be revisited in December for the 1998 operating year.

Meeting closed: 3:50pm on September 26th, 1997

Here is the list of attendees and their email addresses.

Eric Arnold	Hewlett-Packard	earnold@wlv.hp.com	(818) 879-6394
Britt Brooks	Texas Instruments	brittb@.ti.com	(972) 997-6037
Bob Daniels	Avant!	bobd@avanticorp.com	(510) 413-8603
Bhaskar Gadepally	Motorola	Bhaskar_Gadepally@mesaqm.sps.mot.com	(602) 655-3367
Keith Green	Texas Instruments	kgreen@msp.sc.ti.com	(972) 480-7645
Jeffrey Hantgan	Lucent Technologies	jch@bell-labs.com	(610) 712-3329
Chenming Hu	Univ. of California Berkeley	hu@eecs.berkeley.edu	(510) 642-3393
Paul Humphries	Analog Devices	paul.humphries@analog.com	(617) 937-2754
Min-Chie Jeng	Cadence Design Systems	mjeng@cadence.com	(408) 428-5629
Xiaodong Jin	Univ. of California Berkeley	xjin@pauli.eecs.berkeley.edu	(510) 642-1010
Kuntal Joardar	Motorola	rrnr50@email.sps.mot.com	(602) 655-3093
D.B.M. Klaassen	Philips	klaassen@natlab.research.philips.com	+31 40 2744251
Shiuh-Wuu Lee	Intel	shiu@td2cad.intel.com	(408) 765-0217
Dr. Bernd Lemaitre	Siemens AG	lemaitre@hl.siemens.de	49-89-636-3775
William Liu	Texas Instruments	wliu@spdc.ti.com	(972) 995-4493
Weidong Liu	Univ. of California Berkeley	liuwd@pauli.eecs.berkeley.edu	(510) 642-1010
Ferenc Masszi	Digital Equipment Corp	masszi@subpac.enet.dec.com	(978) 568-4956
Mishel Matloubian	Rockwell	mishel.matloubian@rss.rockwell.com	(714) 221-6641
David Newmark	Motorola	ra1697@email.mot.com	(512) 933-6099
Patti Rusher	EIA	pattir@eia.org	(703) 907-7545
Toshiyuki Saito	NEC Corporation	saito@lsi.nec.co.jp	81-044-435-1518
Mahesh Sharma	Motorola	rzvt90@email.mot.com	(512) 933-6462
Richard Taylor	National	Richard.F.Taylor@nsc.com	(408) 721-4884
Tom Vrotsos	Texas Instruments	t-vrotsos@ti.com	(972) 480-7339
Joe Watts	IBM	jwatts@btv.ibm.com	(802) 769-1874
Ping-Chin Yeh	Hewlett-Packard	ypc@hpmozart.corp.hp.com	(415) 857-6198
Larry Yu	Advanced Micro Devices	lyu@grape.amd.com	(408) 749-3249

Thank you all for your participation and attendance.

This meeting was conducted in accordance with the EIA Legal Guides and EIA Manual of Organization and Procedure.