Device for image capture and processing

[0001] The present application is a continuation of U.S. application Ser. No. 10/729,150 filed on Dec. 8, 2003, now issued U.S. Pat. No. 6,948,794, which is a continuation of U.S. application Ser. No. 09/112,774 filed on Jul. 10, 1998 (now abandoned), the entire contents of which are herein incorporated by reference.

[0002] The present invention relates substantially to the concept of a disposable camera having instant printing capabilities and in particular, discloses a printhead re-capping assembly for a digital camera system.

[0003] Recently, the concept of a “single use” disposable camera has become an increasingly popular consumer item. Disposable camera systems presently on the market normally include an internal film roll and a simplified gearing mechanism for traversing the film roll across an imaging system including a shutter and lensing system. The user, after utilizing a single film roll returns the camera system to a film development center for processing. The film roll is taken out of the camera system and processed and the prints returned to the user. The camera system can then be re-manufactured through the insertion of a new film roll into the camera system, the replacement of any worn or wearable parts and the re-packaging of the camera system in accordance with requirements. In this way, the concept of a single use “disposable” camera is provided to the consumer.

[0004] Recently, a camera system has been proposed by the present applicant which provides for a handheld camera device having an internal print head, image sensor and processing means such that images sense by the image sensing means, are processed by the processing means and adapted to be instantly printed out by the printing means on demand. The proposed camera system further discloses a system of internal “print rolls” carrying print media such as film on to which images are to be printed in addition to ink to supplying the printing means for the printing process. The print roll is further disclosed to be detachable and replaceable within the camera system.

[0005] Unfortunately, such a system is likely to only be constructed at a substantial cost and it would be desirable to provide for a more inexpensive form of instant camera system which maintains a substantial number of the quality aspects of the aforementioned arrangement.

[0006] It would be further advantageous to provide for the effective interconnection of the sub components of a camera system.

[0007] According to a first aspect of the invention, there is provided a printhead re-capping assembly for a printer having a chassis, a platen assembly and a pagewidth printhead operatively mounted on the chassis to carry out a printing operation on print media passing over the platen assembly, the re-capping assembly comprising

[0008] a base structure that is mounted on the chassis;

[0009] at least one static solenoid that is mounted on the base structure and that is connected to an electrical power supply of the printer;

[0010] a support member that is actuable by the solenoid to be movable with respect to the chassis between an operative position and an inoperative position; and

[0011] a printhead capping member that is mounted on the support member such that when the support member is in the operative position, the capping member engages the printhead to cap the printhead and when the support member is in the inoperative position, the capping member is disengaged from the printhead.

[0012] The support member may be configured to be normally in the operative position and to move into the inoperative position when the solenoid is energized by the electrical power supply.

[0013] A biasing mechanism may be engaged with the support member to bias the support member into the operative position when the solenoid is de-energized.

[0014] The base structure and the solenoid may both be elongate to correspond with a length of the printhead.

[0015] The support member may also be elongate and may correspond generally with the printhead.

[0016] The capping member may include a length of sponge that is dimensioned to cover the printhead when the support member is displaced into its operative position.

[0017] A sealing member may be positioned on the support member to bound the length of sponge such that, when the length of sponge caps the printhead, the sealing member serves to seal a region about the printhead.

[0018] In accordance with a second aspect of the present invention, there is provided in a camera system comprising: an image sensor device for sensing an image; a processing means for processing the sensed image; a print media supply means for the supply of print media to a print head; a print head for printing the sensed image on the print media stored internally to the camera system; a portable power supply interconnected to the print head, the sensor and the processing means; and a guillotine mechanism located between the print media supply means and the print head and adapted to cut the print media into sheets of a predetermined size.

[0019] Further, preferably, the guillotine mechanism is detachable from the camera system. The guillotine mechanism can be attached to the print media supply means and is detachable from the camera system with the print media supply means. The guillotine mechanism can be mounted on a platen unit below the print head.

[0044] Turning initially simultaneously to FIG. 1 and FIG. 2 there are illustrated perspective views of an assembled camera constructed in accordance with the preferred embodiment with FIG. 1 showing a front perspective view and FIG. 2 showing a rear perspective view. The camera 1 includes a paper or plastic film jacket 2 which can include simplified instructions 3 for the operation of the camera system 1. The camera system 1 includes a first “take” button 4 which is depressed to capture an image. The captured image is output via output slot 6. A further copy of the image can be obtained through depressing a second “printer copy” button 7 whilst an LED light 5 is illuminated. The camera system also provides the usual view finder 8 in addition to a CCD image capture/lensing system 9.

[0045] The camera system 1 provides for a standard number of output prints after which the camera system 1 ceases to function. A prints left indicator slot 10 is provided to indicate the number of remaining prints. A refund scheme at the point of purchase is assumed to be operational for the return of used camera systems for recycling.

