`a single block of material. The individual plates each serve a function and
`provide access for machining during the mold-building process. The removal of
`upper-level plates allows access to features at a lower stack level. Having
`individual plates also potentially allows a single plate to be swapped out if
`damaged. The design of plastic injection molds is itself a specialty, and the
`design of optical molds even more so. Mostplastic optic injection molders have a
`mold designer on staff or work closely with a design house that is experienced in
`the design of optical molds. A poorly designed (or fabricated) mold can have a
`detrimental effect on a project. Two general rules of optical injection moldsare:
`“parts can only be as good as the mold they come from”(although they certainly
`can be worse), and “you get what youpayfor.”
`Most production tools are made out of stainless or hardened tool steel, or a
`combination of the two, which enables a longertoollife along with the ability to
`withstand the large forces that occur during injection molding. As a point of
`reference, during injection the mold may see cavity pressures of 700 kg/cm?
`(10,000 psi) or more. With inserted and moving parts of the mold, two different
`material grades are typically used to avoid galling. For
`lower volume or
`prototype tools, alternate softer materials (such as aluminum) are sometimes
`used.
`
`Figure 3.6 Schematic of an injection mold for producing lenses.
`
`Chapter 3
`
`Core Plate
`
`Attachment Plate
`
`Ejector
`Plate
`
`LZ
`Ejector Bar
`
`
`
`
`
`Optical
`J} Insert
`
`, Cooling
`Channels
`
`Attachment Plate
`
`Backing
`Plate
`
`Cavity Plate
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 301 of 550
`Page 301 of 550
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`
`
`Chapter 3
`
`
`
`‘Manufacturing Methods 47
`
`les, as opposed to
`2 a function and
`;. The removal of
`2k level. Having
`: swapped out if
`pecialty, and the
`m molders have a
`iS experienced in
`mold can have a
`ection moldsare:
`igh they certainly
`
`‘d tool steel, or a
`with the ability to
`g. As a point of
`s of 700 kg/cm?
`dd, two different
`ower volume or
`) are sometimes
`
`tachmentPlate
`/
`
`Optical
`Insert
`
`_— Sprue
`
`& Cooling
`Channels
`
`Referring to Fig. 3.6, which shows a schematic of a lens mold, we discuss the
`roles of each of the tool’s constituent plates. Beginning onthe right-hand side,
`wefirst have an attachmentplate, which supplies support to the mold and is used
`to attach it to the platen. Movingto the left, the next plate is the “cavity” plate.
`This plate houses the features that form one ofthe sides of the lens. Adjacent to
`the cavity plate is the “core” plate. This houses the features that form the other
`side of the lens. The mold splits between the core and cavity plates at the
`boundary referred to as the parting line. Behind the coreplate is a backing plate,
`whichprovides support and structure to the mold. Behind this backing plate is the
`ejection mechanism,discussedlater, followed by another attachmentplate, which
`attaches to the moving platen. Thermalinsulating plates are often mounted on the
`exterior of the attachment plates to help with the thermal management of the
`mold during the manufacturing process.
`Alignmentof the mold halves upon closure of the mold is usually achieved
`through sets of guide pins and taper interlocks. These features can be seen in the
`outer portion of each quadrantof the moldhalf seen in Fig. 3.7. The two sets act
`together as somewhat ofa coarse/fine adjustment scheme. The larger guide pins
`engage while the mold halve faces are still apart, and as the mold faces come
`close to one another, the taper interlocks engage, bringing the mold halves to
`
`their final alignment.
`
`lenses.
`
`Figure 3.7 Eight-cavity mold half showing guide pins and taper interlocks.
`(Photograph courtesyof Alan Symmons.)
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 302 of 550
`Page 302 of 550
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`
`
`
`
`Chapter 3
`
`a hybrid between the two systems, is used. This allows a shorter cold runner,
`
`Wecansee a feature onthe right-hand side of Fig. 3.6 that is labeled “sprue.”
`The sprue is the point where the molten plastic enters the mold. Once into the
`mold, the plastic passes through a series of channels Jeading to the lens area.
`These channels are knownas “runners” and can be clearly seen in Fig. 3.5. The
`runners lead to the “gate,” which is the area where the plastic enters the mold
`cavities. As mentioned previously, the cavities are the empty spaces in the mold
`where the lenses are formed. In the particular case of Fig. 3.5, we would say it is
`a four-cavity mold because spaces for four lenses have beencreated.
