Young v. Cree, Inc.

Filing 139

ORDER (1) GRANTING MOTION TO STRIKE AND EXCLUDE DR. GARY ALLEN; (2) DENYING AS MOOT MOTION TO STRIKE AND EXCLUDE MR. STEFAN BOEDEKER; (3) DENYING RENEWED MOTION TO CERTIFY CLASS by Judge Yvonne Gonzalez Rogers ;denying 115 Administrative Motion to File Under Seal; denying 116 Motion to Certify Class; denying 117 Motion to Strike ; granting 118 Motion to Strike ; denying 127 Administrative Motion to File Under Seal. Joint Statement re scheduling of case is due by 2/19/2021. Compliance Deadline set for 2/26/2021 09:01 AM in Oakland, Chambers before Judge Yvonne Gonzalez Rogers. (fs, COURT STAFF) (Filed on 1/28/2021)

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 1 of 24

1
2
3
4
5

UNITED STATES DISTRICT COURT

6

NORTHERN DISTRICT OF CALIFORNIA

7
8

JEFF YOUNG,
Plaintiff,

9
10

United States District Court
Northern District of California

11
12

Case No. 4:17-cv-06252-YGR

v.
CREE INC.,
Defendant.

13

ORDER: (1) GRANTING MOTION TO STRIKE
AND EXCLUDE DR. GARY ALLEN; (2)
DENYING AS MOOT MOTION TO STRIKE
AND EXCLUDE MR. STEFAN BOEDEKER; (3)
DENYING RENEWED MOTION TO CERTIFY
CLASS
Re: Dkt. Nos. 115, 116, 117, 118, 127

14
15

Plaintiff Jeff Young brings this putative class action lawsuit against defendant Cree, Inc.

16

alleging that Cree engaged in an “unfair and deceptive practice of . . . promising consumers” that

17

Cree’s light-emitting-diode (“LED”) bulbs “will last for particularly long periods of time up to

18

35,000 hours” with a “100% Satisfaction Guarantee” and “yearly energy cost savings ranging

19

from around $0.60 to $2 per blub per year” in violation of California’s Unfair Competition Law

20

(“UCL”), Cal. Bus. Prof. Code §§ 17200, et seq. (Count I); California’s False Advertising Law

21

(“FAL”), Cal. Bus. Prof. Code §§ 17500, et seq. (Count II); Consumers Legal Remedies Act

22

(“CLRA”), Cal. Civ. Code §§ 1750, et seq. (Count III); fraudulent misrepresentation and

23

concealment (Count IV); unjust enrichment (Count V); and breach of express and implied

24

warranties (Count VI). (Dkt. No. 48, Amended Class Action Complaint (“ACAC”).)

25

Now before the Court are the following motions: (1) Young’s renewed motion for class

26

certification (Dkt. No. 116); (2) Cree’s motion to strike the report and exclude opinions of Dr.

27

Stefan Boedeker (Dkt. No. 117); and (3) Cree’s motion to strike the report and exclude opinions of

28

Dr. Gary Allen. (Dkt. No. 118.)

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 2 of 24

1

Having carefully considered the pleadings and the papers submitted, as well as oral

2

argument from counsel on July 28, 2020, and for the reasons set forth more fully below, the Court

3

HEREBY ORDERS as follows: (1) Young’s motion for class certification is DENIED; (2) Cree’s

4

motion to strike the report and exclude opinions of Dr. Stefan Boedeker is DENIED AS MOOT; and

5

(3) Cree’s motion to strike the report and exclude opinions of Dr. Gary Allen is GRANTED.

6

I.

BACKGROUND1

7

A.

8

The record evidence reveals the following:

9

Cree is a manufacturer of LEDs and lighting products. Cree launched its line of consumer

Factual Background2

LED lightbulbs in 2013, and includes several different varieties, including A-type, reflector,

11
United States District Court
Northern District of California

10

downlight, surface and disk mount, and candelabra bulbs. Each bulb type has variations for

12

different features, wattages, color temperatures and sizes, which affect a bulb’s thermal

13

properties.3 Despite these differences, Cree believes its LED bulbs are a “premium product” as

14

opposed to a “high-volume average run-of-the-mill product.”

15

In general, LED bulb consumers have been driven, in part, to these products because the

16

overall value of the purchase driven by claims relating to a LED bulb’s extended lifetime or

17

energy efficiency savings. Indeed, studies in the record have confirmed that such consumers are

18

primarily driven by these longevity and cost saving claims based on energy efficiency.

19
20
21
22
23

1

In conjunction with their motion for class certification and reply in support thereof,
Young has filed two Administrative Motions to Seal. (Dkt. Nos. 115, 127.) Both motions are
“brought solely pursuant to Rule 79-5(e)” based on material designated by defendant as
confidential or privileged. (Dkt. Nos. 115 at 2; 127 at 2.) Because Cree, as the designating party,
has failed to file a declaration establishing that all of the designated material is sealable, as
required by Rule 79-5(e)(1), the Court DENIES Young’s administrative motions to seal.

24

2

25

3

26
27
28

The Court omits record citations in this section to expedite the issuance of this Order.

Young avers, citing to his expert witness Dr. Allen, that “[c]osmetic differences in the
shape, color, size, or wattage of the LED Lightbulbs have no bearing in the analysis as to the
class-wide treatment of Cree’s uniform statements regarding longevity.” (Dkt. No. 115-4 at 11.)
Moreover, Dr. Allen contends that the Cree LED bulbs designs permit him to perform a class-wide
assessment to find defects common to the products at issue. The Court discusses these points in
the relevant section below.
2

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 3 of 24

1

Federal law and regulation control testing and, in part, information conveyed to consumers.

2

First, each of Cree’s lines of bulbs was certified as energy efficient under the federal government’s

3

ENERGY STAR program, which requires each bulb to pass industry-standard tests. Second,

4

ENERGY STAR guidelines require manufacturers to include certain information on the bulb itself

5

and the bulb’s packaging and advertising materials, including a standardized “Lighting Facts” box

6

which requires, among other things: (i) the bulb’s initial light output; (ii) estimated annual energy

7

costs in dollars calculated using the average initial wattage, a usage rate of three hours per day,

8

and $0.11 per kWH; (iii) the life of the lamp express in years, calculated by using the bulb’s L704

9

lifetime in hours and assuming operation of three hours a day; and (iv) the wattage for each lamp

10

United States District Court
Northern District of California

11

expressed as “Energy Used” in average initial wattage.
There is some uniformity in the marketing and labeling of the products but the packaging

12

for each of its LED bulb products are not singular. Although each iteration of Cree’s bulb

13

packaging contains the mandatory Lighting Facts box, each product line of Cree’s bulbs has its

14

own distinct packaging with different images and statements, which changed throughout the

15

proposed class period even within each line of bulbs. Notably, no single iteration of Cree’s LED

16

bulb packaging contained all the statements Young identifies below as problematic in the renewed

17

motion for class certification.

18

Young and the proposed class highlight several claims which they assert are on every LED

19

lightbulb at issue: (1) energy and cost savings, such as “82% Less Energy Consumption” and

20

“$226 Lifetime Energy savings”; (2) longevity and performance comparisons to other LED bulbs,

21

such as “Our LED bulbs work better and last longer”; and (3) “100% Satisfaction Guarantee.”

22

Young also points out that Cree’s non-labeling marketing strategy reinforces these claims, with

23

statements including:


24

“[S]ave money now and save money later” by purchasing the “100-watt
Cree LED bulb [that] looks and lights like a traditional incandescent bulb,

25
26
27
28

4

“L70” is the number of hours of operation after which a bulb can be expected to emit at
least 70% of its initial light output, which Cree avers that the industry considers this metric the end
of an LED bulb’s useful life.
3

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 4 of 24

but uses 82% less energy and is designed to last 25 times longer.”