[0046] Turning now to FIG. 3, the assembly of the camera system is based around an internal chassis 12 which can be a plastic injection molded part. A pair of paper pinch rollers 28, 29 utilized for decurling are snap fitted into corresponding frame holes eg. 26, 27.

[0047] As shown in FIG. 4, the chassis 12 includes a series of mutually opposed prongs eg. 13, 14 into which is snapped fitted a series of electric motors 16, 17. The electric motors 16, 17 can be entirely standard with the motor 16 being of a stepper motor type. The motor 16, 17 include cogs 19, 20 for driving a series of gear wheels. A first set of gear wheels is provided for controlling a paper cutter mechanism and a second set is provided for controlling print roll movement.

[0048] Turning next to FIGS. 5 to 7, there is illustrated an ink supply mechanism 40 utilized in the camera system. FIG. 5 illustrates a back exploded perspective view, FIG. 6 illustrates a back assembled view and FIG. 7 illustrates a front assembled view. The ink supply mechanism 40 is based around an ink supply cartridge 42 which contains printer ink and a print head mechanism for printing out pictures on demand. The ink supply cartridge 42 includes a side aluminium strip 43 which is provided as a shear strip to assist in cutting images from a paper roll.

[0049] A dial mechanism 44 is provided for indicating the number of “prints left”. The dial mechanism 44 is snap fitted through a corresponding mating portion 46 so as to be freely rotatable.

[0050] As shown in FIG. 6, the mechanism 40 includes a flexible PCB strip 47 which interconnects with the print head and provides for control of the print head. The interconnection between the Flex PCB strip an d an image sensor and print head integrated circuit can be via Tape Automated Bonding (TAB) Strips 51, 58. A moulded aspherical lens and aperture shim 50 (FIG. 5) is also provided for imaging an image onto the surface of the image sensor integrated circuit normally located within cavity 53 and a light box module or hood 52 is provided for snap fitting over the cavity 53 so as to provide for proper light control. A series of decoupling capacitors eg. 34 can also be provided. Further a plug 45 (FIG. 7) is provided for re-plugging ink holes after refilling. A series of guide prongs eg. 55–57 are further provided for guiding the flexible PCB strip 47.

[0051] The ink supply mechanism 40 interacts with a platen unit 60 which guides print media under a printhead located in the ink supply mechanism. FIG. 8 shows an exploded view of the platen unit 60, while FIGS. 9 and 10 show assembled views of the platen unit. The platen unit 60 includes a first pinch roller 61 which is snap fitted to one side of a platen base 62. Attached to a second side of the platen base 62 is a cutting mechanism 63 which traverses the platen unit 60 by means of a rod 64 having a screw thread which is rotated by means of cogged wheel 65 which is also fitted to the platen base 62. The screw threaded rod 64 mounts a block 67 which includes a cutting wheel 68 fastened via a fastener 69. Also mounted to the block 67 is a counter actuator which includes a pawl 71. The pawl 71 acts to rotate the dial mechanism 44 of FIG. 6 upon the return traversal of the cutting wheel. As shown previously in FIG. 6, the dial mechanism 44 includes a cogged surface which interacts with pawl 71, thereby maintaining a count of the number of photographs by means of numbers embossed on the surface of dial mechanism 44. The cutting mechanism 63 is inserted into the platen base 62 by means of a snap fit via clips 74.

[0052] The platen unit 60 includes an internal recapping mechanism 80 for recapping the print head when not in use. The recapping mechanism 80 includes a sponge portion 81 and is operated via a solenoid coil so as to provide for recapping of the print head. In the preferred embodiment, there is provided an inexpensive form of printhead re-capping mechanism provided for incorporation into a handheld camera system so as to provide for printhead re-capping of an inkjet printhead.

[0053] FIG. 11 illustrates an exploded view of the recapping mechanism whilst FIG. 12 illustrates a close up of the end portion thereof. The re-capping mechanism 80 is structured around a solenoid including a 16 turn coil 75 which can comprise insulated wire. The coil 75 is turned around a first stationery solenoid arm 76 which is mounted on a bottom surface of the platen base 62(FIG. 8) and includes a post portion 77 to magnify effectiveness of operation. The arm 76 can comprise a ferrous material.

[0054] A second moveable arm 78 of the solenoid actuator is also provided. The arm 78 is moveable and is also made of ferrous material. Mounted on the arm is a sponge portion surrounded by an elastomer strip 79. The elastomer strip 79 is of a generally arcuate cross-section and act as a leaf spring against the surface of the printhead ink supply cartridge 42 (FIG. 5) so as to provide for a seal against the surface of the printhead ink supply cartridge 42. In the quiescent position an elastomer spring unit 87, 88 acts to resiliently deform the elastomer seal 79 against the surface of the ink supply unit 42.