`The runner systems in molds come in two general categories: hot and cold. In
`a cold runnersystem, a groove is cut into the face of the mold plate, whichfills
`with plastic upon injection. Both the part being molded and the plastic in the
`runner (as well as the gate and the sprue) cool and harden during the cooling
`time. When the parts are ejected from the mold, the runner is ejected as well,
`with the parts attached to it. At some later point in time the parts are removed
`from the runner in a process called “degating.” The runner material is typically
`collected and ground up, becoming “regrind” as mentioned earlier.
`In most
`optical parts regrind is not used, so there is a material amount and cost beyond
`just the lens volumethat must be factored in to the lens production. Cold runner
`systems typically have a semicircular or circular profile, known as “half round”
`or “full round” runners. This shape is used instead of a square profile to facilitate
`pulling the runner from the fixed half of the tool during the opening of the mold
`and ejection of the runner during the ejection motion. A square- or sharp-
`cornered profile is more likely to stick in the mold than a round profile during
`pulling or ejecting.
`In a hot runner system, the runneris internal to the mold plates. The runneris
`kept at an elevated temperature (hence it is a hot runner) such that the plastic
`material
`in it remains molten. While the part cools and hardens during the
`cooling time, the runner does not. When the mold is opened, the parts pull with
`the movingplaten, but there is no runnerto be pulled. The parts alone can now be
`ejected from the moving half of the mold. Instead of grabbing the sprue or runner
`uponejection, the parts themselves must now be captured, which is often done
`using suction cups.
`Compared to a cold runner system, a hot runner system has the advantage
`that there is less material used, since a runner system is not produced with each
`part or set of parts. In addition, the parts are already separated from the runner,
`which removes the need for the degating process. However, compared to a hot
`runner system, a mold with a cold runner system is less complex, significantly
`less costly, and easier to maintain and operate. In some cases, the runner system
`in a cold runner mold is specifically designed to be used as a handling feature in
`later operations, such as coating, degating, and/or assembly. Finally, having a
`cold runner system allows for the mold “packing” described earlier, which can
`help produce superior-quality optical parts. As a result of these factors, most
`optical injection molds use cold runner systems. In some cases, a semihot runner,
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 303 of 550
`Page 303 of 550
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`
`
`Chapter3
`
`Manufacturing Methods
`
`49
`
`is labeled “sprue.”
`yld. Once into the
`' to the lens area.
`n in Fig. 3.5. The
`c enters the mold
`ipaces in the mold
`we would sayit is
`ated.
`'s: hot and cold. In
`| plate, whichfills
`the plastic in the
`luring the cooling
`is ejected as well,
`parts are removed
`aterial is typically
`| earlier.
`In most
`t and cost beyond
`ction. Cold runner
`mas “half round”
`orofile to facilitate
`ening of the mold
`square- or sharp-
`and profile during
`
`ates. The runneris
`ch that the plastic
`ardens during the
`the parts pull with
`; alone can now be
`he sprue or runner
`hich is often done
`
`has the advantage
`roduced with each
`J from the runner,
`compared to a hot
`plex, significantly
`the runner system
`iandling feature in
`Finally, having a
`earlier, which can
`ese factors, most
`, asemihot runner,
`iorter cold runner,
`
`which results in a hotter plastic temperature at the cavity, which allows better
`packing.
`to its
`The gates, while seemingly a small part of the mold, are critical
`ultimate performance. The gate size and shape determine how the plastic will
`flow into the mold cavity, as well as impact the freeze-offtime and packing. Poor
`flow canresult in lens defects. To get the desired flow pattern, gates for optical
`parts are typically much larger than for equivalent nonoptical parts. This is
`because the tolerances for optical surfaces are usually muchtighter than surfaces
`on nonoptical parts. Thesize of the gate required must be kept in mind during the
`optical design,as it sets a minimum required edge thickness.
