The Netherland Insuance Company v. HP, Inc.

UNITED STATES DISTRICT COURT

DISTRICT OF MASSACHUSETTS

THE NETHERLANDS INSURANCE

COMPANY as subrogee of GCP CROWN

COLONY, LLC and LIBERTY MUTUAL

FIRE INSURANCE COMPANY as

subrogee of ONEX YORK HOLDINGS

CORP. d/b/a CARE WORKS MANAGED

CARE SERVICES f/k/a MCMC LLC, No. 18-cv-12136-DLC

Plaintiffs,

v.

HP, INC. and INSIGHT DIRECT USA,

INC.,

Defendants.

ORDER ON DEFENDANTS’ MOTION FOR SUMMARY JUDGMENT

Cabell, U.S.M.J.

I. Introduction

This is a products liability case stemming from a printer

fire. The Netherlands Insurance Company (“Netherlands”) is the

subrogee of GCP Crown Colony 1031 LLC and MF Crown Colony 1031

LLC, the owners of the building in which the fire took place.

(Dkt. No. 8-1, pp. 1-2). Liberty Mutual Fire Insurance Company

(“Liberty Mutual”) is the subrogee of Onex York Holdings

Corporation (“Onex”), the tenant in whose office the fire took

place. (Dkt. No. 8-1, p. 2). HP Inc. (“HP”) is a company in the

business of designing, distributing, selling, and supplying

personal computer technology products. (Dkt. No. 8-1, p. 2).

Insight Direct USA, Inc. (“Insight”) is a company in the business

of selling hardware and software products. (Dkt. No. 8-1, p. 2).

Netherlands and Liberty Mutual (collectively, “the plaintiffs”)

raise three sets of claims against HP and Insight (collectively,

“the defendants”): (1) each defendant was negligent as a result of

design, manufacture, and/or distribution defects in the printer

that is alleged to have caught fire; (2) each defendant breached

warranties, both express and implied; and (3) each defendant

violated Massachusetts General Laws ch. 93A. (Dkt. No. 8-1). The

defendants move for summary judgment on all three claims; the

plaintiffs oppose the motion.1 (Dkt. Nos. 64, 81). For the reasons

set forth below, the motion for summary judgment is denied.

II. Factual Background

On April 25, 2017, at approximately 1:00 a.m., a fire broke

out at 300 Crown Colony Drive in Quincy, MA, in office space leased

by Onex. (Dkt. No. 78, ¶ 1). Shortly thereafter, Quincy

firefighter Keith Lentini arrived on the scene. (Dkt. No. 67-4,

p. 3). Lentini first checked the building’s fire alarm panel,

which indicated that a smoke detector had been activated on the

1 In a separate motion, the defendants move to exclude the plaintiffs’ expert

witnesses, which the plaintiffs also oppose. (Dkt. No. 61). Here, the court

concerns itself primarily with the defendants’ motion for summary judgment.

However, insofar as it is necessary to engage with the defendants’ motion to

exclude to fully consider their motion for summary judgment, the court will

do so.

fifth floor of the building, where Onex’s office was located, and

that the sprinkler system was running. (Dkt. No. 67-4, p. 4). He

then proceeded to the fifth floor to find Onex’s office filled

with smoke, and ultimately discovered a printer on fire at the

source of the smoke. (Dkt. No. 67-5, p. 4). Lentini unplugged

the printer and extinguished the fire with the help of another

firefighter. (Dkt. No. 67-5, p. 4). This printer was later

identified as an HP LaserJet 4250n model printer (hereinafter, the

“4250n printer”). (Dkt. No. 67-8, p. 2).

The previous evening, Onex IT technician Jin Lem had moved

the 4250n printer into a cubicle in Onex’s office to repurpose it

for a new department. (Dkt. No. 67-2, p. 5). Lem plugged the

4250n printer into a power source, ran a test print, and left the

printer in standby mode when he left for the night. (Dkt. No. 67-

2, pp. 5-6, 12). The fire took place in that cubicle approximately

five hours later. (Dkt. No. 78, ¶ 9). At the time of the fire,

there was a second printer also located in that cubicle, later

identified as an HP LaserJet p3015n printer (hereinafter, the

“p3015 printer”) (Dkt. No. 67-11, p. 7).2

HP designed and manufactured both printers and Insight sold

them to Onex’s predecessor. (Dkt. No. 78, ¶ 3). Onex’s

2 It remains disputed whether Lem relocated the p3015 printer to the cubicle

where the fire took place around the same time as the 4250n printer, whether

he similarly plugged it in and tested it, and whether he left it in standby

mode before leaving for the night. (Dkt. No. 78, ¶¶ 2, 6-7).

predecessor purchased the 4250n printer in 2007 and the p3015

printer in 2013. (Dkt. No. 78, ¶ 4). Neither printer had a known

history of issues or repairs in the years leading up to the fire.

