UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________________________________________________
`
`TECHNICAL CONSUMER PRODUCTS, INC.,
`NICOR INC., AND AMAX LIGHTING,
`Petitioner,
`
`v.
`
`LIGHTING SCIENCE GROUP CORP.
`Patent Owner
`_______________________________________________________
`
`Case No.: IPR2017-01280
`Patent No.: 8,967,844
`_______________________________________________________
`
`Case No.: IPR2017-01285
`Patent No.: 8,672,518
`_______________________________________________________
`
`Case No.: IPR2017-01287
`Patent No.: 8,201,968
`_______________________________________________________
`
`DECLARATION OF ERIC BRETSCHNEIDER, PH.D.
`_______________________________________________________
`
`PATENT OWNER EXHIBIT 2001
`Page 1
`
`

`

`Exhibit
`A
`B
`
`C
`
`D
`E
`
`Description
`CV - Dr. Eric Bretschneider
`IES Lighting Handbook Application Volume 1981, John E. Kaufman,
`Howard Haynes, Eds.
`Robert E. Simmons, “Simplified formula for Estimating Natural Convection
`Heat Transfer Coefficient on a Flat Plate,” Electronics Cooling vol. 7, No.
`3, p. 12-13, August 2001
`HLMT-PL00 Specification
`IES Lighting Handbook Reference Volume 1984, John E. Kaufman, Jack F.
`Christensen, Eds.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 2
`
`

`

`
`
`I.
`
`INTRODUCTION
`
`I, Eric Bretschneider, declare as follows:
`
`1.
`
`I have been retained as an expert by Lighting Science Group Corp. (“LSG”) in
`
`connection with the above-captioned lawsuit to provide my analyses and conclusions on certain
`
`technical aspects of this dispute.
`
`2.
`
`I have personal knowledge of the matters set forth in this declaration and, if called
`
`upon to do so, would testify to such matters in court. My analyses and conclusions are based on
`
`my review of U.S. Patent Nos. 8,201,968 (the “‘968 Patent”), 8,672,518 (the “‘518 Patent”), and
`
`8,967,844 (the “‘844 Patent”) (collectively, the “Patents at Issue”), their prosecution histories, the
`
`materials cited below, the Petitions filed by Petitioners in these IPR proceedings, any Exhibits to
`
`those Petitions, my professional experience, and my expertise in the field of light-emitting diode
`
`technology.
`
`3.
`
`If asked to do so, I may testify regarding the contents of this declaration, and I
`
`reserve the right to use and rely on certain demonstratives to do so. I also reserve the right to
`
`amend and/or supplement this declaration should additional information or developments that may
`
`affect my opinions become available.
`
`4.
`
`I am being compensated at my customary rate of $400 per hour for my work in
`
`connection with this case. My compensation is not dependent on the contents of this declaration,
`
`the substance of any further analyses, conclusions or testimony that I may give, or the outcome of
`
`this case.
`
`II.
`
`PROFESSIONAL BACKGROUND
`
`5.
`
`My qualifications for forming the conclusions set forth in this declaration are
`
`summarized here and explained in more detail in my curriculum vitae, which is attached as Exhibit
`
`A.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 3
`
`