`It is common for molds to possess a numberofcavities that are a power of
`two, that is, one, two, four, eight, 16, or even 32. This is not a necessity, but it
`works well for having a symmetric Cartesian layout of the cavities within the
`tool. The reason that a symmetric layoutis desired is to have each of the cavities
`the same distance from the sprue, where the plastic enters the mold. This creates,
`theoretically, a situation where eachcavity will receive the molten plastic at the
`sametime, under identical temperature and pressure conditions. The design and
`machining to achieve this state is called “balancing the tool.” Since we generally
`wantall the lenses produced at a given time(in a single shot), and over time as
`well, to be (nearly) identical, it makes sense that we would wanteach cavity to
`have identical conditions.
`Another method to achieve equal flow lengths is to use a radial (or spoke)
`runner system.In this case,it is easy to achieve the equal-length runners because
`the parts lie on a circle with the sprue at the center. The numberofcavities do not
`need to be a power of two but can besetat the desired angular spacing. This form
`of runner system can be useful
`later in the production process. The runner
`system, with lenses attached, can be placed ontoa rotary table for easy manual or
`automated handling of the lenses. Intermediate structures or indexing features
`can also be addedto the runner system to help with automation.
`Having discussed how theplastic gets to the lens cavities, we now consider
`the cavities themselves. Thecavities are the spaces generally (but not necessarily
`exactly) complementary to the shape of the lens. They typically have a region to
`form the mechanical structure of the lens, such as flanging, and another region
`that forms the optical surface. There are three general methods of creating
`cavities. The first, and simplest, is to directly machine the cavity shape into the
`cavity plate. This reduces the complexity of the mold and requires no additional
`pieces to be fabricated. As mentioned previously, most production tools are made
`from hardened materials, which can makeit difficult to achieve an optical quality
`surface. The material can be polished; however, this can be time consuming and
`difficult, particularly if the optic surface is significantly recessed from the face of
`the mold. Another downsideto this methodis that it does not allow for the cavity
`to be easily replaced if it is damaged. Because of these tworeasons, this method
`is morelikely to be used in a tool for prototype or low-volume production when a
`softer mold material and reduced mold lifetime are acceptable.
`
`
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 304 of 550
`Page 304 of 550
`
`
`
`
`
`
`
`50 Chapter3
`
`The second methodis to insert the areas that formthe optical surfaces, while
`putting the mechanical
`features of the lens directly into the mold plate. The
`optical surface is typically formedby the “optic pin” or “optic insert,” whichis a
`pin or rod that
`is inserted into a hole cut
`into the mold plate. The optic pin is
`fabricated separately from the moldplate, typically from a different material or
`material grade. The optic pin may consist ofa solid steel pin with an optical
`surface polished onto the end.
`In the case of production tools,
`this method
`requires the polishing of hardened steel pins. Polaroid extensively used this
`particular methodof fabricating optic pins during the productionofplastic lenses
`for their cameras.
`Alternately, and more commonly,the optic pinstarts as a steel rod into which
`a sphere or an approximate or identical displaced surface to the final optic
`surface is first machined. A layerof nickel is then coated onto the end ofthe pin,
`and the desired optical surface is diamond turnedinto the nickel. Comparedto
`polishing hardenedsteel, the advantages ofthe nickel-plating methodare reduced
`cost and schedule, as well as removing the needfor highly trained opticians or
`other personnel to polish the inserts. If the nickel is sufficiently thick,
`it allows
`for recutting of the optic surface, which may be required if the mold process
`changesorif there is damage to the optic pin. The downsideof the nickel plating
`methodis that diamond-turnable nickel is much softer than hardenedsteel. It
`is
`more susceptible to scratching and other damage that can occur during molding,
`mold maintenance and cleaning, or when swapping inserts (if the same mold
`cavity is used to form different lenses). Because of this,
`it
`is common to have
`additional optic pins on hand as spares. Based on the thickness of the nickel,
`there are a limited numberofrecuts that can occur before the diamond-turning
`tool breaks through the nickel to the steel, usually resulting in damage to the
`diamondtool. It may be possible to machine orstrip the nickel, replate the optic
`pin, and cut a newoptical surface. This can sometimes be required if the
`underlying approximate surface machinedin the steel was not accurate enough.
`or if there is extensive pin damage.
`The availability of modern diamond-turning equipment has made the
`manufacture ofoptic pins by nickel plating and diamondturning afairly routine
`process, and has in many cases eliminated the need forpolishedsteel pins. In the
`case of low-volumeor prototype tools, the steel- and nickel-plating process may
`be eliminated, with the optic pin made directly from a diamond-turnable material,
`such as copper-nickel or aluminum. This will againresult in a softer optical pin,
`which is more prone to damage, but repeating the diamond-turning processfairly
`easily repairs the optic surface ofthis type ofpin.