(Dkt. No. 78, ¶ 5).

The parties dispute how the fire ignited, drawing their

theories from the testimony and reports of first responders, Onex

IT technician Jin Lem, and their own respective expert witnesses.

Both parties rely to varying degrees on the deposition testimony

of Lem and Quincy firefighter Keith Lentini. The plaintiffs

additionally rely on expert witnesses Edward Noonan and Matthew

Elliott to support their account of the fire, while the defendants

rely on expert witnesses Timothy Myers and Donald Galler.

Liberty Mutual retained Edward Noonan, a certified fire

investigator as well as a firefighter in Foxborough, MA, to

investigate and determine the cause and origin of the fire. (Dkt.

No. 79, pp. 34-35). Noonan conducted his on-site investigation on

April 28, 2017, three days after the fire took place. (Dkt. No.

67-8, p. 2). During his investigation, Noonan interviewed Lem,

reviewed Lentini’s deposition, observed the two printers, and took

photos of the printers and the cubicle where the fire took place.

(Dkt. No. 79, pp. 37-40, 43; Dkt. No. 67-8, pp. 15-19). Based on

all the evidence he observed, Noonan concluded the fire originated

within the 4250n printer.

Matthew Elliott, a certified fire and explosion investigator,

serves as the plaintiffs’ electrical engineering expert. (Dkt.

No. 67-13, p. 11). Based on multiple examinations of the damaged

printers, tests performed on exemplar HP 4250 series printers,

additional research, and a review of available testimony and other

court filings (including Noonan’s report), Elliott concluded that

a fault within the 4250n printer’s power supply, namely in and

around the printed circuit board (or “PCB”), resulted in an

“electrical arcing” event within the 4250n printer that then caused

the fire.3 (Dkt. No. 67-13, pp. 3-9). On his way to that

conclusion, Elliott determined that the p3015 printer and its power

cord sustained damage from an external “fire attack,” rejecting

the theory that any part of the p3015 printer, including its power

cord, was the cause of the fire. (Dkt. No. 67-13, pp. 8-9).

Finally, although Elliott concluded that some fault related to the

4250n printer’s PCB caused the fire, the degree of damage to the

circuit board rendered it impossible to identify the exact cause.

(Dkt. No. 67-13, p. 9).

The defendants offer a competing theory of the fire based on

the reports and testimony of their expert witnesses. In short,

3 “Electrical arcing” occurs when an electric current jumps a gap in a circuit,

an event that can produce extreme heat and cause electrical fires. Tanner

Walker, What is an Electric Arc?, FIRETRACE: FIRE PROTECTION BLOG (May 14, 2019),

https://www.firetrace.com/fire-protection-blog/arcing-the-leading-cause-of-

electrical-fires.

the defendants argue the fire was caused by an electrical arcing

event in the power cord of the p3015 printer, which then spread to

consume significant portions of the 4250n printer. The defendants

theorize that the power cord of the p3015 printer had previously

sustained mechanical damage, and that this mechanical damage

created conditions that could support the type of electrical arcing

capable of starting a fire. (Dkt. No. 67-11, p. 20). The

defendants further suggest that, considering the years both

printers functioned without issue, the relocation of one or both

printers the evening before the fire may have dealt the decisive

mechanical damage to the p3015 printer’s power cord that ultimately

caused the fire. (Dkt. No. 67-12, p. 16).

III. Discussion

A. Summary Judgment Standard

The role of summary judgment is “to pierce the pleadings and

to assess the proof in order to see whether there is a genuine

need for trial.” Carrozza v. CVS Pharmacy, Inc., 992 F.3d 44, 56

(1st Cir. 2021) (internal quotation omitted). Summary judgment is

appropriate if there is no genuine dispute of material fact and

the moving party is entitled to judgment as a matter of law. Lima

v. City of E. Providence, 17 F.4th 202, 206 (1st Cir. 2021). The

moving party has the initial burden to prove the absence of such

a genuine dispute. Fed. R. Civ. P. 56(a); Celotex Corp. v.

Catrett, 477 U.S. 317, 322 (1986).

Once the moving party has satisfied its burden, the burden

shifts to the non-moving party to set forth specific facts showing

that there is a genuine, triable dispute. Celotex, 477 U.S. at

324. The Court must view the entire record in the light most

favorable to the non-moving party and indulge all reasonable

inferences in that party's favor. Friends of Merrymeeting Bay v.

Hydro Kennebec, LLC, 759 F.3d 30, 34 (1st Cir. 2014).

B. Daubert in the Context of Summary Judgment

The defendants combine much of their summary judgment

argument with the separate but conceptually related argument that

the court should exclude the plaintiffs’ expert witnesses as

unreliable. Under the familiar Daubert standard, the court weighs

reliability and relevance in assessing “whether the reasoning or

methodology underlying [expert scientific] testimony is

scientifically valid and . . . whether that reasoning or

methodology properly can be applied to the facts in issue.”

Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 592–93 (1993);

see Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147 (1999)

(clarifying that Daubert applies to all expert testimony). That

assessment may consider factors including whether the theory or

technique has been tested; whether it has been subjected to peer

review and publication; in the case of a technique, its known or

potential rate of error as well as the existence and maintenance

of any standards controlling its operation; and the extent of its

acceptance within the relevant scientific community. Daubert, 509

U.S. at 592–94. In a motion to exclude pursuant to Daubert, the

burden on the party who proffers expert testimony is not to prove

that an expert's conclusion is correct but rather that the expert

reached their conclusion in a scientifically sound and

methodologically reliable way. Ruiz-Troche v. Pepsi Cola of Puerto

Rico Bottling Co., 161 F.3d 77, 85 (1st Cir. 1998). Moreover, as

Daubert is careful to underscore, “[v]igorous cross-examination,

presentation of contrary evidence, and careful instruction on the

burden of proof are the traditional and appropriate means of

attacking shaky but admissible evidence.” 509 U.S. at 596.

In the context of fire investigations, courts have often

weighed the relevant Daubert factors in close reference to the

National Fire Protection Association’s Guide for Fire and

Explosion Investigations (hereafter, “NFPA 921”), an industry

standard for conducting fire investigations. See, e.g., United

States v. Hebshie, 754 F. Supp. 2d 89, 111 (D. Mass. 2010) (“NFPA

921 . . . is widely accepted as the standard guide in the field of

fire investigation.”); Workman v. AB Electrolux Corp., No. 03-

4195-JAR, 2005 WL 1896246, at *10 (D. Kan. Aug. 8, 2005) (“[T]he

methodology set forth in NFPA 921 . . . represents the national

standard with regard to appropriate methodology for investigation

by fire science experts.”); Travelers Prop. & Cas. Corp. v. Gen.

Elec. Co., 150 F. Supp. 2d 360, 366 (D. Conn. 2001) (“[T]he court

is convinced that the steps [the expert witness] took to develop

his opinion were nevertheless consistent with the principles of

NFPA 921, a peer reviewed and generally accepted standard in the

fire investigation community.”) (emphasis in original). At the

same time, courts have also cautioned against viewing violations

or ignorance of NFPA 921 as necessarily rendering a fire

investigation unreliable. See, e.g., Schlesinger v. United

States, 898 F. Supp. 2d 489, 504 (E.D.N.Y. 2012) (“[A] failure to

follow the methodology in NFPA 921 does not automatically require

the exclusion of expert testimony on the cause and origin of a

fire.”); Pekarek v. Sunbeam Prod., Inc., 672 F. Supp. 2d 1161,

1175 (D. Kan. 2008) (“The mere fact [the expert] did not cite or

use NFPA 921 as his guide does not necessarily mean he failed to

use a reliable method.”); Am. Fam. Mut. Ins. Co. v. Hewlett-Packard

Co., No. CIV. 07-792 ADM/AJB, 2008 WL 2130217, at *4 (D. Minn. May

19, 2008) (“NFPA 921 is not the exclusive standard for

investigating fire and explosion.”). Thus, while the standards

contained in NFPA 921 may help elucidate whether an expert witness’

fire investigation methodology passes muster under Daubert, they

do not control the court’s determination.

The Daubert analysis contains an extra wrinkle when

confronted at the summary judgment stage. The court can apply

Daubert at summary judgment, but “[b]ecause the summary judgment

process does not conform well to the discipline that Daubert

imposes, the Daubert regime should be employed only with great

care and circumspection at [that] stage.” Cortés-Irizarry v.

Corporacion Insular De Seguros, 111 F.3d 184, 188 (1st Cir. 1997).

The problem is that courts typically have only a “truncated record”

at summary judgment, whereas Daubert requires a “complex factual

inquiry.” Id. Thus, “where Daubert intersects with summary

judgment practice, Daubert is accessible, but courts must be

cautious . . . not to exclude debatable scientific evidence without

affording the proponent of the evidence adequate opportunity to

defend its admissibility.” Id.

Here, the defendants’ motion for summary judgment is so

closely intertwined with their motion to exclude the plaintiffs’

expert witnesses (hereafter, the “Daubert motion”) that the court

cannot rule on the former without essentially ruling on the latter.

While trial typically “provide[s] the best operating environment”

for a Daubert assessment, here the court will assess the evidence

on record pursuant to Daubert to resolve the motion for summary

judgment. Id. In doing so, the court acknowledges the First

Circuit’s caution against excluding evidence without affording an

adequate opportunity to defend its admissibility. Because neither

summary judgment nor exclusion is appropriate here, however, there

is no material risk of defying that caution.

C. Daubert analysis

For the plaintiffs to prevail on their negligence and breach

of warranty claims, they must present evidence showing that there

was a defect in the printer at the time it was sold and that the

defect caused the cited property damage. Mass. Prop. Ins.