`

`
`
`6.
`
`I have over 25 years of experience with lighting and LEDs, including a
`
`comprehensive background on a full range of LED production technologies, including Metal-
`
`Organic Chemical Vapor Deposition (“MOCVD”) hardware/process, fabrication, LED chip and
`
`package testing and reliability, optical design, thermal management, color conversion, and SSL
`
`fixture/lamp design, integration, and reliability. Throughout the course of my career I have
`
`designed and transferred into manufacturing over 150 different LED-based lighting products.
`
`7.
`
`I am currently the Chief Technology Officer at EB Designs & Technology. In that
`
`capacity, I am (among other things) responsible for the design of solid-state lighting technologies
`
`for clients ranging from startups to Fortune 100 companies.
`
`8.
`
`I am also a member and current chair of the University of Florida Department of
`
`Chemical Engineering Advisory Board. I have been a Conference Chair for LED Measurement
`
`and Standards. I am also a member of a number of professional societies, including the
`
`International Society for Optics and Photonics (SPIE), Materials Research Society, and
`
`Illuminating Engineering Society (I am a member of the Science Advisory Panel as well as a
`
`member of numerous committees, most notably the IES Test Procedures Committee where I chair
`
`the Solid-State Lighting subcommittee).
`
`9.
`
`Prior to my position at EB Designs & Technology, I served as the Director of
`
`Engineering at HeathCo, LLC. In that capacity, I was responsible for advanced technology/product
`
`development related to solid-state lighting, sensors, notifications, and control products.
`
`10.
`
`Prior to my position as Director of Engineering at HeathCo, I was at the Elec-Tech
`
`International Co., Ltd., where I held the positions of Chief Engineer, ETi Lighting Research
`
`Institute and VP of Research and Development, ETi Solid State Lighting. In that capacity, my
`
`responsibilities included developing all technology and product roadmaps for markets in North
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 4
`
`

`

`
`
`America, China, Europe, and Japan. I designed and developed LED based lighting products for
`
`all of these markets.
`
`11.
`
`Between 2008 and 2011, I was at LSG, first as a product development manager,
`
`and my responsibilities included developing solid state lighting products, then as VP of Research,
`
`and my responsibilities included developing advanced LED models for product development and
`
`production control. In these roles I was involved in the design and manufacture of numerous LED-
`
`based lighting fixtures and products.
`
`12.
`
`Between 2004 and 2008, I was at Toyoda Gosei North America, where was a sales
`
`manager, and my responsibilities included managing and developing LED die and package sales
`
`accounts form the eastern region of North America. I was also tasked with providing technical
`
`support for the entire western hemisphere. The support I provided included design of LED
`
`packages and design of lighting fixtures and products that incorporated LED packages.
`
`13.
`
`Between 2003 and 2004, I was at Beeman Lighting, where I was Director of Solid
`
`State Lighting Engineering, and my responsibilities included leading development of solid state
`
`lighting systems and materials.
`
`14.
`
`Between 1998 and 2003, I was at Uniroyal Optoelectronics where I held a number
`
`of positions including Team Leader for the Epitaxial Growth and Materials Characterization areas,
`
`Sr. Epi Scientist, Director of Intellectual Property, University Relations and Government
`
`Contracts. My responsibilities included MOCVD hardware modification, epitaxial process
`
`development as well as design, development and testing of new LED chip structures for both
`
`AlInGaP and GaN-based material systems. I was also responsible for providing technical support
`
`and assistance to customers on topics related to use of LED chips, design of LED packages and
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 5
`
`

`

`
`
`design of LED based lighting products and systems. This support included both optical and
`
`thermal design of LED packages and LED lighting products.
`
`15.
`
`From 1996 to 1998 I worked at Emcore Corporation as a Process Engineer. There
`
`I was responsible for qualifying MOCVD capital equipment for production and manufacturing
`
`process development for epitaxial structures. Some of the processes I developed included both
`
`AlInGaP and GaN based LED structures.
`
`16.
`
`I have also authored and presented more than a total of 30 publications,
`
`presentations, and seminars, and I am a named inventor on 30 issued patents and over 25 pending
`
`patents.
`
`17.
`
`I earned a Ph.D. in Chemical Engineering from the University of Florida in 1997,
`
`where my graduate work focused on development of optoelectronic devices, including novel
`
`silicon based visible LEDs and sulfide based TFELD structures and zinc selenide blue LEDs.
`
`18.
`
`In 1989 I worked for Shell Oil Corporation in Norco, LA. I was tasked with
`
`modeling heat exchanger performance for the crude oil distillation column in order to identify cost
`
`saving opportunities. I developed an improved model for determining the operating efficiency of
`
`heat exchange units that determined the economic impact on different operating conditions. The
`
`model was designed to be compatible with all major operating units on site and resulted in costs
`
`savings of over $20 million/year at the Norco Facility. My work was quickly adopted throughout
`
`the entire corporation and has resulted in sustained annual cost savings of approximately $500
`
`million/year.
`
`19.
`
`Based on the above education and experience, I believe that I have a detailed
`
`understanding of the state of the art during the relevant period, as well as a sound basis for opining
`
`how persons of skill in the art at that time would understand the technical issues in this case.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 6
`
`