`Whenusing inserted optic pins, shims are typically usedto adjust their axial
`position. The axial position ofthe optic pins will set both the center thickness of
`the optical element andthe distance from the optic surface to a reference feature
`on the element, such as aflange. This flange offset distance is referred to as the
`“stack.” The standard method of shimming is to use thick metal shims, which
`start at a thickness greater than or equal to that neededtoset the appropriate pin
`
`
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 305 of 550
`Page 305 of 550
`
`
`
`Chapter 3
`
`Manufacturing Methods
`
`51
`
`‘al surfaces, while
`mold plate. The
`insert,” which is a
`. The optic pin is
`ferent material or
`n with an optical
`yols,
`this method
`nsively used this
`n ofplastic lenses
`
`eel rod into which
`to the final optic
`the end ofthe pin,
`‘kel. Compared to
`iethod are reduced
`ained opticians or
`ly thick, it allows
`the mold process
`‘the nickel plating
`irdened steel. It is
`ir during molding,
`if the same mold
`common to have
`iess of the nickel,
`2 diamond-turning
`in damage to the
`l, replate the optic
`ye required if the
`t accurate enough,
`
`nt has made the
`ng a fairly routine
`d steel pins. In the
`lating process may
`|-turnable material,
`softer optical pin,
`ning process fairly
`
`o adjust their axial
`center thickness of
`a reference feature
`5 referred to as the
`yetal shims, which
`the appropriate pin
`
`the shims are adjusted by
`
`locations. Once initial parts have been molded,
`grinding them down to the appropriate length.
`The third method of creating the cavities is to insert the cavity itself into the
`moldplate. In this method,a holeis cut into the mold plate and a separate cavity
`insert is machined. Wheninserting the cavities, it is commonto insert the optic
`pins as well. A hole is machined in the mold plate to take the cavity piece, and a
`hole is machined in the cavity piece to take the optic pin. In this case, the piece
`containing the cavity and the hole for the optic pin is often called a “receiver,”
`“receiver set,” or “cavity set” when referring to the cavity inserts for both sides
`of the mold. A photo of such a cavity set is shown in Fig. 3.8, where the two
`sides of the cavity set are apart.
`This method of creating the cavities has several advantages. With inserted
`cavity sets, each cavity can be machined asa separate piece from the moldplate.
`This allows the production of spare cavities, which can be swapped out in the
`case of damageto one ofthe cavities being used in the mold. Individual cavities
`can also be removed from the mold, to be reworked or repaired, without having
`to remove the mold from the molding machine. In a higher cavitation mold, one
`
`
`
`Figure 3.8 Inserted cavity set for injection mold. Note the tapers on the two
`halves, which interlock the cavity set. (Photograph courtesy of Alan Symmons.)
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 306 of 550
`Page 306 of 550
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`
`
`
`
`
`
`>
`Chapter 3
`
`52
`
`or a few cavities can be machined and tested before a commitment is made to
`machine all
`the cavities.
`In this way, a single mold can be built, with the
`capability to increase the molding capacity as product demand increases. The
`holes where the other cavity sets would go are typically plugged with a receiver
`that has not had the cavity features machined, or by using shutoffs in the runner
`system. In addition to swappingoutcavities that form a single lens, it is possible,
`with reasonably similar lens sizes,
`to swap all
`the cavity sets out and replace
`them with the cavity sets of another lens. In this way, the mold base becomes
`common, and multiple lenses can be made fromit.
`The mold can also be configured with a combinationof cavities for multiple
`lenses. As an example, an eight-cavity mold may have four cavity sets of one
`lens form, and four cavity sets of a different lens form. This situation, where one
`mold is configured to make multiple part forms, is known as a “family mold.” A
`family mold does not require inserted cavities but can be made by any ofthe
`three cavity creation methods. Even though the family mold could theoretically
`make two types of lenses at the same time (of the same material), this is not
`commonly done. The reasonis that the mold process parameters are likely to be
`different for the two lenses, even if the lenses are somewhat similar to one
`another. We mentioned earlier the concept of having a balanced mold. Having
`two different lens form cavities in the mold at the same timeis unlikely to result
`in a balanced mold. If the mold is processed forthe first lens, the other lens will
`likely not meet its quality requirements, and vice versa. Balancing the process
`between the two lenses oftenresults in both lenses being of inferior quality. As a
`result, a family mold is typically run by shutting off the second set of lenses,
`moldingthe first set, then shutting off the first set and molding the secondset of
`lenses.