Underwriting Ass'n v. LG Elecs. U.S.A., Inc., 902 F. Supp. 2d 173,

176 (D. Mass. 2012) (citing Santos v. Sunrise Med., Inc., 351 F.3d

587, 591-92 (1st Cir. 2003)). More specifically, the plaintiffs

must present expert testimony identifying the defect and linking

that defect to the alleged harm because the defect is such that

its presence “cannot be inferred in the absence of expert

testimony.” Id. (quoting Enrich v. Windmere Corp.,

416 Mass. 83, 87 (1993)). Recalling the high bar for exclusion at

the summary judgment stage, the court will consider the methodology

of each challenged expert witness in turn.

1. Daubert Assessment: Noonan

The defendants argue Noonan’s methodology was unreliable

because: (1) he determined the 4250n printer was the origin of the

fire due to the extensive damage it had sustained; (2) he did not

conduct a meaningful investigation; (3) he relied on Lentini’s

deposition testimony in reaching his own conclusion as to the cause

and origin of the fire; (4) his analysis of the burn patterns at

the site of the fire omits key details; and (5) he determined that

the 4250n printer was the cause of the fire before he arrived at

the site to perform his inspection. (Dkt. No. 65, pp. 11-12). As

discussed below, these objections go to the weight of Noonan's

testimony, not its admissibility.

a. Objection 1: Noonan determined the 4250n

printer was the origin of the fire due to the

extensive damage it had sustained.

First, the defendants argue that Noonan’s methodology was

unreliable because he based his conclusions on the significant

damage that the 4250n printer sustained in the fire while also

acknowledging that the extent of fire damage is not always

indicative of a fire’s cause and origin. (Dkt. No. 65, p. 11).

In his deposition, Noonan did indeed articulate that his conclusion

that the fire started in the 4250n printer was based in part on

the significant damage to the printer. (Dkt. No. 67-9, pp. 4-5).

He also agreed that, per NFPA 921, the degree of damage to an

appliance may not be an adequate indication of a fire’s origin.

(Dkt. No. 67-9, p. 6). This, however, was not all he said.

In the full cited quote from Noonan’s deposition, he indicates

that in reaching his conclusion he was not relying solely on the

extent of damage to the printer but on Lentini’s deposition

testimony as well. (Dkt. No. 67-9, pp. 4-5). Furthermore,

Noonan’s full origin-and-cause report cites to additional evidence

to support his conclusion that the fire originated in the 4250n

printer (e.g., other physical evidence found at or around the

cubicle after the fire; Lem’s deposition testimony). See (Dkt.

No. 67-8, pp. 8-12).

Apart from being inaccurate, the defendants’ characterization

of Noonan’s deposition here, in the court’s view, would at most go

to the weight of the evidence. That is, even if Noonan relied on

this evidence as heavily as the defendants suggest, this apparent

weakness would not be grounds for excluding Noonan as an expert

witness altogether. Noonan’s reliance or overreliance on the

extent of the damage to the printer may well be useful fodder for

the defense’s cross-examination, but it is not an appropriate basis

upon which to determine Noonan’s entire methodology was

unreliable.

b. Objection 2: Noonan did not conduct a

meaningful investigation.

Second, the defendants argue that Noonan did not conduct a

“meaningful investigation,” offering by way of example that

Noonan “did not educate himself about the 4250n, and he could

not describe the 4250n’s components, materials, or features.”

(Dkt. No. 65, p. 11). The defendants cite to Noonan’s

deposition testimony, in which he responds in the negative to

several questions probing his familiarity with 4250n printers.

(Dkt. No. 67-9, p. 4). Why this ought to establish whether

Noonan’s investigation was sufficiently “meaningful” for present

purposes, the defendants do not say. In the court’s view,

though, this particular criticism does not provide a basis to

render Noonan’s opinion unreliable.

Noonan does not need to understand the minute functionality

of 4250n printers for his testimony to be admissible because the

plaintiffs do not rely on Noonan for a complete and granular

theory of how the fire started and spread. Rather, Noonan’s

role in the fire investigation was largely aimed at documenting

the scene of the fire and preserving evidence. (Dkt. No. 79-2,

pp. 24, 26). The broad conclusions he draws at the end of his

investigation, which he supports by outlining his adherence to

the scientific method, do not extend beyond that evidence and

documentation. (Dkt. No. 67-8, pp. 6-12). The plaintiffs’

second expert witness, Elliott, then focuses his investigation

on how the fire ignited and spread in more minute detail,

relying on his electrical engineering background in conjunction

with the evidence Noonan had documented and preserved to form

his conclusions. (Dkt. No. 67-13). As such, Noonan’s role is

to establish that the fire started within the printer, not how

exactly the fire started. That determination does not require a

sophisticated understanding of the printer’s inner workings, and

a lack of such understanding does not render Noonan’s

methodology unreliable.

c. Objection 3: Noonan relied on firefighter

Lentini’s deposition testimony in reaching his

own conclusion as to the cause and origin of

the fire.