`

`
`
`III. REVIEW AND USE OF DOCUMENTS
`
`20.
`
`In forming the opinions presented in this report, I have reviewed and relied upon
`
`the following documents:
`
`• U.S. Patent No. 8,201,968;
`
`• U.S. Patent No. 8,672,518;
`
`• U.S. Patent No. 8,967,844;
`
`• U.S. Patent No. 7,828,465 (“Roberge”);
`
`• U.S. Patent No. 7,980,736 (“Soderman”);
`
`• Silescent S100 LP2 Product Sheet and Installation Instructions;
`
`• U.S. Patent No. 7,670,021 (“Chou”);
`
`• Declaration of Dr. Zane Coleman in support of petition for Inter Partes Review of
`
`U.S. Patent No. 8,201,968;
`
`• Declaration of Dr. Zane Coleman in support of petition for Inter Partes Review of
`
`U.S. Patent No. 8,672,518;
`
`• Declaration of Dr. Zane Coleman in support of petition for Inter Partes Review of
`
`U.S. Patent No. 8,967,844;
`
`•
`
`•
`
`•
`
`IPR2017-01287 Petition;
`
`IPR2017-01285 Petition;
`
`IPR2017-01280 Petition;
`
`• U.S. Patent No. 6,616,291 (“Love”);
`
`• U.S. Patent No. 5,463,280 (“Johnson”);
`
`• U.S. Patent No. 7,102,172 (“Lynch”)
`
`• U.S. Patent Application Publication No. US2008/0297060 (“Ko”);
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 7
`
`

`

`
`
`• U.S. Patent No. 7,993,034 (“Wegner”);
`
`• U.S. Patent No. 7,722,227 (“Zhang”);
`
`• Craig DiLouie, “Introducing the LED Driver”, EC&M, September 2004, pp28-32
`
`(“DiLouie”);
`
`• The exhibits to this declaration; and
`
`• All remaining exhibits to IPR2017-01280, IPR2017-01285, and IPR2017-01287.
`
`IV.
`
`LEGAL STANDARDS
`
`21.
`
`I am not an attorney and do not have formal legal training. Counsel for LSG has
`
`informed me about the relevant legal standards.
`
`V.
`
`PERSON OF ORDINARY SKILL IN THE ART
`
`22.
`
`I understand that a person of ordinary skill in the art is one who is presumed to be
`
`aware of all pertinent art, thinks along conventional wisdom in the art, and is a person of ordinary
`
`creativity. A person of ordinary skill in the art (“POSITA”) would have had the knowledge of the
`
`literature concerning LED fixture/lamp design and related arts as of October 2009.
`
`23.
`
`Based on my review of the patent specification and file history of the Patents at
`
`Issue, in my opinion POSITA would have at least a B.S. degree or equivalent in electrical
`
`engineering, mechanical engineering, chemical engineering, physics, or a related field and at least
`
`2-3 years of experience in designing LED lighting products or fixtures. I have applied this
`
`perspective throughout my declaration. I would have been a POSITA by at least December 2002.
`
`VI.
`
`TECHNICAL BACKGROUND
`
`A.
`
`24.
`
`Lighting Fixture Design
`
`At the time of the Patents at Issue, recessed cans were a common type of lighting
`
`fixture. Many different classes of can light fixtures existed for different lighting applications. As
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 8
`
`