`the potentially higher cost
`is
`The downside to cavity-inserted molds
`associated with the additional pieces and machining. In the case of a common
`mold base used with multiple lens form cavity sets, there is the risk of stopping
`production on multiple lens elements if the mold base is damaged and downfor
`repair. However, the flexibility that the inserted cavity sets allow, particularly in
`high-rate production, often outweighs the initial additional cost andrisk of this
`mold form.
`In addition to the insertion of the optic pins and cavities, the gates of the
`mold can also be inserted. Inserting the gates allows them to be adjusted without
`having to machine the mold plate directly. This allows for different sizes and
`shapes of gates to be evaluated, without the risk of opening up the gate too much.
`In the case of a mold with a large numberofcavities, inserting the gate may help
`with balancing the tool by allowing the individual cavity gates to be adjusted. It
`also allows the gate to be replaced in the event of damage.
`Since the mold is not normally operated under vacuum conditions, when first
`closed,
`the runners, gates, and cavities are filled with air. When the injection
`cycle begins, plastic flows through the runners, the gates, and into the cavities.
`The air in the runners, gates, and cavities, if not allowed to escape, will end up
`being trapped in the mold. In order to allow the air to escape, “vents” are often
`
`1)
`
`i
`|
`
`~~
`
`Ld
`
`
`
`(i
`
`|
`
`;
`t
`
`4
`0
`
`x
`
`C
`t
`t
`i
`
`§
`t
`a
`
`o
`
`
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 307 of 550
`Page 307 of 550
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`
`
`—___Chapter3
`
`Manufacturing Methods
`
`53
`
`1itment is made to
`be built, with the
`ind increases. The
`ted with a receiver
`t offs in the runner
`lens, it is possible,
`!s out and replace
`old base becomes
`
`vities for multiple
`cavity sets of one
`uation, where one
`“family mold.” A
`ide by any of the
`‘ould theoretically
`terial), this is not
`rs are likely to be
`at similar to one
`‘ed mold. Having
`unlikely to result
`he other lens will
`icing the process
`‘rior quality. As a
`md set of lenses,
`the secondset of
`
`ally higher cost
`ise of a common
`' risk of stopping
`‘ed and down.for
`W, particularly in
`t and risk of this
`
`the gates of the
`adjusted without
`fferent sizes and
`e gate too much.
`1e gate may help
`o be adjusted.It
`
`tions, whenfirst
`en the injection
`nto the cavities.
`ape, will end up
`vents” are often
`
`incorporated. Venting usually takes place in one or both of two ways. Thefirst
`methodis to cut a series of shallow grooves into the face of the mold. This type
`of vent can be seen in Fig. 3.5 as the diagonal grooves coming fromthe cavities,
`as well as the grooves coming from the runners. A vent
`is also visible (the
`horizontal groove) on the left-hand side of the cavity set shown in Fig. 3.8. The
`deeper groove on the right-handside ofthe receiveris the end ofthe runner and
`the gate.
`The second methodofventing is to have the air go around and downthepin
`that forms the optical surface. If there is improper venting, the air in the mold
`will be trapped and will end up being compressed by the injected plastic. This
`can result in “burning” due to the rapid compression, which leads to a scorched
`optic pin surface or the part. In somecases, a poor venting condition will appear
`as a small spherical imprint in the part due to the bubble of highly compressed
`trapped air. Improper venting can occur due to poor vent design, or due to the
`vents gradually clogging asthe tool is used. Injection moldsare regularly taken
`out of service for a short period of time to receive cleaning and maintenance in
`order to prevent this and otherpotential failures.
`Another feature of the mold is a system of heating and cooling channels.
`These channels
`run through the mold plates,
`typically near
`the cavities.