[0055] When it is desired to operate the printhead unit, upon the insertion of paper, the solenoid coil 75 is activated so as to cause the arm 78 to move down to be adjacent to the end plate 76. The arm 78 is held against end plate 76 while the printhead is printing by means of a small “keeper current” in coil 75. Simulation results indicate that the keeper current can be significantly less than the actuation current. Subsequently, after photo printing, the paper is guillotined by the cutting mechanism 63 of FIG. 8 acting against Aluminium Strip 43, and rewound so as to clear the area of the re-capping mechanism 80. Subsequently, the current is turned off and springs 87, 88 return the arm 78 so that the elastomer seal is again resting against the printhead ink supply cartridge.

[0056] It can be seen that the preferred embodiment provides for a simple and inexpensive means of re-capping a printhead through the utilisation of a solenoid type device having a long rectangular form. Further, the preferred embodiment utilises minimal power in that currents are only required whilst the device is operational and additionally, only a low keeper current is required whilst the printhead is printing.

[0057] Turning next to FIGS. 13 and 14, FIG. 13 illustrates an exploded perspective of the ink supply cartridge 42 whilst FIG. 14 illustrates a close up sectional view of a bottom of the ink supply cartridge with the printhead unit in place. The ink supply cartridge 42 is based around a pagewidth printhead 102 which comprises a long slither of silicon having a series of holes etched on the back surface for the supply of ink to a front surface of the silicon wafer for subsequent ejection via a micro electro mechanical system. The form of ejection can be many different forms such as those set out in the tables below.

[0058] Of course, many other inkjet technologies, as referred to the attached tables below, can also be utilised when constructing a printhead unit 102. The fundamental requirement of the ink supply cartridge 42 is the supply of ink to a series of colour channels etched through the back surface of the printhead 102. In the description of the preferred embodiment, it is assumed that a three colour printing process is to be utilised so as to provide full colour picture output. Hence, the print supply unit includes three ink supply reservoirs being a cyan reservoir 104, a magenta reservoir 105 and a yellow reservoir 106. Each of these reservoirs is required to store ink and includes a corresponding sponge type material 107–109 which assists in stabilising ink within the corresponding ink channel and inhibiting the ink from sloshing back and forth when the printhead is utilised in a handheld camera system. The reservoirs 104, 105, 106 are formed through the mating of first exterior plastic piece 110 and a second base piece 111.

[0059] At a first end 118 of the base piece 111 a series of air inlet 113–115 are provided. Each air inlet leads to a corresponding winding channel which is hydrophobically treated so as to act as an ink repellent and therefore repel any ink that may flow along the air inlet channel. The air inlet channel further takes a convoluted path assisting in resisting any ink flow out of the chambers 104–106. An adhesive tape portion 117 is provided for sealing the channels within end portion 118.

[0060] At the top end, there is included a series of refill holes (not shown) for refilling corresponding ink supply chambers 104, 105, 106. A plug 121 is provided for sealing the refill holes.

[0061] Turning now to FIG. 14, there is illustrated a close up perspective view, partly in section through the ink supply cartridge 42 of FIG. 13 when formed as a unit. The ink supply cartridge includes the three colour ink reservoirs 104, 105, 106 which supply ink to different portions of the back surface of printhead 102 which includes a series of apertures 128 defined therein for carriage of the ink to the front surface.

[0062] The ink supply cartridge 42 includes two guide walls 124, 125 which separate the various ink chambers and are tapered into an end portion abutting the surface of the printhead 102. The guide walls 124, 125 are further mechanically supported by block portions eg. 126 which are placed at regular intervals along the length of the ink supply unit. The block portions 126 leave space at portions close to the back of printhead 102 for the flow of ink around the back surface thereof.

[0063] The ink supply unit is preferably formed from a multi-part plastic injection mould and the mould pieces eg. 110, 111 (FIG. 13) snap together around the sponge pieces 107, 109. Subsequently, a syringe type device can be inserted in the ink refill holes and the ink reservoirs filled with ink with the air flowing out of the air outlets 113–115. Subsequently, the adhesive tape portion 117 and plug 121 are attached and the printhead tested for operation capabilities. Subsequently, the ink supply cartridge 42 can be readily removed for refilling by means of removing the ink supply cartridge, performing a washing cycle, and then utilising the holes for the insertion of a refill syringe filled with ink for refilling the ink chamber before returning the ink supply cartridge 42 to a camera.

[0064] Turning now to FIG. 15, there is shown an example layout of the Image Capture and Processing integrated circuit (ICP) 48.

[0065] The Image Capture and Processing integrated circuit 48 provides most of the electronic functionality of the camera with the exception of the print head integrated circuit. The integrated circuit 48 is a highly integrated system. It combines CMOS image sensing, analog to digital conversion, digital image processing, DRAM storage, ROM, and miscellaneous control functions in a single integrated circuit.

[0066] The integrated circuit is estimated to be around 32 mm² using a leading edge 0.18 micron CMOS/DRAM/APS process. The integrated circuit size and cost can scale somewhat with Moore's law, but is dominated by a CMOS active pixel sensor array 201, so scaling is limited as the sensor pixels approach the diffraction limit.