`

`
`
`their name implies, can light fixtures are hollow cylindrical housings installed in holes in a ceiling.
`
`The lower rim of the cylinder usually has a flared trim element that sits flush with the surface of
`
`the ceiling. The upper end of the cylinder that resides above the ceiling is closed. Can light fixtures
`
`include an Edison socket that allows installation of a standard format light bulb. Can light fixtures
`
`also include metal junction boxes mounted on the outside of the can structure and both structures
`
`are located above the level of the ceiling. The junction box provides a safe location for making an
`
`electrical connection between the Edison socket and AC mains. Can light fixtures are mounted
`
`using screws, nails or the like to joists, rafters or other physical structures that reside above the
`
`ceiling.
`
`25.
`
`Prior to the development of LED lighting fixtures, the actual lighting units installed
`
`in can light fixtures were light bulbs of various formats including A19 bulbs, PAR30 bulbs, and
`
`PAR38 bulbs. These light sources were installed in the Edison sockets inside can light fixtures.
`
`Incandescent bulbs emit significant amounts of radiant heat which is absorbed by the body of can
`
`light fixtures. Because of this the body of can light fixtures have an inherent ability to dissipate
`
`significant amounts of heat above the level of the ceiling.
`
`26.
`
`Ceiling mounted light fixtures are a separate class of luminaires designed to be
`
`mounted to a junction box. In contrast to the body of can light fixture which are installed above
`
`the level of the ceiling, the body of a ceiling mounted light fixture resides entirely on the occupied
`
`(lower) side of the ceiling. A junction box installed in the ceiling provides a location for
`
`connection to electrical mains and provides the physical mounting mechanism for a ceiling
`
`mounted light fixture.
`
`27.
`
`Ceiling mounted light fixtures are the antithesis of can light fixtures. While both
`
`have significant physical dimensions and require at least one light source to produce light, can
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 9
`
`

`

`
`
`fixtures reside at and above the ceiling and ceiling mounted fixtures reside at and below the ceiling.
`
`The ceiling is a common interface for both classes of fixtures. They are otherwise exclusive.
`
`
`
`[Exhibit B at 10-20. Examples of recessed can lighting fixtures.]
`
`28.
`
`Both fixtures also require junction boxes, but the junction box is an integral part of
`
`a can light fixture while the junction box for ceiling mounted fixtures is an independent component
`
`that must be installed before a ceiling mounted fixture can be installed. In both cases the junction
`
`box is used for electrical connections. No circuitry resides in the junction box.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 10
`
`

`

`
`
`
`[Exhibit B at 10-22. Examples of typical ceiling mounted light fixtures.]
`
`29.
`
`In light of these facts, claim language directed towards covering and sizing of
`
`components to fit within junction boxes and can light fixtures is understood by a POSITA to
`
`require that the fixture be installed using both can light fixtures and junction boxes. Merely
`
`requiring an overlap of outlines would improperly sweep in completely unrelated products such as
`
`streetlamps and even automotive headlamps.
`
`B.
`
`30.
`
`LED Drive Circuits
`
`A drive circuit was commonly used to convert AC mains to a low voltage DC
`
`current suitable for powering LEDs. In general, the efficiency of the drive circuit is inversely
`
`related to the difference in the supply side (input) voltage and the output (LED) side voltage.
`
`Safety and regulatory requirements limit the output voltage to about 42V. Higher output voltages
`
`increase the cost due to increased safety concerns. These safety and regulatory issues limit the
`
`drive circuits to Class 2, isolated designs. These are the most economical and, more importantly,
`
`are required by customers. The requirement for a Class 2, isolated drive circuit limits the
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 11
`
`