Third, the defendants argue that Noonan’s investigation

relied on Lentini’s investigation, which was a flawed

investigation, which means that Noonan’s investigation was flawed

as well. In particular, the defendants argue that “Lentini simply

decided that the fire originated in the 4250n because he found it

on fire when he arrived at the scene.” (Dkt. No. 65, p. 11).

Setting aside whether this is an accurate characterization of

Lentini’s investigation, Noonan did not rely exclusively on

Lentini’s account in reaching his own conclusion. (Dkt. No. 67-

8, pp. 2-6, 10-12). As outlined above, Noonan relied on a range

of evidence.

Nevertheless, the plaintiffs counter that Noonan was right to

rely on Lentini’s investigation to the extent that he did, arguing

NFPA 921 directly supports this practice. (Dkt. No. 81, p. 7).

Section 18.1.2 of the NFPA 921 states in part:

Determination of the origin of the fire involves the

coordination of information derived from one or more of

the following:

(1) Witness Information and/or Electronic Data. The

analysis of observations reported by persons who

witnessed the fire or who were aware of the condition

present at the time of the fire as well as the analysis

of electronic data such as security camera footage,

alarm system activation, or other such data recorded in

and around the time of the fire event.

(Dkt. No. 81, pp. 5-6). The defendants, meanwhile, have not

established that Noonan’s partial reliance on Lentini’s account

was methodologically unsound. The above language from NFPA 921

suggests the opposite.

d. Objection 4: Noonan’s analysis of the burn

patterns at the site of the fire omits key

details.

The defendants argue that Noonan’s opinion is unreliable

because “his conclusion ignores the fact that the 4250n was not

the only item on the cubicle desk and that the arced power cord

was actually closer to the ‘V’ pattern than the 4250n” printer.

(Dkt. No. 65, p. 12). This criticism again misses the mark for

exclusion under Daubert even assuming it is true. The pertinent

inquiry is not whether an expert witness’ conclusion is “correct”

but only whether it was “arrived at in a . . . methodologically

reliable fashion.” Ruiz-Troche 161 F.3d at 85. Here, the

defendants argue that their theory of the fire’s cause and origin

is correct, and that Noonan’s is not. Insofar as the defendants

allege that Noonan’s interpretations of the burn patterns are

incorrect, they will be free to cross-examine him along those lines

at trial. Under the Daubert assessment, however, the defendants

do not establish that Noonan’s methodology was unreliable in this

respect.

e. Objection 5: Noonan determined that the 4250n

printer was the cause of the fire before he

arrived at the site to perform his inspection.

Last, the defendants argue that Noonan arrived at the site of

the fire to perform his investigation with his conclusion already

formed, in violation of NFPA 921 standards. (Dkt. No. 65, p. 12).

In support, the defendants quote and emphasize Noonan’s deposition

testimony, where he says, “I may have made a preliminary thought

determination for myself without any further electrical

engineering that would be necessary to confirm that.” (Dkt. No.

65, p. 12). The defendants then cite to two sections of NFPA 921,

NFPA 921 4.3.8 and NFPA 921 4.3.9, that warn fire investigators

against letting premature determinations or presumptions with

respect to fire origin dictate their investigation processes.4

(Dkt. No. 65, p. 12).

A more comprehensive quote from Noonan’s deposition at the

very least casts doubt on the defendants’ interpretation:

Q. Did you consider the possibility that the fire started

outside the 4250n printer?

A. At the fire scene I did, but my thought process

brought me back to the 4250 printer itself.

Q. And how did you make that determination?

4 In their memorandum in support of their motion for summary judgment, the

defendants cite to the 2017 version of NFPA 921 4.3.7. (Dkt. No. 65, p. 12).

However, the parenthetical quote they include (“All investigations of fire

and explosion incidents should be approached by the investigator without

presumption as to origin...”) comes from NFPA 921 4.3.8, which is titled

“Avoid Presumption.” (Dkt. No. 65, p. 12). The defendants do not appear to

be relying on NFPA 921 4.3.7 in their argument here.

A. Again, just by the damage that I seen [sic] within

the unit.

Q. Can you be more specific?

A. Again, the thermal damage within the unit, even the

surrounding area. I guess you talk about the tabletop

and cubicle. The burn pattern on that to me it's

directional and it comes from the 4250 printer. So I

think in totality I took that into account, and, again,

on that day, you know, I was really there to preserve

the evidence, so I may have made a preliminary thought

determination for myself without any further electrical

engineering that would be necessary to confirm that.