1


2

The bulbs “look and light like an incandescent, while using up to 85% less

3

energy and lasting 25 times longer,” which are “Built to light and last.

4

Covered by Cree’s industry leading 10-year limited warranty.”

5



“Unlike some others, Cree doesn’t compromise”

6



“Long LED lifetime lasts 22+ years (25,000 hours), up to 2x as long as the
cheap LED bulbs.”

7


8

lifetime lasts 22+ years (25,000 hours), up to 6x as long as the cheap LED

9

bulbs.”

10

United States District Court
Northern District of California

11

With respect to its BR40 85W Replacement LED lightbulbs: “Longer LED

Cree has some influence over the pricing of its products through a Manufacturer’s

12

Suggested Retail Price (“MSRP”), but has no further control as to the actual retail pricing of its

13

products. The Home Depot (“THD”) was the exclusive retailer of Cree’s products when Cree

14

began selling its consumer bulbs in 2013. Today, however, while THD accounts for a significant

15

portion of its retail sales (approximately 85%), Cree also sells some bulbs through Amazon.com

16

and specialty lighting website 1000bulbs.com. At these retailers, Cree has no control over

17

whether these entities place its products on sale or clearance, or on how THD merchandises its

18

bulbs within stores. Furthermore, the pricing of Cree consumer bulbs varies from line to line,

19

though, pricing of all of its consumer bulbs has generally decreased over time. The pricing of

20

each product can also be affected by the sale of multipack of bulbs, and rebates offered by various

21

utility companies.

22

For each of Cree’s lines of bulbs, Cree has implemented multiple generational changes

23

between 2013 and today. As a result of Cree’s manufacturing improvements and generational

24

changes, bulbs within the same product line may be from different generations, and bulbs within

25

the same generation may have been manufactured using different processes or components.

26

Retailers do not keep track of generational changes in bulbs, so there is no way to track which

27

generation of bulb each consumer purchased. As retailers control the sale of and stocking of Cree

28

products, consumers may see on the same day different generations of Cree products or packaging.
4

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 5 of 24

1

Cree highlights that consumers may ultimately use the bulbs differently. For instance,

2

consumers may be using bulbs in a manner that is contrary to a bulb’s instructions and warnings.

3

Moreover, consumers may also be selling bulbs on the “gray market”—buying large quantities of

4

bulbs and reselling them online for profit.

5

Cree has also offered consumers various warranties since 2013, the length and terms of
which have changed over time and varies from product line to product line. Those bulbs

7

purchased from THD can return the products to THD subject to their return policy or initiate a

8

warranty claim directly to Cree. Since 2013, Cree’s return and warranty claim rates have both

9

been less than 2%, which Cree asserts is very low by industry standards. Although the specific

10

warranties may have changed, Cree’s customer service department has followed a “satisfaction

11
United States District Court
Northern District of California

6

guaranteed” policy and offers replacement bulbs to dissatisfied customers even if they do not

12

comply with or fall within the warranty terms.

13

To Cree’s knowledge, it is not aware of any product of manufacturing defects that affect

14

the expected lifetime of its consumer bulbs, nor has it been the subject of what it defines as

15

unusually high consumer complaints or warranty returns. In short, Cree avers that its consumer

16

bulbs have never experienced catastrophic failure at an unusually high rate.

17

With regards to Young specifically: In April 2015, Young purchased three 100W

18

replacement dimmable LED lightbulbs from a THD store in Windsor, California, paying

19

approximately $15 to $20 per bulb. Young avers that he reviewed the represents on the label,

20

including those about longevity, cost savings, and warranty, and further recalls viewing television

21

advertisements containing longevity representations. Young inserted these bulbs into his wall

22

sconces, where they burned out within six months to a year of use. In August 2016, Young

23

emailed Cree regarding the failed LED bulbs, but never received a response.

24

Based on the foregoing, Young and the proposed class assert that they paid a premium

25

because of these allegedly false, deceptive and misleading marketing representations regarding

26

longevity. As alleged in the operative complaint, Cree LED “failed well before the promised time

27

frame.” (ACAC ¶ 34.)

28
5

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 6 of 24

1

B.

2

The parties had previously engaged in briefing on class certification and related Daubert

3

motions. (See Dkt. Nos. 79, 85, 93.) On May 28, 2019, the Court heard oral argument on these

4

motions. (See Dkt. No 107.) The following day, for reasons stated on the record at the oral

5

argument, the Court denied without prejudice the motion for class certification, and denied as

6

moot the pending Daubert motions. (See Dkt. No. 109.) The Court thereafter set a briefing

7

schedule on this round of briefing (see Dkt. No. 113), which was later modified due to the onset

8

and continuation of the ongoing coronavirus (COVID-19) pandemic. (See Dkt. No. 124.)

9

Procedural Background

Young now seeks in his renewed motion to certify the following class: “All persons in
California who purchased Cree LED Lightbulbs for end use, and not resale, during the period from

11
United States District Court
Northern District of California

10

March 2013 to the present (the “Class”).” (Dkt. No. 116 at 9-10.) Cree’s opposition contends that

12

Young’s claims fail to satisfy commonality, typicality, and adequacy of representation, in addition

13

to the requirements under section (b)(3). (See generally Dkt. No. 119.) In support of this motion,

14

Young provides two expert reports: first, a report of expert Dr. Stefan Boedeker, who Young hired

15

to “ascertain if it is possible to quantify economic losses to [p]laintiff and [putative] class

16

members” and “provide a framework for the computation of class-wide damages” (Dkt. No. 115-

17

18 (opening report), ¶ 14; see also Dkt. No. 116-2 at 9-170 (rebuttal report)); second, a report of

18

expert Dr. Gary R. Allen, who Young hired “to determine whether a common failure mode exists

19

for a discrete set of LED lamps[.]” (Dkt. No. 115-17 (“Allen Report”), ¶ 2; see also Dkt. No. 126-

20

1 at 3-51 (“Allen Rebuttal”).) Cree seeks to strike the reports of and exclude opinions of Young’s

21

expert witnesses, Boedeker and Allen. (Dkt. Nos. 117, 118.)

22

II.

LEGAL FRAMEWORK

23

A.

24

Federal Rule of Evidence 702 permits opinion testimony by an expert as long as the

25

witness is qualified and their opinion is relevant and reliable. An expert witness may be qualified

26

by “knowledge, skill, experience, training, or education.” Fed. R. Evid. 702. The proponent of

27

expert testimony has the burden of proving admissibility in accordance with Rule 702. Fed. R.

28

Evid. 702, Advisory Committee Notes (2000 amendments). An expert should be permitted to

Daubert Motion Standard

6

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 7 of 24

1

testify if the proponent demonstrates that: (i) the expert is qualified; (ii) the evidence is relevant to

2

the suit; and (iii) the evidence is reliable. See Thompson v. Whirlpool Corp., 2008 WL 2063549,

3

at *3 (W.D. Wash. 2008) (citing Daubert v, Merrell Dow Pharm., Inc., 509 U.S. 579, 589-90

4

(1993) (Daubert I)). Moreover, at the class certification stage, the Court does not make an

5

ultimate determination of the admissibility of an expert’s opinions for purposes of a dispositive

6

motion or trial. Dukes v. Wal–Mart Stores, Inc. (Dukes II), 603 F.3d 571, 602 n. 22 (9th Cir.

7

2010) rev’d on other grounds by 564 U.S. 338 (2011); Millenkamp v. Davisco Foods Int’l, Inc.,

8

562 F.3d 971, 979 (9th Cir. 2009). Rather, the court considers only whether the expert evidence is

9

“useful in evaluating whether class certification requirements have been met.” Tait v. BSH Home
Appliances Corp., 289 F.R.D. 466, 495–96 (C.D. Cal. 2012) (citing Ellis v. Costco Wholesale

11
United States District Court
Northern District of California

10

Corp., 657 F.3d 970, 982 (9th Cir. 2011)); see also Rai v. Santa Clara Valley Trans., 308 F.R.D.