`

`
`
`maximum efficiency of the drive circuit. At the time of the Patents at Issue, efficiencies for Class
`
`2, isolated drive circuits ranged from about 86% to about 91%, with most designs falling in the
`
`range of 87-89%.
`
`C.
`
`31.
`
`Efficiency Estimates
`
`Each inefficiency in the components of a fixture results in production of heat. The
`
`efficiencies above can be used to calculate the relative heat produced by each component. As can
`
`be seen in the table below, the vast majority of the heat that must be dissipated in an LED light is
`
`generated by the LEDs.
`
`Component
`
`Efficiency
`
`Input 20W electrical power
`LED Driver
`88%
`LEDs
`29%
`
`Heat
`Production
`-
`2.40 W
`12.50 W
`
`% Heat
`
`-
`16.1%
`83.9%
`
`[Table 1. Heat in a solid-state lighting fixture at the priority date of the Patents at Issue.]
`
`32.
`
`The lifetime of both LEDs and the components of the LED driver are reduced by
`
`exposure to elevated temperatures. Since the LEDs and the LED driver both produce heat, there
`
`is an inherent drive for fixture designers to keep these components separated to reduce the
`
`operating temperature of each.
`
`33.
`
`There is an additional driving force for fixture designers to physically separate the
`
`LEDs and the driver. The efficiency of an LED decreases as temperature increases which means
`
`that at higher temperatures LEDs will produce even greater amounts of heat. Without a very
`
`specific reason, production costs dictate that LED designers would use the smallest possible heat
`
`sinks for given applications.
`
`34.
`
`A suggestion that a POSITA could simply choose to use fewer LEDs ignores the
`
`reality that lighting applications have requirements for the number of lumens delivered by a fixture.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 12
`
`

`

`
`
`Reducing the number of LEDs would require increasing the current through the remaining LEDs
`
`to compensate for the lower lumen output of each LED. This by itself would increase the heat
`
`dissipated by each LED and increase the operating temperature of each LED with no other design
`
`changes.
`
`35.
`
`A suggestion that designers could merely use a more efficient, cooler-running
`
`driver is a gross mischaracterization of the situation. The efficiency analysis above shows that the
`
`driver is responsible for a small portion of the heat generated in an LED lighting fixture at the time
`
`of the Patents at Issue. As shown above, LEDs generated over 80% of the heat that needed to be
`
`dissipated by an LED lighting fixture. Electrical drivers were already rather efficient and
`
`increasing the efficiency by a few percent would not significantly change the heat generated by
`
`the entire fixture.
`
`36. While the efficiency of LEDs has improved over time, these improvements have
`
`come about as a result of the efforts of hundreds of scientists and engineers at dozens of companies
`
`over decades of time and at a cost of billions of dollars. Improving the efficiency of an LED
`
`requires significant technical and scientific knowledge in one or more fields. These fields include:
`
`1) MOCVD process development, 2) semiconductor structure physics/structure design, 3)
`
`semiconductor device fabrication, 4) LED package thermal design, 5) LED package optical design,
`
`6) LED phosphor synthesis, or 7) LED phosphor applications. There is nothing in the Patents at
`
`Issue nor any of Petitioner’s cited references about the methods and technology necessary to
`
`produce a more efficient LED and thus such an endeavor is well beyond the ability of a POSITA.
`
`37.
`
`Alternatively, a position based on a POSITA deliberately choosing inefficient
`
`LEDs that would increase power consumption and heat dissipation requirements, which would in
`
`turn require a larger and more expensive fixture and reduce fixture lifetime is incorrect. Two of
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 13
`
`