`Occasionally, channels are also run up the center of the optic pin, where they are
`referred to as “bubblers.” In the channels, water or oil is circulated through the
`mold. For optical parts oil
`is more common, as the higher mold temperatures
`associated with them can turn the water to steam. In addition, depending on the
`material used to construct the mold, using oil
`instead of water will prevent
`corrosion. The channels are connected to hoses, which are in turn connected to a
`thermal conditioning device known as a thermolater. The thermolater maintains
`the proper oil or water temperature. As mentionedearlier, once the moltenplastic
`is injected, heat transfer determines how long the mold must remain closed. The
`fluid moving throughthe channels will be at a lower temperature than theplastic,
`so it will draw heat from the molded part. Since the oil or water is also hotter
`than room temperature, it will also heat the mold base. Depending upon the mold
`and molding application, electrical heating elements may also be added to the
`mold, or separate heating and cooling channels may be used. These additional
`mold complexities would typically be used for moredifficult parts, such as those
`containing high-aspect-ratio microstructures.
`Having discussed several features of the mold and how the molten plastic
`arrives to form the part, we now discuss how the (cooled) parts are removed (or
`ejected) from the mold. As mentioned earlier, when the part
`is sufficiently
`hardened, the mold is openedandthe parts are “pulled” with the moving half of
`the mold. The needto pull the parts is considered during the design of the mold.
`For instance, depending on the part shape and profile, it may be oriented in the
`mold with a particular part side on the fixed side of the mold. The runner system
`can also be used to help pull the parts. If a half-round runneris put into the
`moving side of the mold, the runner is much morelikely to stay with the moving
`half than with the fixed half whenthe mold opens.If there is difficulty getting the
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 308 of 550
`Page 308 of 550
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`
`
`
`
`54
`
`Chapter 3
`
`parts to pull, “grippers” can be added to mold. These are small features put into
`the part or runner that provide added pulling power. If they are addedtothe part,
`they are typically placed in an inconspicuous, noncritical area. Sometimes
`additional pulling power can be achieved by slightly roughening an overly
`polished noncritical surface.
`With the mold opened and pulling achieved, the parts must now beejected
`from the mold. There are three main ejection methods used, sometimes in
`conjunction with each other. The first method relies upon pins, blades, sleeves or
`plates to push the parts out of the moving half of the mold. Ejection (or ejector)
`pins are the most commonofthese devices, so we generically use the term pin in
`the rest of our discussion. The ejector pins run back through the core and backing
`plate to the ejector mechanism orejector plate. When the time comestoeject the
`parts,
`the ejector plate is driven forward by the ejector bar. This moves the
`ejector pins forward, pushing the molded parts out of the moving mold half
`cavities. The ejector pins are typically fairly small in diameter compared to the
`part, and often several of them are positioned around the part. Pushing from
`several points aroundthe part helps preventtilt during the ejection motion, which
`could cause the part to become stuck in the cavity. Closing the mold on a part
`that did not properly eject can damage the cavity or optic inserts. Sensors are
`sometimes used to prevent the mold from closing if all of the cavities are not
`clear.
`The ejector pins are normally positionedat or slightly off the face of part. As
`a result, the pin leaves a witness mark on the part. During the design of the mold
`and the part, “keep-out zones” for ejector marks must be considered if ejector
`pins are to be used. In addition to pushing the part out of the cavity, in a mold
`with a cold runner system (the most commoncase) the runner must be pushed out
`of the mold as well. Additional ejector pins are positioned along the runner
`system to push it out. These pins are normally attached to the same ejector plate
`as those that push out the part, so the runner and part comeouttogether.
`The second ejection method uses the optic pin as the ejection pin. This
`method is often referred to as “optical eject.” In this case, the optic pin runs
`through the mold plates and is connected to the ejection plate. Similar to the
`ejector pin case, the ejection plate moves forward, and the optic pin pushes the
`part out of the cavity. Figure 3.9 showsa cavity insert with the optic pin pushed
`forward. The bar attaching the optic pin to the ejector mechanism can be seen on
`the far left side of the photograph. Since the optic surface is often a significant
`percentage ofthe areaofthe lens, ejecting with the optic pin reduces the chance
`of the parttilting during ejection. As in the previous method,in the case of a cold
`runner system, additional ejector pins can be usedto eject the runners.