12

245, 264 (N.D. Cal. 2015). At class certification, “the relevant inquiry is a tailored Daubert

13

analysis which scrutinizes the reliability of the expert testimony in light of the criteria for class

14

certification and the current state of the evidence.” Id. For scientific opinions, they must be based

15

on scientifically valid principles. Daubert I, 509 U.S. at 589.

16

B.

17

Under Federal Rule of Civil Procedure 23(a), a court may certify a class only where “(1)

Motion for Class Certification

18

the class is so numerous that joinder of all members is impracticable; (2) there are questions of law

19

or fact common to the class; (3) the claims or defenses of the representative parties are typical of

20

the claims or defenses of the class; and (4) the representative parties will fairly and adequately

21

protect the interests of the class.” Fed. R. Civ. P. 23(a). Courts refer to these four requirements as

22

“numerosity, commonality, typicality[,] and adequacy of representation.” Mazza v. Am. Honda

23

Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012).

24

Once the threshold requirements of Rule 23(a) are met, plaintiffs must then show “through

25

evidentiary proof” that a class is appropriate for certification under one of the provisions in Rule

26

23(b). Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013). Here, plaintiff seeks certification

27

under Rule 23(b)(2) and Rule 23(b)(3). Rule 23(b)(3) requires a plaintiff to establish “that the

28

questions of law or fact common to class members predominate over any questions affecting only
7

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 8 of 24

1

individual members, and that a class action is superior to other available methods for fairly and

2

efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). The predominance inquiry

3

focuses on “whether proposed classes are sufficiently cohesive to warrant adjudication by

4

representation.” Hanlon v. Chrysler Corp., 150 F.3d 1011, 1022 (9th Cir. 1998) (quoting Amchem

5

Prods., Inc. v. Windsor, 521 U.S. 591, 623 (1997)).
“[A] court’s class-certification analysis must be ‘rigorous’ and may ‘entail some overlap

6
7

with the merits of the plaintiff’s underlying claim.’” Amgen, Inc. v. Conn. Ret. Plans & Trust

8

Funds, 568 U.S. 455, 456-66 (2013) (quoting Wal-Mart Store, Inc. v. Dukes, 564 U.S. 338, 351

9

(2011) (Dukes)); see also Mazza, 666 F.3d at 588. The Court considers the merits to the extent
they overlap with the Rule 23 requirements. Ellis, 657 F.3d at 983. The Court must resolve

11
United States District Court
Northern District of California

10

factual disputes as “necessary to determine whether there was a common pattern and practice that

12

could affect the class as a whole.” Id. (emphasis in original). When resolving such factual

13

disputes in the context of a motion for class certification, district courts must consider “the

14

persuasiveness of the evidence presented.” Ellis, 657 F.3d at 982. “A party seeking class

15

certification must affirmatively demonstrate [its] compliance with the Rule.” Dukes, 564 U.S. at

16

350. Ultimately, the Court exercises its discretion to determine whether a class should be

17

certified. Califano v. Yamasaki, 442 U.S. 682, 703 (1979).

18

III.

ANALYSIS

19

The Court first address the pending Daubert motion regarding Dr. Allen given its

20

importance for the underlying motion for class certification. For the reasons set forth below, the

21

Court finds that the Daubert motion as to Dr. Allen is well-taken and that its exclusion from the

22

class certification analysis is fatal to the request to certify the class. Consequently, the Court

23

declines to rule on the Daubert motion as to Dr. Boedeker as the issue is moot.

24

A.

25

Young primarily relies on Dr. Allen to identify alleged defects that are common to the

Daubert Motion as to Dr. Allen

26

architecture of the Cree LED lightbulbs in order to support his burden of the commonality factor

27

under Rule 23(a) and the predominance inquiry under Rule 23(b)(3). Cree does not challenge Dr.

28

Allen’s credentials, experience, or his background. Instead, Cree focuses on his evaluations and
8

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 9 of 24

1

assessments, including as to the procedures, assumptions and ultimate opinions. The Court

2

summarizes his background and the relevant substance of both his opening and rebuttal reports, as

3

well as other relevant evidence in the record, before analyzing his reports under the Daubert

4

standard.

5
6

1.

Relevant Background and Summary

Dr. Allen has a Ph.D. and a Masters of Science in astrophysical sciences from Princeton
University and in addition to degrees from Pennsylvania State University. (Allen Report ¶ 5.) Dr.

8

Allen conceived and designed the world’s first LED replacement lamp to be certified under the

9

ENERGY STAR program. (Id. ¶ 4.) As described in depth in his report, he has decades of

10

experience with research and industrial institutions in the fields of lighting, applied physics,

11
United States District Court
Northern District of California

7

product design, and patent generation and analysis as the principal engineer and engineering

12

manager of GE Lighting and as a research physicist with Sylvania Lighting. (Id. ¶¶ 6-10.) Dr.

13

Allen has primarily been an expert witness in patent infringement cases, but has also testified as an

14

expert witness in a breach of a non-disclosure agreement matter. (Id. ¶ 1.)

15

As disclosed in his expert report, Dr. Allen was hired “to determine whether a common

16

failure mode exists for a discrete set of LED lamps[,]”5 which he identifies as thirty Cree LED

17

bulbs.6 (Id. ¶ 2.)

18

While Dr. Allen’s report is quite technical, his fundamental methodology is quite

19

simplistic. To conduct his analysis, Dr. Allen grouped together similar Cree LED bulbs, tested a

20
21
22
23
24
25
26
27
28

5

Dr. Allen utilizes the term “lamp” in its technical meaning used in the lighting
manufacture industry which defines “lamp” as “a self-contained light source powered by
electricity and producing light” which is “equivalent to what is commonly called a “light bulb.”
(Id. ¶ 12.) Dr. Allen further clarifies that he defines an “LED lamp” as a lamp whose light is
produced by an LED in contrast to an LED which “is generally contained inside the [l]amp.” (Id.)
Dr. Allen also defines “fixture” as “hardware that holds the [l]amp in place in its application,”
more commonly recognized as a “table lamp, floor lamp, desk lamp, etc. having an Edison-style
socket that the lamp screws into [sic].” (Id.) The Court provides these defined terms to clarify
quoted sections from Dr. Allen’s report. This Order uses the term “bulb” or “lightbulb” and
“lamp” interchangeably unless otherwise noted.
6

Dr. Allen clarifies that his opinions only extend insofar as those lamps listed in Table 2
of his opening report. (See Allen Rebuttal ¶¶ 1 n.1, 18.) Cree notes that it makes more LED
lamps than those identified in this table. (See Dkt. No. 132-2 (“Vollers II Decl.”) ¶ 59.)
9

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 10 of 24

1

fraction of them, and then extrapolated his conclusions across a wide array of Cree products.

2

Thus, Dr. Allen reviewed the packaging of 43 types of LED bulbs sold during the class period,

3

and sorted them by manufacturer part number. (Id. ¶ 12.) Next, he narrowed the 43 bulbs down to

4

30 on the ground that each of the 30 provided either a 5- or 10-year warranty and had some type of

5

claimed lifetime energy savings. (Id.)