[0067] The ICP 48 includes CMOS logic, a CMOS image sensor, DRAM, and analog circuitry. A very small amount of flash memory or other non-volatile memory is also preferably included for protection against reverse engineering.

[0068] Alternatively, the ICP can readily be divided into two integrated circuits: one for the CMOS imaging array, and the other for the remaining circuitry. The cost of this two integrated circuit solution should not be significantly different than the single integrated circuit ICP, as the extra cost of packaging and bond-pad area is somewhat cancelled by the reduced total wafer area requiring the color filter fabrication steps.

The ICP preferably contains the following functions:

[0069]

	Function
	1.5 megapixel image sensor
	Analog Signal Processors
	Image sensor column decoders
	Image sensor row decoders
	Analogue to Digital Conversion (ADC)
	Column ADC's
	Auto exposure
	12 Mbits of DRAM
	DRAM Address Generator
	Color interpolator
	Convolver
	Color ALU
	Halftone matrix ROM
	Digital halftoning
	Print head interface
	8 bit CPU core
	Program ROM
	Flash memory
	Scratchpad SRAM
	Parallel interface (8 bit)
	Motor drive transistors (5)
	Clock PLL
	JTAG test interface
	Test circuits
	Busses
	Bond pads

[0070] The CPU, DRAM, Image sensor, ROM, Flash memory, Parallel interface, JTAG interface and ADC can be vendor supplied cores. The ICP is intended to run on 1.5V to minimize power consumption and allow convenient operation from two AA type battery cells.

[0071] FIG. 15 illustrates a layout of the ICP 48. The ICP 48 is dominated by the imaging array 201, which consumes around 80% of the integrated circuit area. The imaging array is a CMOS 4 transistor active pixel design with a resolution of 1,500×1,000. The array can be divided into the conventional configuration, with two green pixels, one red pixel, and one blue pixel in each pixel group. There are 750×500 pixel groups in the imaging array.

[0072] The latest advances in the field of image sensing and CMOS image sensing in particular can be found in the October, 1997 issue of IEEE Transactions on Electron Devices and, in particular, pages 1689 to 1968. Further, a specific implementation similar to that disclosed in the present application is disclosed in Wong et. al, “CMOS Active Pixel Image Sensors Fabricated Using a 1.8V, 0.25 μm CMOS Technology”, IEDM 1996, page 915.

[0073] The imaging array uses a 4 transistor active pixel design of a standard configuration. To minimize integrated circuit area and therefore cost, the image sensor pixels should be as small as feasible with the technology available. With a four transistor cell, the typical pixel size scales as 20 times the lithographic feature size. This allows a minimum pixel area of around 3.6 μm×3.6 μm. However, the photosite must be substantially above the diffraction limit of the lens. It is also advantageous to have a square photosite, to maximize the margin over the diffraction limit in both horizontal and vertical directions. In this case, the photosite can be specified as 2.5 μm×2.5 μm. The photosite can be a photogate, pinned photodiode, charge modulation device, or other sensor.

[0074] The four transistors are packed as an ‘L’ shape, rather than a rectangular region, to allow both the pixel and the photosite to be square. This reduces the transistor packing density slightly, increasing pixel size. However, the advantage in avoiding the diffraction limit is greater than the small decrease in packing density. The transistors also have a gate length which is longer than the minimum for the process technology.

[0075] These have been increased from a drawn length of 0.18 micron to a drawn length of 0.36 micron. This is to improve the transistor matching by making the variations in gate length represent a smaller proportion of the total gate length.

[0076] The extra gate length, and the ‘L’ shaped packing, mean that the transistors use more area than the minimum for the technology. Normally, around 8 μm² would be required for rectangular packing. Preferably, 9.75 μm² has been allowed for the transistors.

[0077] The total area for each pixel is 16 μm², resulting from a pixel size of 4 μm×4 μm. With a resolution of 1,500×1,000, the area of the imaging array 101 is 6,000 μm×4,000 μm, or 24 mm².

[0078] The presence of a color image sensor on the integrated circuit affects the process required in two major ways:

• • [0079] The CMOS fabrication process should be optimized to minimize dark current Color filters are required. These can be fabricated using dyed photosensitive polyimides, resulting in an added process complexity of three spin coatings, three photolithographic steps, three development steps, and three hardbakes.

[0080] There are 15,000 analog signal processors (ASPs) 205, one for each of the columns of the sensor. The ASPs amplify the signal, provide a dark current reference, sample and hold the signal, and suppress the fixed pattern noise (FPN).

[0081] There are 375 analog to digital converters 206, one for each four columns of the sensor array. These may be delta-sigma or successive approximation type ADC's. A row of low column ADC's are used to reduce the conversion speed required, and the amount of analog signal degradation incurred before the signal is converted to digital. This also eliminates the hot spot (affecting local dark current) and the substrate coupled noise that would occur if a single high speed ADC was used. Each ADC also has two four bit DAC's which trim the offset and scale of the ADC to further reduce FPN variations between columns. These DAC's are controlled by data stored in flash memory during integrated circuit testing.