`

`
`
`the most desirable benefits of LED lighting are reduced power consumption and long lifetime. In
`
`a similar vein, the steps necessary to significantly improve the efficiency of the LED drive circuit
`
`are beyond the ability of POSITA.
`
`VII. REFERENCES CITED BY PETITIONER
`
`A.
`
`38.
`
`CHOU - U.S. Patent No. 7,670,021
`
` Chou discloses a lighting assembly for installation in a can fixture. The assembly
`
`includes a thermally conductive trim that resides on the outside of the can fixture and a heat sink
`
`that resides inside the can fixture.
`
`39.
`
`Chou asserts that its invention dissipates the majority of its heat via the trim. (Chou
`
`at 5:8-10, 7:14-19). According to the calculations of Chou, this is because 65% of the heat energy
`
`is dissipated by the trim. This does not diminish the importance of dissipating the remaining 35%
`
`of the heat energy, which is accomplished by the heat sink structure located inside the recessed
`
`can.
`
`40.
`
`Chou calculates a temperature of 64.1°C for the trim in a 25°C ambient, when
`
`dissipating 65% of 15 W of heat energy. This gives a thermal resistance of (64.1°C - 25°C)/(0.65
`
`x 15 W) = 4.01°C/W. The heat sink dissipates the remaining heat or about 5.25 W of heat.
`
`41.
`
`At a first approximation it may be assumed that the thermal resistance of the trim
`
`is constant with temperature. Thus, if the trim dissipated the entire heat load, the temperature
`
`would increase by 4.01°C/W x 5.25 W = 21.1°C. This would give a temperature of 64.1°C +
`
`21.1°C = 85.2°C. This would place the LEDs at or above reasonable design temperature limits
`
`and thus any modifications of Chou which exclude the heat sink structure that resides in the
`
`recessed can would render Chou inoperable.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 14
`
`

`

`
`
`42.
`
`A review of the calculations of Chou reveals many inconsistencies, which call into
`
`question claims that the trim dissipates the majority of the heat produced by the fixture.
`
`43.
`
`The standard equation for convective heat transfer is:
`
`Q=hA∆T
`
`where Q is the heat flux
`
`
`
`
`
`
`
`h is the convective heat transfer coefficient
`
`A is the area
`
`ΔT is the temperature difference between the surface and the ambient
`
`44.
`
`This is different from the equation used by Chou (Chou at 5:35):
`
`
`
`Q=ηh∆T
`
`where Q is the heat flux
`
`
`
`
`
`
`
`η is the efficiency
`
`h is the convective heat transfer coefficient
`
`ΔT is the temperature difference between the surface and the ambient
`
`45.
`
`Chou’s efficiency η is a parameter that accounts for the fin efficiency. An ideal fin
`
`would have a constant temperature from its base to edge. However, in real materials the
`
`temperature decreases with distance from the heat sink and is lowest at the distal edge of a fin.
`
`The efficiency of a fin depends on the geometry (thickness, length) and thermal conductivity of
`
`the fin material. According to Chou the efficiency is calculated by:
`
`where η = efficiency and mL is given by:
`
`
`
`(cid:2015)=tanh ((cid:1865)(cid:1838))mL
`(cid:1865)(cid:1838)= (cid:3436)ℎ(cid:1863)(cid:1872)(cid:1838)(cid:3440)(cid:2868).(cid:2873)× (cid:1838)(cid:2869).(cid:2873)
`
`where presumably h = convective heat transfer coefficient
`
` k = thermal conductivity
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 15
`
`

`

`
`
` t = fin thickness
`
` L = fin length
`
`46.
`
`From this it follows that short wide fins will have the highest efficiency.
`
`47. While there are no issues with Chou’s equation for convective heat transfer, issue
`
`is taken with Chou’s use of the same convective heat transfer coefficient for the trim and the heat
`
`sink. Convective heat transfer occurs when the fluid (here, air) adjacent to a heat sink is heated.
`
`Increasing the temperature of the fluid results in a decrease in density. Density differences in turn
`
`result in a buoyant force due to gravity.
`
`48.
`
`As long as fluid remains in contact with the heat sink surface while it is rising, its
`
`temperature continues to increase and the buoyant force increases. When a heat sink is oriented
`
`vertically as is the case for the fin structures of Chou, the fluid (air) maintains contact with the fin
`
`surface and is able to efficiently extract heat. In contrast, the trim in Chou is nearly horizontal.
`
`The trim itself impedes buoyancy driven flow and would be expected to result in a much lower
`
`convective heat transfer coefficient.
`
`49.
`
`Paradoxically, Chou assumes the same convective heat transfer coefficient for the
`
`trim and the heat sink fins. This goes against common knowledge and expectations for convective
`
`heat transfer. As shown by Simmons (Exhibit C) below, the convective heat transfer coefficient
`
`for vertical surfaces at 45°C would be about 5 W/°C m2 which is in line with Chou’s estimate for
`
`the trim. However, the trim is a near horizontal surface that is facing down. At the same
`
`temperatures, the convective heat transfer coefficient would be expected to be at least a factor of
`
`2 lower than that for a vertical surface.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 16
`
`