6

Dr. Allen then grouped the 30 by focusing on the design elements related to reliability and
useful life of the product in consumer applications. (Id.) From this grouping, Dr. Allen concluded

8

that many of these designs were “functionally equivalent” designs that share the following

9

attributes: bulb shape, lamp architecture, lumens, wattage equivalent, LED wattage, and rated

10

lifetime. (Id. ¶ 14.) Dr. Allen first categorized the 30 Cree LED bulbs into seven “Cree LED

11
United States District Court
Northern District of California

7

Lamp Designs.” (Id.) Thereafter, Dr. Allen further narrowed the groupings into four “LED Lamp

12

Design Architectures” “based on their fundamental design architecture and visual appearance”:

13

(1) the FT21 Design Architecture; (2) the FT19 Design Architecture; (3) the 4F19 Design

14

Architecture; and (4) the BRHS Design Architecture. (Id. ¶ 15.)7 Notably, Dr. Allen’s

15

categorizations were not based on any physically inspection of even a single Cree LED lamp, but

16

rather resulted from reviewing photographs or PDF images. (Id. ¶¶ 12-16; Dkt. No. 120-10

17

(Richter Decl. Ex. 10) (“Allen Dep.”) at 40:18-42:14.)8

18

Dr. Allen then moved into a testing phase. He purchased approximately ten Cree LED

19

bulbs from third-party internet sellers,9 which he contends are representative of the seven LED

20
21
22
23

7

Dr. Allen notes that at the time of his opening report, two of the seven Cree LED Lamp
Designs, the FT18 and the FT9.0, were not commercially available. (Id. at ¶ 16.) Dr. Allen noted
that the FT18 is functionally equivalent to a FT18/3w (3-way lamp) that is functionally equivalent
when operating on the 100 W setting. (Id.) For the FT9.0, Dr. Allen notes there is a FT9.5 that
has 0.5W difference in operating power, which he discounts in his later analysis. (Id.)

24
8

25
26
27
28

The use of photographs and the narrowing of designs into design groups and then
further into architecture design groups are neither an industry standard or industry practice, nor is
it based upon any peer reviewed or generally scientific or technical principles. (See Allen Report
¶¶ 12-16; Allen Dep. at 41:5-42:14; Dkt. No. 119-2 (“Vollers I Decl.”) ¶¶ 18, 20.)
9

Dr. Allen provides no evidence on the storage conditions of the LED lamps prior to
purchase, nor is he aware of whether these lamps were used previously, although he testified that
these LED lamps were in the original packaging at the time of purchase. (Allen Dep. at 58:310

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 11 of 24

1

Lamp Design and four LED Lamp Design Architectures. The ten Cree LED bulbs represent five

2

of the seven identified Cree LED Lamp Designs, specifically: four FT19 lamps (two from the

3

FT9.5 Lamp Design, and two from the FT6 Lamp Design), two FT21 lamps (FT18/3w Lamp

4

Design), two 4F19 lamps (4F10 Lamp Design), and two BRHS lamps (BR30 Lamp Design), all

5

manufactured in different years from 2013 to 2016. (Allen Report ¶¶ 3, 15, 16, 42 (Table 6).) Dr.

6

Allen did not take into consideration the various generations of each Lamp Architecture, and

7

instead focused solely on samples from only one or two manufacturing weeks in 2013 and 2016.

8

(Allen Dep. at 44:18-45:14; 88:12-89:6.)
Dr. Allen subjected these ten bulb samples to: (1) a visual inspection of the LED bulbs;

9

(2) temperature measurement and testing; and (3) a comparison to market competitors to

11
United States District Court
Northern District of California

10

corroborate the results of his testing. (Allen Report ¶¶ 12-16, 17, 62-69.) In considering the data

12

generated by his tests, he compared his own data with Cree’s internal laboratory test results,

13

including those showing temperature data. (Id. ¶ 20.)
With respect to the thermal tests, Dr. Allen took infrared images and temperature

14
15

measurements of certain basic components and subsystems that are found in all LED lamps,

16

including the LEDs themselves, drivers, electrolytic capacitors, heat sink, and the envelope or

17

optics. (Id. ¶¶ 17-19.) Specifically, Dr. Allen dissembled one of the two LED lamps that he

18

purchased for each available Lamp Design, and recorded temperature measurements on LED

19

lamps operating for approximately 2.5 hours at: (i) the heat sink; (ii) electrolytic capacitor; and

20

(iii) the printed circuit board (“PCB”), upon which the LEDs are soldered. (Id. ¶ 42.) Dr. Allen

21

performed these measurements in 25° C (75° F) ambient air, and thereafter extrapolated those

22

temperatures to expected performance at 45° C (113° F) ambient air. (Id. ¶¶ 41-46.) During his

23

deposition, Dr. Allen conceded that to adequately test the expected lifetime of Cree LED lamps, he

24

would have needed to perform extended reliability testing on a large number of Cree LED lamps.

25

(Allen Dep. at 18:6-15.) He did not do; nor did he explain why. Importantly, Dr. Allen’s report

26

did not reveal that any of the Cree LED lamps that he tested catastrophically failed during his

27
28

59:11.)
11

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 12 of 24

1

inspection.10

2

Instead, Dr. Allen merely created a purported “methodology” to compare the temperature

3

measurements recorded after 2.5 hours of use to what he refers to as a “Target Maximum.” This

4

purported methodology was specifically developed for this litigation and is not based on any

5

industry standard, peer-reviewed analysis, or generally accepted scientific or technical principles.

6

(Allen Report ¶¶ 22, 23; Allen Dep. at 111:24-112:13.)

7

By way of background, in the lighting industry, component manufacturers typically

8

publish specification sheets that state the maximum operating temperature at which the component

9

should operate in order to achieve a target lifetime. (Vollers I Decl. ¶¶ 35-39.) Dr. Allen’s Target
Maximums are as much as 20° C, or 68° F, below the rated temperature for these components.

11
United States District Court
Northern District of California

10

(Allen Report ¶¶ 42-45.) The “Target Maximums” are based solely on Dr. Allen’s own

12

conservative, personal design opinions for good guidelines. He concedes that no publication or

13

source reflects any maximum temperature threshold. (Allen Report ¶¶ 22, 23; Allen Dep. at

14

111:24-112:13.)11

15

Dr. Allen concluded, based on the foregoing categorizations and temperature analyses that,

16

despite cosmetic variation, all the LED bulbs run “too hot” which stresses the electronic

17

components beyond what they can handle and results in premature “failures” and shortened

18
19
20
21
22
23
24
25
26
27
28

10

The parties briefing and the record reflects that there are generally two types of failures:
the first, catastrophic failure, is where the LED lamp fails completely. The second, parametric
failure, includes LED lamp failure on other metrics, including lumen depreciation, color shift, and
flicker. Importantly, Dr. Allen does not provide a methodology to evaluate LED lamps for
failures on a class wide basis, and concedes that some people may have increased ability to
visually detect flicker in lights than others. (See Allen Rebuttal ¶ 80; Vollers II Decl. ¶ 71.)
11

For example: Dr. Allen opines that he believes the LED junction temperature, which is
measured at the PCB onto which the LEDs are soldered, should not exceed 105° C for Cree LED
lamps, despite that the LEDs used in the evaluated lamps are rated to exceed 25,000-hour
ENERGY STAR requirements at 125° C. (Allen Report ¶ 26.) As for the electrolytic capacitor,
Dr. Allen selected either 72° C or 82° C as the Target Maximum even though these capacitors that
were evaluated have a 25,000 hour rated lifetime at 92° C. (Allen Report ¶ 27; Dkt. No. 132-1
(“Fang II Decl.”) ¶¶ 17-18.) For the heat sink, Dr. Allen noted that there is no industry standard
for the maximum operating temperature of a heat sink, but that he adopted a “personal guideline”
to strive to keep the heat sink temperature below 75° C in order to keep the “touch point,” or
temperature of the exterior of an operating bulb, sufficiently low, for the sole purpose of ensuring
that a consumer can comfortably touch the lamp’s exterior. (Allen Report ¶ 28.) How touch is
relevant is never made clear.
12