[0082] The column select logic 204 is a 1:1500 decoder which enables the appropriate digital output of the ADCs onto the output bus. As each ADC is shared by four columns, the least significant two bits of the row select control 4 input analog multiplexors.

[0083] A row decoder 207 is a 1:1000 decoder which enables the appropriate row of the active pixel sensor array. This selects which of the 1000 rows of the imaging array is connected to analog signal processors. As the rows are always accessed in sequence, the row select logic can be implemented as a shift register.

[0084] An auto exposure system 208 adjusts the reference voltage of the ADC 205 in response to the maximum intensity sensed during the previous frame period. Data from the green pixels is passed through a digital peak detector. The peak value of the image frame period before capture (the reference frame) is provided to a digital to analogue converter(DAC), which generates the global reference voltage for the column ADCs. The peak detector is reset at the beginning of the reference frame. The minimum and maximum values of the three RGB color components are also collected for color correction.

[0085] The second largest section of the integrated circuit is consumed by a DRAM 210 used to hold the image. To store the 1,500×1,000 image from the sensor without compression, 1.5 Mbytes of DRAM 210 are required. This equals 12 Mbits, or slightly less than 5% of a 256 Mbit DRAM. The DRAM technology assumed is of the 256 Mbit generation implemented using 0.18 μm CMOS.

[0086] Using a standard 8F cell, the area taken by the memory array is 3.11 mm². When row decoders, column sensors, redundancy, and other factors are taken into account, the DRAM requires around 4 mm².

[0087] This DRAM 210 can be mostly eliminated if analog storage of the image signal can be accurately maintained in the CMOS imaging array for the two seconds required to print the photo. However, digital storage of the image is preferable as it is maintained without degradation, is insensitive to noise, and allows copies of the photo to be printed considerably later.

[0088] A DRAM address generator 211 provides the write and read addresses to the DRAM 210. Under normal operation, the write address is determined by the order of the data read from the CMOS image sensor 201. This will typically be a simple raster format. However, the data can be read from the sensor 201 in any order, if matching write addresses to the DRAM are generated. The read order from the DRAM 210 will normally simply match the requirements of a color interpolator and the print head. As the cyan, magenta, and yellow rows of the print head are necessarily offset by a few pixels to allow space for nozzle actuators, the colors are not read from the DRAM simultaneously. However, there is plenty of time to read all of the data from the DRAM many times during the printing process. This capability is used to eliminate the need for FIFOs in the print head interface, thereby saving integrated circuit area. All three RGB image components can be read from the DRAM each time color data is required. This allows a color space converter to provide a more sophisticated conversion than a simple linear RGB to CMY conversion.

[0089] Also, to allow two dimensional filtering of the image data without requiring line buffers, data is re-read from the DRAM array.

[0090] The address generator may also implement image effects in certain models of camera. For example, passport photos are generated by a manipulation of the read addresses to the DRAM. Also, image framing effects (where the central image is reduced), image warps, and kaleidoscopic effects can all be generated by manipulating the read addresses of the DRAM.

[0091] While the address generator 211 may be implemented with substantial complexity if effects are built into the standard integrated circuit, the integrated circuit area required for the address generator is small, as it consists only of address counters and a moderate amount of random logic.

[0092] A color interpolator 214 converts the interleaved pattern of red, 2× green, and blue pixels into RGB pixels. It consists of three 8 bit adders and associated registers. The divisions are by either 2 (for green) or 4 (for red and blue) so they can be implemented as fixed shifts in the output connections of the adders.

[0093] A convolver 215 is provided as a sharpening filter which applies a small convolution kernel (5×5) to the red, green, and blue planes of the image. The convolution kernel for the green plane is different from that of the red and blue planes, as green has twice as many samples. The sharpening filter has five functions:

• • [0094] To improve the color interpolation from the linear interpolation provided by the color interpolator, to a close approximation of a sinc interpolation.
  • [0095] To compensate for the image ‘softening’ which occurs during digitization.
  • [0096] To adjust the image sharpness to match average consumer preferences, which are typically for the image to be slightly sharper than reality. As the single use camera is intended as a consumer product, and not a professional photographic products, the processing can match the most popular settings, rather than the most accurate.
  • [0097] To suppress the sharpening of high frequency (individual pixel) noise. The function is similar to the ‘unsharp mask’ process.
  • [0098] To antialias Image Warping.

[0099] These functions are all combined into a single convolution matrix. As the pixel rate is low (less than 1 Mpixel per second) the total number of multiplies required for the three color channels is 56 million multiplies per second. This can be provided by a single multiplier. Fifty bytes of coefficient ROM are also required.