`

`
`
`[Exhibit C at Fig. 1]
`
`
`
`50.
`
`Chou thus uses an unreasonably high convective heat transfer coefficient for the
`
`trim, which overestimates the contribution of the trim unit for heat dissipation. Further, Chou does
`
`not account for the area on the backside of the trim between the fins, which would dissipate heat
`
`into the recessed can. The convective heat transfer coefficient for this surface would be
`
`significantly higher than the convective heat transfer coefficient of the outer surface of the trim
`
`and only slightly lower than the convective heat transfer coefficient for the heat sink.
`
`51. With respect to radiative heat transfer, Chou inexplicably uses significantly
`
`different values for emissivity of the trim (0.90) and the heat sink (~0.30). If the heat sink were
`
`designed to dissipate heat, a person of ordinary skill would not use a material with such a low
`
`emissivity. It would be expected that the heat sink and the trim would have approximately the
`
`same emissivity.
`
`52.
`
`One possibility for the discrepancy is that Chou assumed the heat sink emissivity
`
`was the same as new galvanized steel (~0.28). Galvanized steel is the most common material used
`
`for construction of can lighting fixtures. However, after production, the tin coating on galvanized
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 17
`
`

`

`
`
`steel quickly oxidizes and the emissivity increases to about 0.88 which is essentially the same as
`
`the emissivity of the trim used by Chou.
`
`53.
`
`By ignoring the orientation difference between the trim and the heat sink, Chou
`
`grossly overestimates the convective heat dissipation capability of the trim. Similarly, Chou
`
`ignores the convective heat transfer from the backside of the trim between the fins of the heat sink,
`
`which would further increase heat dissipation inside the recessed can. Chou grossly
`
`underestimates radiative heat transfer from the heat sink inside the recessed can by using an
`
`unreasonably low value for emissivity.
`
`54.
`
`Chou’s claims that the majority of heat is dissipated by the trim cannot be accepted;
`
`they are based on improper assumptions and parameters that significantly overestimate the amount
`
`of heat dissipated by the trim.
`
`55.
`
`The heatsink of Chou is in direct contact with the trim and heat is transferred
`
`between the components. Chou admits this: “[h]eat energy in remaining in [the] trim is transferred
`
`into [the] heatsink. As such [the] heatsink may be regarded as acting as a heatsink for [the] trim
`
`rather than the light source directly.” (Id. at 5:14-17). This means that the heatsink component of
`
`Chou participates in spreading heat generated by the light source to the edge of the trim. This is
`
`shown schematically below. The light source is highlighted in green and red arrows indicate the
`
`approximate paths of heat flow.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 18
`
`

`

`
`
`[Chou at Fig. 2b. Annotated.]
`
`
`
`56.
`
`Embodiments of Chou are described as using an LED based PAR lamp (Id. at 2:65-
`
`67, 3:12-20). A common PAR lamp used in can light housings is a PAR38 lamp. The
`
`nomenclature refers to the diameter of the lamp in increments of 1/8”. Therefore a PAR38 lamp
`
`is 38/8 = 4.75” in diameter. Chou implicitly acknowledges these dimensions: “[f]ixture 10 is
`
`configured to install into both conventional 12.7 cm (5 inch) and 15.24 cm (6 inch) recessed can
`
`housings.” (Id. at 3:65-67) An attempt to modify Chou to fit within a 4 in can light housing
`
`requires removal of significant portions of the PAR lamp component and would not be attempted
`
`by a POSITA following the teachings of Chou.
`
`57.
`
`The fixture of Chou is designed to be installed inside a can light fixture. Can light
`
`fixtures include integrated electrical junction boxes on the outside of the body of the can in the
`
`space above the ceiling as shown below in Figure 8. Chou does not disclose a light fixture to be
`
`installed within a junction box.
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 19
`
`