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 13 of 24

1

lifespans. (Allen Report ¶¶ 3, 24, 42 & Table 6.) Specifically, Dr. Allen concluded that some of

2

the Cree LED lamps would exceed the Target Maximums when operated at 25° C (75° F), and

3

many more would exceed his Target Maximums when he extrapolated measurements to 45° C

4

(113° F). (Id. ¶¶ 3, 43-46.)
Utilizing these measurements, Dr. Allen calculated the difference between the expected

6

lifetime of a Cree LED lamp and his conservative target design that would have used his Target

7

Maximums by applying what he refers to as a “Rule of Thumb,” namely “a[n operating]

8

temperature increase of 10C is generally accepted as cutting the life of [an electronic component

9

in half.” (Id. ¶ 22.) In applying this rule of thumb, Dr. Allen focused particular attention on the

10

electrolytic capacitor, and opined that the operating temperature of Cree LED lamps would cause

11
United States District Court
Northern District of California

5

the electrolytic capacitors to fail prematurely. (Id. ¶ 46.) Dr. Allen found that the premature

12

failure is so significant that the LED bulbs will “commonly experience failure in advance of the

13

advertised product life ([less than one] year)” with a “failure rate [that] would be abnormally

14

high.” (Id. ¶ 3; Allen Dep. at 19:17-18.)12

15

While Dr. Allen found that Cree LED bulbs would commonly experience failure within

16

one year, Dr. Allen’s report has noteworthy caveats. First, Dr. Allen could not quantify the

17

number or percentage of Cree LED Lamps that would, in his opinion, fail in less than one year.

18

(Id. at 17:25-18:20, 22:5-17.) Indeed, while Dr. Allen avers that his report confirms that there is a

19

defect, his report assumes an inference that Cree LED lamps are defective based on the allegations

20

of this lawsuit. (Id. at 83:19-84:24.) Second, Dr. Allen did not reconcile or address Cree’s own

21

reliability testing results, reflecting that Cree LED lamps perform within stated temperature

22

tolerances and are not expected to commonly experience premature failures. (Id. at 78:4-79:12.)13

23
24
25
26
27
28

12

Although Dr. Allen does not elaborate in his report what “failure” the defect causes, the
record reflects that, in general: (1) an LED failure would mean that an LED lamp would emit less
light, or color shift, which means that the color temperature of the Lamp may shift warmer
(Vollers I Decl. ¶ 57); and (2) a failure of the electrolytic capacitor would result in a flicker that
most, but not all, humans cannot perceive. (Dkt. No. 119-4 (“Fang I Decl.”) ¶¶ 21-24.) Dr. Allen
did testify in his deposition that higher temperatures of the electrolytic capacitor “would mean
higher stress on [it], and likely a shorten lifetime.” (Allen Dep. at 107:1-3.)
13

Dr. Allen did not address the evidence reflecting a low return and warranty claim rate on
Cree LED lamps. (Id.at 20:9-21:9; 91:25-92:8; 96:5-97:3.)
13

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 14 of 24

Finally, in Dr. Allen’s rebuttal report, he introduces additional arguments in support of his

2

opening report, including: (1) citing to a new methodology and studies that subject LED lamps to

3

on-off switching that Dr. Allen avers proves that Cree LED lamps will fail prematurely; and

4

(2) performed statistical analysis of THD’s records of returned Cree LED lamps. (Allen Rebuttal

5

¶¶ 38-44, 76-77.) Further, Dr. Allen withdrew two lamp types set as defective from his analysis:

6

specifically, Dr. Allen states that he “did a separate thermal analysis of the 40 W and 60 W version

7

of the FT19 and found that the 40 W version” – which Dr. Allen refers to as the FT6.0 lamp type –

8

“is not overheated enough for [him] to opine that it will fail prematurely.” (Id. ¶¶ 21, 78; see also

9

Vollers II Decl. ¶ 64 (noting Dr. Allen’s err in assuming similar thermal problems with different

10

wattage LED lamps).). Additionally, Dr. Allen clarified that although he included the lamp type

11
United States District Court
Northern District of California

1

FT9.0 in his FT19 Lamp Architecture Group, he does not actually have an opinion about that

12

lamp, stating: “There are no Opinions provided for the FT9.0 because there are no samples

13

available for procurement of that LED Lamp.” (Allen Rebuttal ¶ 64.)

14
15

2.

Analysis

Cree challenges two of Dr. Allen’s methodologies and assessments made in his opinions:14

16

(a) the LED Lamp Design and LED Lamp Design Architecture methodology and classification;

17

and (b) the Target Maximum methodology and conclusions. Cree further moves to strike new

18

opinions and methodologies that were absent in Dr. Allen’s Rebuttal, namely: (i) Dr. Allen’s

19

citations to new methodology and studies that subject LED lamps to on-off switching that he

20

claims proves Cree LED lamps will fail premature; and (ii) Dr. Allen’s purported statistical

21

analysis of the THD’s records of returned Cree LED lamps. The Court addresses each in turn.15

22
14

23
24
25
26
27
28

Cree does not challenge Dr. Allen’s credentials. Dr. Allen is a well-credentialed,
knowledgeable, and experienced individual in the area of LED lamp design.
15

As an initial matter, Young seeks to exclude declarations from Jonathan Vollers and
Yuan Frank Fang on the basis that Cree did not attempt to qualify them or submit their testimony
as experts. With respect to Vollers, the Court DENIES this request where Young did not object to
Vollers earlier declaration during the briefing on the first motion for class certification (see Dkt.
No. 86-3), and, being on notice of the same, has not shown any prejudice or denial of a deposition.
With respect to Fang, the Court DENIES AS MOOT this request, where Fang’s declarations do not
materially impact the Court’s analysis, and citations to Fang’s declarations otherwise are in
addition to citations from other sources in the record.
14

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 15 of 24

1
2

a.

LED Lamp Design and LED Lamp Design Architecture

With respect to the LED Lamp Design and LED Lamp Design Architecture methodology

3

and classification, Cree argues that Dr. Allen’s analysis is not based on any accepted methodology

4

and was solely created for purposes of this litigation. Young disagrees claiming that Dr. Allen’s

5

categorization is based on solid engineering principles and does not need to be based on a peer-

6

reviewed methodology.

7

To determine whether a scientific opinion is sufficiently reliable to be admissible, the
Court must “analyze not what the expert[s] say[]s, but what basis they have for saying it.”

9

Daubert v. Merrell Dow Pharm., Inc., 43 F.3d 1311, 1216 (9th Cir. 1995) (Daubert II). Scientific

10

methodology can be reliable if: (i) the methodology can and has been tested; (ii) the methodology

11
United States District Court
Northern District of California

8

has been subjected to peer review; (iii) the known or potential rate of error for the technique has

12

been addressed; or (iv) the methodology has a general degree of acceptance in the relevant

13

scientific community. Daubert I, 509 U.S. 593-94. An expert must “employ[] in the courtroom

14

the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.”

15

Kumho Tire Co. v. Carmichael, 526 U.S. 137, 150 (1999). In some situations, “[p]eer reviewed

16

scientific literature may be unavailable because the issue may be too particular, new, or of

17

insufficiently broad interest to be in the literature.” Primiano v. Cook, 598 F.3d 558, 565 (9th Cir.

18

2010). In those circumstances, expert opinion testimony “is reliable if the knowledge underlying

19

it has a reliable basis in the knowledge and experience of the relevant discipline.” Id. Of course,

20

evaluating the design and predicting the lifetime of an electrical system is, however, a scientific

21

inquiry that can be proven by actual data. Kumho Tire, 526 U.S. at 150 (pointing to engineering

22

as the type of testimony that “rests upon scientific foundations”).