[0100] A color ALU 113 combines the functions of color compensation and color space conversion into the one matrix multiplication, which is applied to every pixel of the frame. As with sharpening, the color correction should match the most popular settings, rather than the most accurate.

[0101] A color compensation circuit of the color ALU provides compensation for the lighting of the photo. The vast majority of photographs are substantially improved by a simple color compensation, which independently normalizes the contrast and brightness of the three color components.

[0102] A color look-up table (CLUT) 212 is provided for each color component. These are three separate 256×8 SRAMs, requiring a total of 6,144 bits. The CLUTs are used as part of the color correction process. They are also used for color special effects, such as stochastically selected “wild color” effects.

[0103] A color space conversion system of the color ALU converts from the RGB color space of the image sensor to the CMY color space of the printer. The simplest conversion is a 1's complement of the RGB data. However, this simple conversion assumes perfect linearity of both color spaces, and perfect dye spectra for both the color filters of the image sensor, and the ink dyes. At the other extreme is a tri-linear interpolation of a sampled three dimensional arbitrary transform table. This can effectively match any non-linearity or differences in either color space. Such a system is usually necessary to obtain good color space conversion when the print engine is a color electrophotographic.

[0104] However, since the non-linearity of a halftoned ink jet output is very small, a simpler system can be used. A simple matrix multiply can provide excellent results. This requires nine multiplies and six additions per contone pixel. However, since the contone pixel rate is low (less than 1 Mpixel/sec) these operations can share a single multiplier and adder. The multiplier and adder are used in a color ALU which is shared with the color compensation function.

[0105] Digital halftoning can be performed as a dispersed dot ordered dither using a stochastic optimized dither cell. A halftone matrix ROM 216 is provided for storing dither cell coefficients. A dither cell size of 32×32 is adequate to ensure that the cell repeat cycle is not visible. The three colors—cyan, magenta, and yellow—are all dithered using the same cell, to ensure maximum co-positioning of the ink dots. This minimizes ‘muddying’ of the mid-tones which results from bleed of dyes from one dot to adjacent dots while still wet. The total ROM size required is 1 KByte, as the one ROM is shared by the halftoning units for each of the three colors.

[0106] The digital halftoning used is dispersed dot ordered dither with stochastic optimized dither matrix. While dithering does not produce an image quite as ‘sharp’ as error diffusion, it does produce a more accurate image with fewer artifacts. The image sharpening produced by error diffusion is artificial, and less controllable and accurate than ‘unsharp mask’ filtering performed in the contone domain. The high print resolution (1,600 dpi×1,600 dpi) results in excellent quality when using a well formed stochastic dither matrix.

[0107] Digital halftoning is performed by a digital halftoning unit 217 using a simple comparison between the contone information from the DRAM 210 and the contents of the dither matrix 216. During the halftone process, the resolution of the image is changed from the 250 dpi of the captured contone image to the 1,600 dpi of the printed image. Each contone pixel is converted to an average of 40.96 halftone dots.

[0108] The ICP incorporates a 16 bit microcontroller CPU core 219 to run the miscellaneous camera functions, such as reading the buttons, controlling the motor and solenoids, setting up the hardware, and authenticating the refill station. The processing power required by the CPU is very modest, and a wide variety of processor cores can be used. As the entire CPU program is run from a small ROM 220[.], program compatibility between camera versions is not important, as no external programs are run. A 2 Mbit (256 Kbyte) program and data ROM 220 is included on integrated circuit. Most of this ROM space is allocated to data for outline graphics and fonts for specialty cameras. The program requirements are minor. The single most complex task is the encrypted authentication of the refill station. The ROM requires a single transistor per bit.

[0109] A Flash memory 221 may be used to store a 128 bit authentication code. This provides higher security than storage of the authentication code in ROM, as reverse engineering can be made essentially impossible. The Flash memory is completely covered by third level metal, making the data impossible to extract using scanning probe microscopes or electron beams. The authentication code is stored in the integrated circuit when manufactured. At least two other Flash bits are required for the authentication process: a bit which locks out reprogramming of the authentication code, and a bit which indicates that the camera has been refilled by an authenticated refill station. The flash memory can also be used to store FPN correction data for the imaging array. Additionally, a phase locked loop rescaling parameter is stored for scaling the clocking cycle to an appropriate correct time. The clock frequency does not require crystal accuracy since no date functions are provided. To eliminate the cost of a crystal, an on integrated circuit oscillator with a phase locked loop 224 is used. As the frequency of an on-integrated circuit oscillator is highly variable from integrated circuit to integrated circuit, the frequency ratio of the oscillator to the PLL is digitally trimmed during initial testing. The value is stored in Flash memory 221; This allows the clock PLL to control the ink-jet heater pulse width with sufficient accuracy.

[0110] A scratchpad SRAM is a small static RAM 222 with a 6T cell. The scratchpad provided temporary memory for the 16 bit CPU. 1024 bytes is adequate.