`

`
`
`
`[Chou at Fig. 8. Red rectangle outlines location of integral junction box on a recessed can
`
`housing.]
`
`ROBERGE - U.S. Patent No. 7,828,465
`
`Roberge discloses a surface mounted lighting fixture. Roberge is designed to take
`
`B.
`
`58.
`
`advantage of a chimney effect for thermal dissipation. The chimney effect is based on a
`
`fundamental phenomenon associated with convective heat transfer—air is heated by contact with
`
`a hot surface, the air expands and begins to rise due to buoyant forces. As the heated air rises, its
`
`temperature increases while in contact with the hot surface. Increasing the temperature of the air
`
`increases the buoyant forces, which also increases the velocity of the air. (See Roberge at 24:65-
`
`25:2). Further, “a ‘chimney effect’ (also known as a ‘stack effect’) is a movement of air into and
`
`out of structures, e.g. buildings or containers, driven by buoyancy, occurring due to a difference
`
`between interior and exterior air density result from temperature and moisture differences.” (Id.
`
`at 24:65-25-2).
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 20
`
`

`

`
`
`59.
`
`Once it is no longer in contact with the heated surface, the higher temperature air
`
`dissipates the heat it absorbed into the environment. When the air rises, cooler air is drawn from
`
`below to continue the process.
`
`60.
`
`The chimney effect is predicated on maintaining contact of the air with the heated
`
`surface as it moves. Reducing the height of any heat exchange surfaces reduces the driving force
`
`for movement of air and in turn reduces heat transfer.
`
`61.
`
`The driving force for heat transfer to the air is minimized when the hot surface is
`
`horizontally positioned above the air. It is maximized when the hot surface is oriented vertically.
`
`62.
`
`Roberge is aware of this and directs “extraneous surface area of one or more heat-
`
`dissipating elements, not along the trajectory of the cooing air, is omitted.” (Id. at 22:50-52).
`
`Roberge further states that “up to 90% or more of the heat-dissipating surface area is configured
`
`to be within the air flow trajectory through the fixture.” (Id. at 22:57-59).
`
`63.
`
`Indeed, the outer dimension of Roberge is defined by a bezel (330). This must
`
`remain a significant distance from the ceiling in order for the heated air to be dissipated. As the
`
`gap between the ceiling and the bezel is reduced, the driving force for the chimney effect is
`
`reduced—it would be zero if the gap between the bezel and the ceiling were eliminated.
`
`64.
`
`Roberge also recognizes the importance of determining the location and eliminating
`
`regions of stagnant air flow that do not participate in heat transfer in order to make a “more
`
`compact configuration of the heat sink.” (Id. at 25:28-35). In fact, the chimney effect would be
`
`virtually non-existent inside a recessed can.
`
`65.
`
`This is evidenced by Roberge itself: “one object of the invention disclosed herein
`
`is to provide a shallow surface-mount fixture—as shallow as 1”-2” overall height—to alleviate the
`
`undesirable constraints of shallow recess depths for many designers; in fact, it could help many
`
`
`
`PATENT OWNER EXHIBIT 2001
`Page 21
`
`

`

`
`
`projects reclaim up to 6” of ceiling height.” (Id. at 2:12-17). Thus, Roberge is envisioned as
`
`eliminating the use of recessed cans, not as a fixture to be installed in recessed cans.
`
`C.
`
`66.
`
`LOVE - U.S. Patent No. 6,616,291
`
`Love is directed to an underwater lighting assembly operating off of 12VAC or
`
`12VDC. The supply voltage is significantly different from what is used for recessed can fixtures
`
`in ceilings and standard electrical junction boxes. The fixture of Love is designed to be installed
`
`using a water sealing gasket and outer house retaining nut. (Love at 4:29-32). It is designed to be
`
`installed underwater and