23

Dr. Allen’s analysis of LED Lamp Designs and LED Lamp Design Architectures has not

24

shown to be at the appropriate level of intellectual rigor required to meet the Daubert standard.

25

Importantly, the issue is not reviewed in a vacuum. The record is replete with the business

26

community, government regulators, and academics designing for, testing, and projecting long-

27

term LED lamp lifetimes, and many groups spending years investigating processes and procedures

28

subject to peer review, public comment, and government approval with respect to these issues.
15

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 16 of 24

1

(See Vollers II Decl. ¶ 86; Dkt. No. 132-3 (“Pattinson Decl.”) ¶¶ 12-14.)16 Young’s explanation

2

that there is no relevant literature or peer review studies to cite because there is very little interest

3

in testing LED lamps and that Dr. Allen is relying on his wealth of experience and his knowledge

4

of engineering best practices is not supported. Despite the wealth of analysis, there is a total

5

absence of a process similar to that of Dr. Allen. Even if one could justify the expediency in

6

categorizing LED lamps by merely reviewing packaging and photos, Young has failed to show

7

that Dr. Allen’s methods of broad extrapolations and temperature measurements are “generally

8

accepted in the relevant engineering community,” especially where the comparative metric is self-

9

created.17 Nor is it probable that he can. Kumho Tire, 526 U.S. at 151.

10

Dr. Allen’s categorization impermissibly and overbroadly sweeps in different models and

United States District Court
Northern District of California

11

versions of LED lamps for which he has done the actual testing. Indeed, Dr. Allen’s

12

categorization from 42 identified packaging, to 30 identified LED Lamps, to 7 LED Lamp

13

Designs, to 4 LED Lamp Design Architectures appears to be nothing more than a naked attempt to

14

take the findings and conclusions made from the ten (10) purchased LED lamps, which consisted

15

of only four unique (4) LED lamp types, and extend these findings and conclusion to a broad set

16

of LED lamps. Dr. Allen nor Young have shown that this is a generally accepted practice within

17

the lighting community, especially where such tests and measurements already exist for individual

18

LED lamps.

19

Moreover, Dr. Allen’s Rebuttal Report itself reveals the error of being so overinclusive.

20

Indeed, the Court is generally aware of the foundational principle that more power and wattage

21

would typically mean higher temperatures for that electrical system. That Dr. Allen ignored this

22

in his Opening Report when making his categories reveals that Dr. Allen overlooked and failed to

23
24
25
26
27
28

16

As one classification example, ENERGY STAR distinguishes LED lamps through
wattage, whereby a manufacturer must seek new qualification if it changes the input wattage of a
qualified lamp by more than 10%. (Vollers II Decl. ¶¶ 65-66; Pattison Decl. ¶ 21.)
17

For instance, the studies referenced in the record show testing on individual LED lamps
and conclusions being drawn for that particular LED lamp. Dr. Allen does not explain why he did
not or could not undertake a similar endeavor of individually testing each LED lamp to arrive at a
conclusion for each LED lamp of which the class finds at issue.
16

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 17 of 24

1

apply this basic understanding when creating this novel methodology. It is unsurprising therefore

2

that Dr. Allen concedes in his Rebuttal Report that some of the lower wattage LED lamps in some

3

of the Designs and Architectures do not in fact fail, despite placing them within the same groups

4

as higher wattage models and versions.18 Dr. Allen’s post hoc withdrawal does not save the initial

5

analysis. Experience and engineering practices alone do not convert a made-up methodology into

6

one that meets the threshold requirements of admissibility. See Grodzitsky v. American Honda

7

Motor Co.,Inc., 957 F.3d 979, 986-87 (9th Cir. 2020) (affirming exclusion of design defect expert

8

offered in support of class certification where the expert’s testimony suffered from scientific and

9

methodological flaws despite being based on “general product engineering principles”).

10

United States District Court
Northern District of California

11

Accordingly, the Court finds that exclusion of Dr. Allen’s opinions as to the LED Lamp
Designs and LED Lamp Design Architectures is appropriately excluded under Daubert.19

12
13

b.

Target Maximums

With respect to the Target Maximums, Cree challenges the analysis on two grounds. First,

14

Cree contends that Dr. Allen’s Target Maximums are an arbitrary conservative preference

15

untethered to any industry standard or accepted research. Second, the Target Maximum analysis

16

offers no causal link between the purported defect and liability and fails to quantify any specific

17

number of failures from Cree LED lamps. In opposition Young disputes these grounds with

18

similar arguments as in the prior section. Namely, Young avers that Dr. Allen’s opinions are

19

supported by and based on engineering best practices related to design for reliability, and that

20

there are no industry standards or regulations concerning the temperatures of the electronic

21

components of LED lamps.

22
23

Young does not persuade. First, the record reflects that the Target Maximums are not a
commonly accepted practice within the lighting community and are merely a reflection of Dr.

24
25
26
27
28

18

Additionally, as Cree highlights, Dr. Allen’s methodology ignores the fact that some of
Cree’s LED lamps have undergone some changes and updates throughout the class period.
19

Were this, Dr. Allen’s overbroad categorization, the only flaw in Dr. Allen’s opinions,
the Court would have considered his results as to those LED lamps that he actually tested.
However, as will be discussed in the next section, given the problems with his Target Maximums,
Dr. Allen’s opinions are appropriately stricken.
17

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 18 of 24

1

Allen’s personal design preference. Dr. Allen must “explain precisely how he went about

2

reaching his conclusions and point to some objective source—a learned treatise, the policy

3

statement of a professional association, a published article in a reputable scientific journal, or the

4

like—to show that he has followed the scientific method, as it is practiced by (at least) a minority

5

of scientists in his field.” Daubert II, 43 F.3d at 1318-19; United States v. Rincon, 28 F.3d 921,

6

924 (9th Cir. 1994) (research must be described “in sufficient detail that the district court [can]

7

determine if the research was scientifically valid.”).

8
9

Here, Dr. Allen admits that there is no industry standard for maximum temperature of a
heat sink, and he does not cite to any sources that recommend temperature thresholds. (Allen
Report ¶ 28.) Dr. Allen also admits that these Target Maximums, while based on a general

11
United States District Court
Northern District of California

10

engineering rules of thumb, are not industry standard, but rather are a personal preference

12

reflecting his conservative design targets. (Id. ¶¶ 12, 26; Allen Dep. at 109:11-24; 111:6-112:13.)

13

Moreover, Dr. Allen concedes that he does not cite to one source that supports his Target

14

Maximum analysis. (Allen Dep. at 111:-24-112:13.) Indeed, Dr. Allen states that he developed

15

this methodology solely for the purpose of this litigation and that it is not peer reviewed. (Allen

16

Report ¶ 12; Allen Dep. at 42:11-14.) Cree’s witnesses further testified that they have never heard

17

of Dr. Allen’s target methodology, and that the industry standard methodology for LED lamp

18

design permits LED lamp designers to rely upon the component manufacturers’ specifications

19

sheets for components’ rated lifetimes. (See, e.g., Vollers I Decl. ¶¶ 17-20.)

20

This is not a situation where there is absence of testing or literature in the relevant field.