(Dkt. No. 67-9, p. 7). As this more fulsome excerpt suggests,

Noonan considered a range of evidence and factors in reaching his

conclusion. In context, when he said he “made a preliminary

thought determination,” it is more likely he meant that the

evidence he observed at the scene of the fire suggested a

particular cause that would still need to be confirmed with

“further electrical engineering” expertise. Regardless, Noonan’s

deposition remarks here do not clearly demonstrate that he made a

premature determination of the cause and origin of the fire before

conducting his full investigation.

The defendants’ cited NFPA 921 standards do not alter this

conclusion. Based on the portion of Noonan’s deposition that the

defendants selectively quote, Noonan did not appear to be making

any premature determinations or specific hypotheses about how the

fire occurred, let alone final conclusions unduly influenced by

presumptions or expectations. (Dkt. No. 67-9, p. 7). This coheres

with Noonan’s emphasis on the preservation and documentation of

evidence in his investigation, which is likewise supported by NFPA

921. See NFPA 921, § 4.1 (2017 ed.) (“With few exceptions, the

proper methodology for a fire or explosion investigation is to

first determine and establish the origin(s), then investigate the

cause: circumstances, conditions, or agencies that brought the

ignition source, fuel, and oxidant together.”).

For the foregoing reasons, the court finds that the defendants

have not established that Noonan’s methodology was unreliable

under Daubert.

2. Daubert Assessment: Elliott

The defendants argue Elliott’s methodology was unreliable

because: (1) his use of a butane torch in his testing violated

NFPA 921 and proved nothing; (2) he did not identify a specific

component failure in the 4250n printer or articulate with enough

precision how a failure in the printer could generate enough heat

to start a fire, which violated NFPA 921; (3) he did not

appropriately account for the 4520n’s redundant safety features or

how they could have failed in such a way that allowed for a fire

to start, which violated NFPA 921; and (4) his rejection of the

defendants’ theory of the fire cause and origin is based on

insufficient evidence. As with the defendants’ objections to

Noonan’s investigation, these objections go more to the potential

weight of Elliott’s testimony rather than its admissibility, and

they do not establish that his methodology was unreliable under

Daubert.

a. Objection 1: Elliott’s use of a butane torch

in his testing violated the NFPA 921 and

proved nothing.

First, the defendants argue that “Elliott’s butane torch

experiment did not comply with NFPA 921, which requires a fire

investigator to demonstrate ‘how the appliance generated

sufficient heat energy to cause ignition.’” (Dkt. No. 65, p. 12).

Here, the defendants refer to the portion of Elliott’s

investigation in which he gathered similar models of the HP 4250

series printer (or “exemplar” printers) to test theories of how

the fire ignited and spread in the manner that it did. (Dkt. No.

67-13, p. 5). On the two documented occasions where Elliott

performed such testing, he used a butane torch to ignite certain

components of the exemplar printers. (Dkt. No. 67-13, p. 5). If

Elliott’s use of the butane torch was intended to serve as the

“sufficient heat” in his testing, as the defendants suggest, then

perhaps this objection would cast doubt on his methodology and,

likewise, the plaintiffs’ theory of how the printer generated

enough heat to ignite. This, however, is not the case.

Elliott’s theory posits that a “board level component failure

or heating connection” caused an electrical arcing event, and that

arcing event then caused the fire. (Dkt. No. 67-13, p. 9). Elliott

points to the presence of melted copper in the original 4250n

printer’s PCB as an indication that electrical arcing created

temperatures well above the ignition temperatures for the polymer

components in the printer. (Dkt. No. 67-13, p. 9). Thus,

according to Elliott’s theory, it was the electrical arcing that

“generated sufficient heat energy to cause ignition.” See NFPA

921, § 26.4.1.1 (2017 ed.). This is the part of the causal chain

that Elliott needed to demonstrate through his testing.

In testing his second exemplar printer, Elliott used the

butane torch to instigate an arcing event that in turn produced

sufficient heat to cause ignition in the printer. (Dkt. No. 67-

13, pp. 5, 9). In effect, the butane torch replaced the board-

level fault that allegedly caused the arcing event in the destroyed

4250n printer. The resulting arcing event, not the torch, ignited

the exemplar printer, which demonstrated that such an arcing event

could indeed produce “sufficient heat energy to cause ignition.”

If there is an issue with Elliott’s methodology, it is that he

used an external heat source to induce the arcing event inside the

printer. As the defendants are quick to point out, “[t]he 4250n

does not contain a butane torch inside it.” (Dkt. No. 65, p. 12).

However, the defendants do not point to any provision in NFPA 921

that prohibits the use of an external heat source to induce an

arcing event. Further, given that the original 4250n printer’s

PCB was destroyed by the fire and impossible to analyze, see (Dkt.

No. 79-4, p. 9), it was not unreasonable for Elliott to use the

butane torch as an alternative to trying to recreate the unknown

“board-level fault.” As such, the use of the butane torch does

not warrant excluding Elliott’s testimony.

b. Objection 2: Elliott did not identify a

specific component failure in the 4250n

printer or articulate with enough precision

how a failure in the printer could generate

enough heat to start a fire, which violated

NFPA 921.