[0111] A print head interface 223 formats the data correctly for the print head. The print head interface also provides all of the timing signals required by the print head. These timing signals may vary depending upon temperature, the number of dots printed simultaneously, the print medium in the print roll, and the dye density of the ink in the print roll.

[0112] The following is a table of external connections to the print head interface:

[0113]

Connection	Function	Pins
DataBits[0–7]	Independent serial data to the eight segments of the print	8
	head
BitClock	Main data clock for the print head	1
ColorEnable[0–2]	Independent enable signals for the CMY actuators,	3
	allowing different pulse times for each color.
BankEnable[0–1]	Allows either simultaneous or interleaved actuation of two	2
	banks of nozzles. This allows two different print
	speed/power consumption tradeoffs
NozzleSelect[0–4]	Selects one of 32 banks of nozzles for simultaneous	5
	actuation
ParallelXferClock	Loads the parallel transfer register with the data from the	1
	shift registers
Total		20

[0114] The print head utilized is composed of eight identical segments, each 1.25 cm long. There is no connection between the segments on the print head integrated circuit. Any connections required are made in the external TAB bonding film, which is double sided. The division into eight identical segments is to simplify lithography using wafer steppers. The segment width of 1.25 cm fits easily into a stepper field. As the print head integrated circuit is long and narrow (10 cm×0.3 mm), the stepper field contains a single segment of 32 print head integrated circuits. The stepper field is therefore 1.25 cm×1.6 cm. An average of four complete print heads are patterned in each wafer step.

[0115] A single BitClock output line connects to all 8 segments on the print head. The 8 DataBits lines lead one to each segment, and are clocked into the 8 segments on the print head simultaneously (on a BitClock pulse). For example, dot 0 is transferred to segment₀, dot 750 is transferred to segment₁, dot 1500 to segment₂ etc simultaneously.

[0116] The ParallelXferClock is connected to each of the 8 segments on the print head, so that on a single pulse, all segments transfer their bits at the same time.

[0117] The NozzleSelect, BankEnable and ColorEnable lines are connected to each of the 8 segments, allowing the print head interface to independently control the duration of the cyan, magenta, and yellow nozzle energizing pulses. Registers in the Print Head Interface allow the accurate specification of the pulse duration between 0 and 6 ms, with a typical duration of 2 ms to 3 ms.

[0118] A parallel interface 125 connects the ICP to individual static electrical signals. The CPU is able to control each of these connections as memory mapped I/O via a low speed bus.

[0119] The following is a table of connections to the parallel interface:

[0120]

	Connection	Direction	Pins
	Paper transport stepper motor	Output	4
	Capping solenoid	Output	1
	Copy LED	Output	1
	Photo button	Input	1
	Copy button	Input	1
	Total		8

[0121] Seven high current drive transistors eg. 227 are required. Four are for the four phases of the main stepper motor, two are for the guillotine motor, and the remaining transistor is to drive the capping solenoid. These transistors are allocated 20,000 square microns (600,000 F) each. As the transistors are driving highly inductive loads, they must either be turned off slowly, or be provided with a high level of back EMF protection. If adequate back EMF protection cannot be provided using the integrated circuit process chosen, then external discrete transistors should be used. The transistors are never driven at the same time as the image sensor is used. This is to avoid voltage fluctuations and hot spots affecting the image, quality. Further, the transistors are located as far away from the sensor as possible.

[0122] A standard JTAG (Joint Test Action Group) interface 228 is included in the ICP for testing purposes and for interrogation by the refill station. Due to the complexity of the integrated circuit, a variety of testing techniques are required, including BIST (Built In Self Test) and functional block isolation. An overhead of 10% in integrated circuit area is assumed for integrated circuit testing circuitry for the random logic portions. The overhead for the large arrays the image sensor and the DRAM is smaller.

[0123] The JTAG interface is also used for authentication of the refill station. This is included to ensure that the cameras are only refilled with quality paper and ink at a properly constructed refill station, thus preventing inferior quality refills from occurring. The camera must authenticate the refill station, rather than vice versa. The secure protocol is communicated to the refill station during the automated test procedure. Contact is made to four gold plated spots on the ICP/print head TAB by the refill station as the new ink is injected into the print head.

[0124] FIG. 16 illustrates a rear view of the next step in the construction process whilst FIG. 17 illustrates a front view.

[0125] Turning now to FIG. 16, the assembly of the camera system proceeds via first assembling the ink supply mechanism 40. The flex PCB is interconnected with batteries 84 only one of which is shown, which are inserted in the middle portion of a print roll 85 which is wrapped around a plastic former 86. An end cap 89 is provided at the other end of the print roll 85 so as to fasten the print roll and batteries firmly to the ink supply mechanism.

[0126] The solenoid coil is interconnected (not shown) to interconnects 97, 98 (FIG. 8) which include leaf spring ends for interconnection with electrical contacts on the Flex PCB so as to provide for electrical control of the solenoid.