`Thethird ejection mechanism uses compressed air instead of physical ejector
`pins. This method is less commonlyused,as it requires additional equipment and
`is not as easily controlled as the motion of an ejector pin. Compressed air is
`sometimesused as anassisting ejection mechanism for the above two methods.
`
`
`
`Me
`
`,
`
`|
`
`(aa
`|
`
`Figars
`metig
`Th
`methd
`distort
`diamel
`back d
`its hold
`eases,
`saa
`optical
`acceptd
`prefere
`Td
`molded
`sSurtacg
`
`es
`hotizd
`horizo
`varving
`“We
`to cred
`“straig
`the ling
`
`4
`
`Q
`
`j
`
`|
`4
`
`!
`|
`|
`
`‘
`
`4
`
`
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 309 of 550
`Page 309 of 550
`
`
`
`
`
`>
`
`____Chapter 3
`
`small features put into
`y are addedto the part,
`tical area. Sometimes
`roughening an overly
`
`s must now be ejected
`ls used, sometimes in
`pins, blades, sleeves or
`d. Ejection (or ejector)
`ally use the term pin in
`th the core and backing
`time comesto eject the
`t bar. This moves the
`the moving mold half
`meter compared to the
`he part. Pushing from
`ejection motion, which
`ing the mold on a part
`‘ic inserts. Sensors are
`of the cavities are not
`
`off the face of part. As
`the design of the mold
`e considered if ejector
`f the cavity, in a mold
`ner must be pushed out
`yned along the runner
`the same ejector plate
`out together.
`the ejection pin. This
`ise, the optic pin runs
`n plate. Similar to the
`ie optic pin pushes the
`th the optic pin pushed
`hanism can be seen on
`e is often a significant
`pin reduces the chance
`yd, in the case of a cold
`the runners.
`ead of physical ejector
`ditional equipment and
`vin. Compressed air is
`above two methods.
`
`
`
`
`
`Manufacturing Methods 55
`
`
`
`Figure 3.9 Cavity insert with optic pin moved forward showing optical ejection
`method. (Photograph courtesy of Alan Symmons.)
`
`There is sometimes a debate among or within vendorsasto the best ejection
`method to use. Some argue that using ejector pins on optical parts can introduce
`distortion of the optic surface due to nonuniform push-out forces over the part
`diameter. On the other hand, using optical eject requires the optic pin to slide
`back and forth, which requires a slightly larger gap betweenthe optic insert and
`its hole, possibly resulting in a larger decenter of the optical surface. In some
`cases, the part size may rule out one methodor the other. For instance, on very
`small optics, there may not be room to use ejector pins, requiring the use of the
`optical
`injection method. When either method (ejector pins or optic pin) is
`acceptable,
`the choice may come down to the vendor’s experience and
`preference.
`To enable pulling upon the mold opening, and to allow for easy injection,
`molded parts are typically designed with “draft” on them. Draft is the angling of
`surfaces that would otherwise be parallel to the mold opening direction. For a
`standard horizontal molding machine, as shownearlier, draft would be added to
`horizontal surfaces. The amount ofdraft required depends uponthe length ofthe
`horizontal surface, as well as the material that is used, as different materials have
`varying tendenciesto stick in the mold cavities.
`Wediscussed earlier that one of the advantages ofplastic opticsis the ability
`to create complex parts. The molds that we have discussed so far are simple
`“straight-draw” tools. By straight draw, we meanthat there is only one motion:
`the linear translation of the moving mold half involved in opening the mold. This
`
`
`
`Apple Ex. 1022
`Apple Ex. 1022
`
`Apple v. Corephotonics
`Apple v. Corephotonics
`
`Page 310 of 550
`Page 310 of 550
`
`
`
`56 Chapter 3
`
`: | a|~
`
`
`type of mold is the most common,and it works well for standard lens elements,
`adjust
`However, due to their complexity, certain parts cannot be molded using a
`, mechit
`straight-draw tool.
`from tll
`Figure 3.10 shows an example of a part that would not normally be molded
`with the
`with a simple straight-draw tool. This is a four-channel telecom device, where a
`out off
`fiber-optic ferrule wouldbe inserted into the left-hand side. After passing through
`threadal
`the collimating lens, the four wavelengths oflight emerging from the fiber would
`large a
`reflect off the TIR mirror surface and pass through a filter/mirror assembly,
`range.
`