Second, the defendants argue Elliott’s lack of specificity in

his conclusion of component failure is a violation of NFPA 921,

citing specifically to NFPA 921 26.4.6. (Dkt. No. 65, pp. 12-13).

The defendants quote part of this section of NFPA 921, claiming

that it “requires an investigator to ‘establish the validity of

the proposed ignition scenario. If the ignition scenario requires

the failure or malfunction of one or more appliance components,

this can also be tested for validity on [an] exemplar.’” (Dkt.

No. 65, p. 13).

As discussed above, Elliott’s use of the butane torch in

testing appears to have established the validity of his proposed

ignition scenario. That is, Elliott’s theory is that a “board

level component failure or heating connection” caused the fire

within the 4250n printer, and his testing revealed that such a

fault resulted in “flaming combustion when the power supply is

energized.” (Dkt. No. 67-13, p. 9). In this respect, Elliott

seems to have met the NFPA 921 standard cited by the defendants.

As to the specificity of his theory, Elliott homes in on the

PCB as the approximate origin of the fire. In his report, he

posits:

While the specific board level component failure or

heating connection could not be identified due to the

extent of damage, both will compromise the insulation

integrity of the PCB and result in arc tracking. Arc

tracking and arcing through a carbonized path are

generally-accepted principles and will result when the

PCB insulation integrity is compromised due to a board

level component failure or heating connection. The

melting of the copper board level traces reveals

temperatures observed during tracking were in excess of

1981°F, above the ignition temperature of listed

polymers 910°F – 1220°F.

(Dkt. No. 67-13, p. 9) (citations omitted). In his deposition,

Elliott adds that he was unable to offer a more specific

identification than this because, “given the extent of damage,

there were no materials left to analyze.” (Dkt. No. 79-4, p. 9).

He does, however, further support his theory through testing and

by attempting to eliminate several competing theories, including

that put forward by the defendants.

Furthermore, the plaintiffs offer a different NFPA 921

standard that seems to contradict the defendants’ argument with

respect to NFPA 921 26.4.6. The plaintiffs refer to AmGUARD Ins.

Co. v. Richmond, which quotes NFPA 921 19.6.5 to explain that “NFPA

921 does approve a process of eliminating alternative causes of a

fire so long as a ‘[d]etermination of the ignition source [is]

based on data or logical inferences drawn from that data’ or

‘derived from evidence, observations, calculations, experiments,

and the laws of science.’” AmGUARD Ins. Co. v. Richmond, No. CV

19-11529-RGS, 2021 WL 1857180, at *3 (D. Mass. May 10, 2021).

Here, Elliott determined the ignition source and subsequent spread

of the fire by drawing inferences from physical and photo evidence,

temperature calculations, and testing, all of which he framed in

terms of the scientific method. (Dkt. No. 67-13). For example,

Elliott inferred the high temperatures caused by the theorized

arcing event by observing the melted copper traces on the circuit

board, knowing that the melting point of copper is significantly

higher than the ignition temperature for the listed component

polymers. Consequently, the defendants’ argument that the

plaintiffs violated NFPA 921 by failing to establish an ignition

source through testing fails to persuade.

c. Objection 3: Elliott did not appropriately

account for the 4520n printer’s redundant

safety features or how they could have failed,

which violated NFPA 921.

Third, the defendants argue that “Elliott failed to account

for the 4520n’s multiple redundant safety features or to explain

how they all could have failed at the same time,” which they claim

violates section 19.7.3 of NFPA 921. (Dkt. No. 65, p. 13). The

defendants quote this section of NFPA 921 as stating, “Safety

devices and features are often engineered and built to prevent

fires from occurring. The cause determination will need to account

for the actions of safety devices.” (Dkt. No. 65, p. 13).

Elliott’s theory accounts for the fuses in the 4250n printer

failing to prevent the fire. His theory, which he supports with

evidence, is that an arcing event was able to bypass the fuses,

which is why the board-level fuses in the 4250n printer were not

blown during the fire. (Dkt. No. 67-13, p. 9). Just as the board-

level fuses remained intact in the original 4250n printer, Elliott

noted that the PCB fuses remained intact after he induced an arcing

event through testing with an exemplar printer. (Dkt. No. 67-13,

p. 5). Thus, in making his cause determination, Elliott accounted

for the actions of safety features, and the defendants have not

demonstrated that his methodology was unreliable in this respect.

d. Objection 4: Elliott’s rejection of the

defendants’ theory of the fire cause and

origin is based on insufficient evidence.

