Free Letter - District Court of Delaware - Delaware

Case 1:07-cv-00633-JJF-LPS

Document 88

Filed 05/02/2008

Page 1 of 2

FISH & RICHARDSON P.C.
Suite II00

Frederick P. Fish 1855-1930 W.K. Richardson 1859-1951

919 N. Market Street P.o. Box 1114 Wilmington, Delaware 19899-1114

VIA ECF May 2, 2008
The Honorable Leonard P. Stark United States District Court

Telephone 3oz 652-5070 Facsimile 302 652-0607 Web Site www.fr.com William J. Marsden, Jr. 302 778-8401 Email [email protected]

for the District of Delaware 844 King Street Wilmington, DE 19801

m
ATLANTA AUSTIN BOSTON DALLAS DELAWARE MUNICH NEW YORK SAN DIEGO SILICON VALLEY TWIN CITIES WASHINGTON, DC

Re:

Power Integrations, Inc. v. BCD Semiconductor Corporation USDC-D. Del. - C.A. No. 07-633 JJF-LPS

Dear Magistrate Judge Stark: Power Integrations submits these comments on Defendants' ("BCD") technology tutorial (D.I. 81) pursuant to Your Honor's Scheduling Order. (D.I. 44 at ¶13.) First, despite the Court's order that the tutorials "should focus on the technology in issue and should not be used to argue the parties' claims construction contentions" (id.), BCD's tutorial does just that by assuming constructions that are contrary to (and/or were rejected by) those adopted by the Court in the Fairchild litigation. BCD also makes reference to "the claim interpretations that Plaintiff has advanced," but Power Integrations notes that the Court has construed the claims, not Power Integrations. Power Integrations disagrees with BCD's suggestion that the Court's prior constructions have no bearing in this matter. Second, BCD's tutorial suggests that there is some issue in this case regarding what it refers to as "pseudo-random" frequency variation, but "pseudo-random" is not a term in any of the asserted claims, and it is not relevant to the frequency jitter inventions as recited in the patents-in-suit. BCD's effort to inject a dispute regarding whether the output of a D-to-A converter ("DAC") used to vary an oscillator might be "pseudorandom" is a red-herring intended to mischaracterize the differences between the invention and the prior art as explained by Power Integrations' expert in the prior litigation. The actual issue in this case is whether the counter in BCD's circuit is coupled to its DAC (it is) and whether the output of the DAC is coupled to the oscillator (it is).1 Power Integrations further objects to (1) BCD's suggestion that "a sequence of values that repeat in a cyclic manner" that "are not linear" is somehow "pseudo-random" and (2) BCD's characterization of the repeating waveform shown in BCD's presentation as "pseudo-random." The IEEE Dictionary (5th Ed.) defines pseudo-random as "pertaining to the approximation of true, statistical randomness" I The fact that preferred embodiment of Power Integrations' '876 patent shows a monotonic frequency variation signal does not mean the claim is so limited.

Case 1:07-cv-00633-JJF-LPS

Document 88

Filed 05/02/2008

Page 2 of 2

FISH & RICHARDSON P.C.

The Honorable Leonard P. Stark May 2, 2008 Page 2

(see Ex. A), and a sequence of 8 or even 16 numbers that repeats over and over again in the same order (which, coincidentally, is the way BCD implements the patented invention) cannot be characterized as "pseudo-random" under any reasonable definition; nor can it be reasonably asserted to be the same as what is practiced in the prior art. Third, BCD's tutorial suggests that the '851 patent is limited to "an analog circuit for jittering or dithering the switching frequency of a switched mode power supply", but the patent makes clear that there are both analog and digital implementations of the jitter circuitry disclosed in the '851 patent. Specifically, the '851 patent specification explains that frequency variation signals can include "ramp signals, counter output signals or other signals that vary in magnitude during a fixed period of time." ('851 patent, Col. 6, 11. 34-38.) The explicit statement about using a counter output signal as the frequency variation signal demonstrates that the '851 patent covers both digital and analog implementations, as only digital jitter circuits would use counter outputs in the claimed manner. Lastly, PI objects to BCD's suggestion that the oscillator control input of claim 1 of the '876 patent is limited to the summing node 123 shown in Figure 1 of the patent. As noted above with respect to the frequency variation signal of the '851 patent, the disclosure of the preferred embodiment in the '876 patent does not limit the scope of the claimed invention. Power Integrations is also available to address any further questions the Court might have regarding the technology at issue in this case. Respectfully submitted,

cc:

B Steven J lies, Esq. Erik R. Puknys, Esq.

50479818.doc

Case 1:07-cv-00633-JJF-LPS

Document 88-2

Filed 05/02/2008

Page 1 of 4

Exhibit A

Case 1:07-cv-00633-JJF-LPS

Document 88-2

Filed 05/02/2008

Page 2 of 4

Case 1:07-cv-00633-JJF-LPS

Document 88-2

Filed 05/02/2008

Page 3 of 4

Case 1:07-cv-00633-JJF-LPS

Document 88-2

Filed 05/02/2008

Page 4 of 4