21

As previously discussed, there are accepted LED lamp lifetime projection tests, and commonly

22

accepted processes and procedures in conducting such tests. (See Vollers II Decl. ¶ 86; Pattinson

23

Decl. ¶¶ 12-14.) There are ample accepted peer-reviewed and vetted industry standards for

24

reliability testing based on long-term durational testing of a large number of LED lamps, including

25

LM-80-08, LM-84, and ENERGY STAR program requirements whereby such qualification

26

requires 6,000 hours of testing. (Vollers I ¶ 27-30; Vollers II ¶¶ 79-86; Pattison ¶¶ 15-18, 25, 41,

27
28
18

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 19 of 24

1

42, 53.)20 That Dr. Allen disagrees with these standards does not mean that they are irrelevant,

2

especially where Dr. Allen’s alternative methodology is not based on any accepted practices, and

3

are, in fact, the opposite of these practices. (See Allen Rebuttal ¶¶ 24-26, 32-34, 40, 45, 62, 68)

4

“Something doesn’t become ‘scientific knowledge’ just because it’s uttered by a scientist; nor can

5

an expert’s self-serving assertion that his conclusions were ‘derived by the scientific method’ be

6

deemed conclusive.” In re Ford Motor Co. DPS6 Powershift Transmission Prods. Liab. Litig.,

7

No. ML 18-02814 AB, 2019 WL 7177984 at *1 (C.D. Cal. Dec. 2, 2019) (citing Daubert II, 43

8

F.3d at 1318). In short, Dr. Allen’s personal design preferences cannot form the basis of an

9

admissible expert opinion given the context of the industry. See Daubert II, 43 F.3d at 1319
(expert’s personal opinion is not sufficient to meet Daubert standards); Claar v. Burlington R. Co.,

11
United States District Court
Northern District of California

10

29 F.3d 499, 502 (9th Cir. 1994).

12

Second, the Target Maximum analysis offers no causal link between the purported defect

13

and liability and fails to quantify any specific number of failures from Cree LED lamps. Dr. Allen

14

concludes that there is a higher risk of failure due to the heat on certain components of the LED

15

lamp. However, not only does Dr. Allen not define failure, he cannot quantify, or even

16

approximate, the specific failure rate. See Gonzalez v. Corning, 885 F.3d 186, 198 (3d Cir. 2018)

17

(noting that liability determination in purported design defect class action where alleged defect is

18

not latent in all products would require “distinguish[ing] [the products] that are likely to fail before

19

the end of their warranty periods from [the] ones that are likely to perform as expected (i.e., that

20

are not defective)”).

21
22
23
24
25
26
27
28

20

These tests, in connection with the processes explained by Cree’s witnesses, would not
only yield results in lifetime projections, but would allow individuals to determine the cause or
causes of failure. These tests do not assume a defect, but rather seek to test the LED lamp for its
performance.
By contrast, and as discussed, Dr. Allen concedes he has assumed a defect based merely on
the filing of this lawsuit by Young, and then employs a methodology that assumes this premise:
that there is in fact a defect in the Cree LED lamps. Instead of seeking to determine whether a
defect exists, Dr. Allen merely seeks to explain where the assumed defect would be. In other
words, Dr. Allen does not necessarily prove the existence of a defect: only that, assuming there is
one, it is likely due in his opinion to the high temperature of the LED lamps at certain
components.
19

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 20 of 24

1

Nor does Dr. Allen link the defect with the ultimate liability in this action. As Cree’s

2

witnesses explain, a failure of the components identified by Dr. Allen would more often lead to a

3

parametric failure than a catastrophic failure—the latter of which is the failure that is pled in this

4

action. (See ACAC ¶¶ 34, 35, 37, 41.) Unlike a catastrophic failure, a parametric failure is highly

5

subjective, depending on the individual’s sensitivity to lumen depreciation, color shift, and flicker.

6

(See Vollers II Decl. ¶¶ 68-71; Pattinson Decl. ¶¶ 48-50.) Dr. Allen’s opinions focus on a

7

parametric failure while overlooking the actual pled failure: catastrophic.

8
9
10

United States District Court
Northern District of California

11

Accordingly, the Court finds that exclusion of Dr. Allen’s opinions as to Target
Maximums is appropriately excluded under Daubert.
c.

Additional Considerations

Cree further moves in its reply to strike from Dr. Allen’s Rebuttal his citation to new

12

methodology and studies concerning on-off switching that he asserts prove Cree LED lamps will

13

fail and statistical analysis performed on the THD records of returned Cree LED lamps.

14

Specifically, Cree highlights that these new opinions and testing do not rebut any of its expert

15

witness opinion, and that it was not properly raised in the opening report. Additionally, Cree

16

contends that the Dr. Allen has no background or expertise in statistics, misconstrues the cited

17

studies, and reaches an incorrect conclusion with his statistical analysis.

18

The Court agrees. As an initial hurdle, Dr. Allen’s opinions incorporating these additional

19

citations and arguments were not properly raised or disclosed in his opening brief, and they do not

20

respond to any of the opinions of Cree’s expert witnesses. See Matthew Enterprise, Inc. v.

21

Chrysler Group, LLC, No. 13-cv-04236-BLF, 2016 WL 4272430 at *2 (N.D. Cal. Aug. 15, 2016

22

(“[R]ebuttal expert testimony is limited to new unforeseen facts brought out in the other side’s

23

case.”); Huawei Techs., Co., Ltd. v. Samsung Elec. Co., Ltd., 340 F. Supp. 3d 934, 995 (N.D. Cal.

24

2018) (“The rest of whether an expert’s opinion constitutes rebuttal or a new opinion . . . [is]

25

whether a rebuttal attempts to put forward new theories outside the scope of the report it claims to

26

rebut.”); Shenon v. New York Life Ins. Co., No. 2:18-cv-240-CAS, 2020 WL1317722, at *10 (C.D.

27

Cal. Mar. 16, 2020 (finding that attempts to “add new opinion, or deepen or strengthen existing

28

opinions” are not proper on rebuttal); see also Fed. R. Civ. P. 26(a)(2)(D); Fed. R. Civ. P.
20

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 21 of 24

1

37(c)(1). Indeed, as Dr. Allen concedes, “Mr. Voller [Cree’s employee] did not address on-off

2

switching, nor testing in luminaries that represent consumer applications.” (Allen Rebuttal ¶ 45.)

3

For this plain reason, they are inappropriate and should indeed be stricken.21

4

Finally, while Dr. Allen might be expert in statistical analysis, the record does not support

5

the conclusion. Under Federal Rule of Evidence 702,“[t]he burden is on the proponent of the

6

expert testimony to show, by a preponderance of the evidence,” that the expert is qualified and

7

meets the necessary admissibility requirements. Contour IP Holding LLC v. GoPro Inc., 2021

8

WL 75666, at *10 (N.D. Cal. Jan. 8, 2021) (citing Lust By & Through Lust v. Merrell Dow

9

Pharm., Inc., 89 F.3d 594, 598 (9th Cir. 1996)). An expert’s qualifications in one discipline do
not make him qualified in a separate field. In short, Young has not shown that Dr. Allen is

11
United States District Court
Northern District of California

10

qualified to perform that type of statistical analysis on the THD materials that he has done. On

12

this basis too is Dr. Allen’s new opinions made in his rebuttal also appropriately stricken.

13
14
15

d.

Conclusion

In sum, for the foregoing reasons, the Court concludes that Dr. Allen’s opinions do not
meet the Daubert standard. Accordingly, the Court GRANTS the Daubert motion as to Dr. Allen.22

16

B.