Finally, the defendants argue there is no evidence to support

Elliott’s dismissal of the defendants’ theory that the fire was

caused by an arcing event that occurred in the power cord of the

p3015 printer. “The basis for Elliott’s [dismissal] was that a

circuit breaker had tripped, which would have stopped any prolonged

arcing sufficient to start a fire.” (Dkt. No. 65, p. 13). A

tripped circuit breaker does indeed seem to factor into Elliott’s

dismissal of the power cord as a cause of the fire, but the

defendants’ objections to this theory go to the weight of the

evidence and are not grounds for excluding Elliott’s testimony.

In the course of his analysis of potential fire causes,

Elliott eliminates several alternatives to his theory that the

fire originated in the 4250n printer. (Dkt. No. 67-13, pp. 8-9).

Citing a lack of sufficient evidence, he rules out separate

theories that the fire was caused by an over-voltage condition,

faulty wiring of the cubicle or other surrounding devices, and

anything internal to the p3015 printer. (Dkt. No. 67-13, pp. 8-

9). Elliott also rules out a fault in either printer’s power cord

as causes of the fire, theorizing that the damage to the cords was

instead the result of an external fire attack. (Dkt. No. 67-13,

p. 8). He notes, among other things, that the absence of “evidence

of a sustained arcing event” on the p3015 printer’s power cord “is

consistent with the operation of the circuit breaker protecting

the circuit supplying the printers.” (Dkt. No. 67-13, p. 8).

The defendants take issue with this inference. Pointing to

various depositions and expert reports (including Elliott’s own

report), which all indicated that there was no sign that circuit

breakers tripped, they claim “Elliott has no evidence that circuit

breakers had tripped.” (Dkt. No. 65, p. 13) (emphasis in the

original). This is untrue. Elliott inferred that the circuit

breaker must have tripped from the absence of evidence of a

sustained arcing event on the p3015 printer cord. (Dkt. No. 67-

13, p. 8). According to Elliott, if the breaker had not tripped,

the arcing event would have continued and caused more damage to

the cord than it did. (Dkt. No. 67-14, 48:20-49:16). The

defendants may feel confident that the lack of hard evidence of a

circuit breaker tripping discredits Elliott’s theory. However, it

is for the trier of fact to determine whether the theory is

correct. At this stage, the court is only concerned with whether

Elliott’s inference is methodologically unsound. It is not.

As noted above, the bar for excluding expert witnesses is a

high one. The court finds that the defendants have not cleared

that bar here, and the court rejects their arguments that either

Noonan or Elliott should be excluded. As a result, their argument

for summary judgment in their favor on the basis that the

plaintiffs lack the necessary expert witness testimony fails.

D. The defendants are not entitled to summary judgment on

the grounds that the plaintiffs’ claims fail even with

their expert testimony

The defendants also argue that the plaintiffs’ claims fail

for lack of evidence even if Noonan’s and Elliott’s expert

testimony is considered. (Dkt. No. 65, pp. 15-19). In this

regard, the defendants principally take issue with Elliott’s

conclusions, arguing that “they are not supported by any facts in

the record.” (Dkt. No. 65, p. 19). Without repeating the

foregoing, though, this argument fails for the same reason that

many of their arguments against Elliott fail to warrant exclusion

under a Daubert assessment: their arguments go to the weight rather

than admissibility of the evidence, which is not what the court is

tasked with considering at this stage. At the summary judgment

stage, the court’s concern is whether a genuine issue of material

fact exists between the parties. As it relates to the expert

testimony of Noonan and Elliott, the court finds that such a

genuine issue of material fact does indeed exist.

The defendants argue that the plaintiffs have not identified

a specific defect or component failure in the 4250n printer; that

Elliott’s testing with the butane torch did not establish a cause

of the fire; and that the defendant’s own theory of the fire has

not been disproven by the plaintiffs’ expert testimony. (Dkt. No.

65, pp. 16-18). In rejecting these arguments, the court notes

first that Elliott puts forth a theory of the fire that identifies

the PCB in the 4250n printer as the locus of the defect or component

failure while eliminating competing theories. To be sure, the

defendants take issue with that theory but there are at least some

facts to support it, as discussed above. Second, as previously

addressed, the defendants misinterpret Elliott’s use of a butane

torch in his testing. The torch’s purpose was not to ignite the

4250n printer as a simulation of the fire in and of itself but to

induce an arcing event to see if that arcing event was

independently capable of igniting a fire within the printer. As

Elliott’s testing revealed, it was. Finally, the defendants’

argument that Elliott’s theory of the fire fails to disprove their

own theory does not indicate the absence of a genuine issue of

material fact. The plaintiffs do not need to disprove the

defendants’ competing theory to survive summary judgment. Rather,

the appropriate place for the parties to prove or disprove their

theories is before a trier of fact.

IV. Conclusion

For the foregoing reasons, the defendants’ motion for summary

judgment is DENIED.

So Ordered. /s/ Donald L. Cabell

DONALD L. CABELL, U.S.M.J.

DATED: December 5, 2022