17

Young moves to certify a California class of consumers who have purchased specific Cree

18

Motion for Class Certification

LED bulbs. In opposition, Cree challenges the commonality, typicality, and adequacy factors of

19
20
21
22
23
24
25
26
27
28

21

Indeed, even considering the cited California Public Utilities Commission (“CPUC”)
study, it is not facially apparent whether Make #5, a model with some issues, is a Cree LED lamp
where the study anonymized the subject lamps. As Cree’s employee, Vollers, declared, he could
not definitively conclude that Cree is the manufacturer of Make #5 as some of the specifications
do not match up with attributes of its consumer products. (Vollers II Decl. ¶¶ 34-37.) Dr. Allen
was not involved in this CPUC study and Dr. Allen does not otherwise explain why he arrived at
his conclusion that Make # 5 is a Cree lamp. Instead, Dr. Allen has made an unfounded
assumption without explanation. See Diviero v. Uniroyal Goodrich Tire Co., 114 F.3d 851, 853
(9th Cir. 1997) (“Rule 702 demands that expert testimony relate to scientific, technical, or other
specialized knowledge, which does not include unsubstantiated speculation and subjective
beliefs.”). Furthermore, neither new cited study reflects on-off cycles that are common or typical
in the field. (See Vollers II Decl. ¶¶ 40, 44, 51.)
22

As the Court discusses below, without Dr. Allen’s expert testimony, Young cannot
satisfy the necessary elements for class certification, even without consideration of Cree’s Daubert
motion as to Dr. Boedeker. Accordingly, the Court DENIES AS MOOT the Daubert motion as to
Dr. Boedeker.
21

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 22 of 24

1

Rule 23(a), in addition to the requirements under Rule 23 (b)(3). The Court begins by addressing

2

the commonality factor.23 Thus:

3

Under Federal Rule of Civil Procedure 23(a), a court may certify a class only where “(1)

4

the class is so numerous that joinder of all members is impracticable; (2) there are questions of law

5

or fact common to the class; (3) the claims or defenses of the representative parties are typical of

6

the claims or defenses of the class; and (4) the representative parties will fairly and adequately

7

protect the interests of the class.” Fed. R. Civ. P. 23(a). Young must demonstrate that these four

8

requirements are met by a preponderance of the evidence. Here, Young fails to establish that there

9

are common questions of law or fact.

10

The commonality requirement is satisfied where the claims of all members of the proposed

United States District Court
Northern District of California

11

class “depend upon a common contention . . . of such a nature that it is capable of classwide

12

resolution—which means that determination of its truth or falsity will resolve an issue that is

13

central to the validity of each one of the claims in one stroke.” Dukes, 564 U.S. at 350. Courts

14

generally construe the commonality requirement “permissively” and find that “[a]ll questions of

15

fact and law need not be common to satisfy the rule.” Ellis, 657 F.3d at 981 (alteration in original)

16

(internal quotations omitted); see also Rodriguez v. Hayes, 591 F.3d 1105, 1122 (9th Cir. 2010)

17

(“common” does not mean “complete congruence”). However, in a case alleging claims of a

18

defective product and misrepresentations regarding that defect, “[w]ithout any evidence of a

19

common defect, there are no ‘common questions of law or fact’ binding the proposed class

20

together.” Kramer v. Toyota Motor Co., 688 F. App’x 765, 766 (9th Cir. 2016).

21

Here, the Court’s ruling on the Daubert motion as to Dr. Allen are fatal to a finding of

22

commonality under the proposed class. Young’s common issues are premised on a purported

23
24
25
26
27
28

23

Cree does not challenge the numerosity requirement under Rule 23(a). Nor could Cree,
based on the record before the Court, as it is clear that there are a sufficient number of consumers
of Cree LED bulbs based on the numerosity requirement. See, e.g., Akaosugi v. Benihana Nat.
Corp., 282 F.R.D. 241, 253-54 (N.D. Cal. 2012) (“[C]ourts generally find the numerosity
requirement satisfied when a class includes at least forty members.”); Hernandez v. Cnty. of
Monterey, 305 F.R.D. 132, 153 (N.D. Cal. 2015) (“A class or subclass with more than 40
members raises a presumption of impracticability based on numbers alone.” (internal quotation
marks omitted)).
22

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 23 of 24

misrepresentation about the longevity of Cree LED bulbs. Young cites a number of omission

2

cases where consumers sued manufactures of a product with an undisclosed, fundamental, and

3

identified design defect. See, e.g., Wolin v. Jaguar Land Rover N. Am. LLC, 617 F.3d 1168, 1172

4

(9th Cir. 2010). Young relies exclusively on Dr. Allen’s opinions purporting to establish common

5

evidence of a defect relating to overheating in Cree LED lamps.24 Without this evidence, Young

6

does not articulate an alternative method of proving why or how his own bulbs failed, much less

7

one across multiple product lines and generations. Without such evidence, Young does not show

8

commonality in support of class certification. See Kramer, 688 F. App’x at 766; see also

9

Gonzalez, 885 F.3d at 196-199 (holding that where “[d]effective design [was] an essential element

10

of [p]laintiffs’ misrepresentation-based claims” and plaintiffs “do not identify a particular defect”

11
United States District Court
Northern District of California

1

and therefore “invite [the court] to equate the existence of a defect with the mere possibility that

12

one might exist” there “no support in Rule 23 or caselaw for certification on such a speculative

13

basis”); Grodzitsky v. American Honda Motor Co., Inc., No. 12-cv-001142-SVW-PLA, 2017 WL

14

8943159, at *6 (N.D. Cal. Oct. 30, 2017) (“All they have is a series of [products] that may or may

15

not have broken before they were supposed to, and these breakages may or may not have been

16

caused by a common defect which may or may not exist. This is manifestly inadequate to satisfy

17

Rule 23.”).

18

Instead, the issue of falsity or Cree’s alleged misrepresentations turn on issues that are not

19

common to the class as a whole: whether each consumer’s bulb failed, the manner of the failure,

20

the extent of longevity before failure, and what ultimately caused the failure. See Kramer, 668 F.

21

App’x at 766 (“Without any evidence of a common defect, there are no ‘common question of law

22

or fact’ binding the proposed class together.”); Gonzalez, 885 F.3d at 198 (“If proponents of the

23
24
25
26
27
28

24

As discussed under the Daubert analysis, it is not even clear that Dr. Allen’s report
establishes that there is, in fact, a defect within the identified Cree LED lamps at issue. Dr. Allen
testified that he essentially was provided an “inference” that these lamps were defective based on
the filing of this lawsuit by Young. (Allen Dep. at 83:19-84:24.) Dr. Allen posits reasons for this
defect, assuming it is exists, but fails to grapple with evidence in the record that shows that since
2013, Cree has not had any excessive premature failure, especially within less than one year of
operation. (Vollers I Decl. ¶ 56.) In other words, Dr. Allen’s report provides, at best, the likeliest
source of a defect, but does not establish that there is actually a defect and that the defect has any
link to the misrepresentations at issue.
23

Case 4:17-cv-06252-YGR Document 139 Filed 01/28/21 Page 24 of 24

1

class do not allege a defect common the class, the defectiveness of a given product is, by

2

necessity, not susceptible to proof by classwide evidence.”). This is manifestly an individual

3

determination, and therefore inappropriate to satisfy the commonality requirement under the Rule

4

23 analysis.25
Accordingly, the Court DENIES the motion for class certification.

5
6
7

IV.

CONCLUSION
Accordingly, the for the foregoing reasons, the Court HEREBY ORDERS as follows:

8



Cree’s Daubert motion as to Dr. Allen is GRANTED;

9



Young’s motion for class certification is DENIED; and

10



Cree’s Daubert motion as to Dr. Boedeker is DENIED AS MOOT.

United States District Court
Northern District of California

11

Moreover, in light of this Order, the Court hereby SETS a compliance deadline for 9:01

12

a.m. on Friday, February 26, 2021. Five (5) business days prior to the date of the compliance

13

deadline, the parties shall file a JOINT STATEMENT setting forth the parties’ position with respect

14

to the scheduling of this case. If compliance is complete, the compliance deadline will be taken

15

off calendar.

16

This Order terminates Docket Numbers 115, 116, 117, 118, and 127.

17

IT IS SO ORDERED.

18

Dated: January 28, 2021

19
20
YVONNE GONZALEZ ROGERS
UNITED STATES DISTRICT JUDGE

21
22
23
24
25
26
27
28

25

The Court declines to alternatively consider the remaining Rule 23(a) and 23(b)(3)
factors in light of the court’s conclusion as to commonality.
24