[0127] Turning now to FIGS. 17–19 the next step in the construction process is the insertion of the relevant gear trains into the side of the camera chassis. FIG. 17 illustrates a front view, FIG. 18 illustrates a rear view and FIG. 19 also illustrates a rear view. The first gear train comprising gear wheels 22, 23 is utilised for driving the guillotine blade with the gear wheel 23 engaging the gear wheel 65 of FIG. 8. The second gear train comprising gear wheels 24, 25 and 26 engage one end of the print roller 61 of FIG. 8. As best indicated in FIG. 18, the gear wheels mate with corresponding pins on the surface of the chassis with the gear wheel 26 being snap fitted into corresponding mating hole 27.

[0128] Next, as illustrated in FIG. 20, the assembled platen unit 60 is then inserted between the print roll 85 and aluminium cutting blade 43.

[0129] Turning now to FIG. 21, by way of illumination, there is illustrated the electrically interactive components of the camera system. As noted previously, the components are based around a Flex PCB board and include a TAB film 58 which interconnects the printhead 102 with the image sensor and processing integrated circuit 48. Power is supplied by two AA type batteries 83, 84 and a paper drive stepper motor 16 is provided in addition to a rotary guillotine motor 17.

[0130] An optical element 31 is provided for snapping into a top portion of the chassis 12. The optical element 31 includes portions defining an optical view finder 32, 33 which are slotted into mating portions 35, 36 in view finder channel 37. Also provided in the optical element 31 is a lensing system 38 for magnification of the prints left number in addition to an optical pipe element 39 for piping light from the LED 5 for external display.

[0131] Turning next to FIG. 22, the assembled unit 90 is then inserted into a front outer case 91 which includes button 4 for activation of printouts.

[0132] Turning now to FIG. 23, next, the unit 90 is provided with a snap-on back cover 93 which includes a slot 6 and copy print button 7. A wrapper label containing instructions and advertising (not shown) is then wrapped around the outer surface of the camera system and pinch clamped to the cover by means of clamp strip 96 which can comprise a flexible plastic or rubber strip.

[0133] Subsequently, the preferred embodiment is ready for use as a one time use camera system that provides for instant output images on demand. It will be evident that the preferred embodiment further provides for a refillable camera system. A used camera can be collected and its outer plastic cases removed and recycled. A new paper roll and batteries can be added and the ink cartridge refilled. A series of automatic test routines can then be carried out to ensure that the printer is properly operational. Further, in order to ensure only authorised refills are conducted so as to enhance quality, routines in the on-integrated circuit program ROM can be executed such that the camera authenticates the refilling station using a secure protocol. Upon authentication, the camera can reset an internal paper count and an external case can be fitted on the camera system with a new outer label. Subsequent packing and shipping can then take place.

[0134] It will be further readily evident to those skilled in the art that the program ROM can be modified so as to allow for a variety of digital processing routines. In addition to the digitally enhanced photographs optimised for mainstream consumer preferences, various other models can readily be provided through mere re-programming of the program ROM. For example, a sepia classic old fashion style output can be provided through a remapping of the colour mapping function. A further alternative is to provide for black and white outputs again through a suitable colour remapping algorithm. Minimum colour can also be provided to add a touch of colour to black and white prints to produce the effect that was traditionally used to colourize black and white photos. Further, passport photo output can be provided through suitable address remappings within the address generators. Further, edge filters can be utilised as is known in the field of image processing to produce sketched art styles. Further, classic wedding borders and designs can be placed around an output image in addition to the provision of relevant clip arts. For example, a wedding style camera might be provided. Further, a panoramic mode can be provided so as to output the well known panoramic format of images. Further, a postcard style output can be provided through the printing of postcards including postage on the back of a print roll surface. Further, cliparts can be provided for special events such as Halloween, Christmas etc. Further, kaleidoscopic effects can be provided through address remappings and wild colour effects can be provided through remapping of the colour lookup table. Many other forms of special event cameras can be provided for example, cameras dedicated to the Olympics, movie tie-ins, advertising and other special events.

[0135] The operational mode of the camera can be programmed so that upon the depressing of the take photo a first image is sampled by the sensor array to determine irrelevant parameters. Next a second image is again captured which is utilised for the output. The captured image is then manipulated in accordance with any special requirements before being initially output on the paper roll. The LED light is then activated for a predetermined time during which the DRAM is refreshed so as to retain the image. If the print copy button is depressed during this predetermined time interval, a further copy of the photo is output. After the predetermined time interval where no use of the camera has occurred, the onboard CPU shuts down all power to the camera system until such time as the take button is again activated. In this way, substantial power savings can be realized.

Ink Jet Technologies

[0136] The embodiments of the invention use an ink jet printer type device. Of course many different devices could be used. However presently popular ink jet printing technologies are unlikely to be suitable.

[0137] The most significant problem with thermal inkjet is power consumption. This is approximately 100 times that required for high speed, and stems from the energy-inefficient means of drop ejection. This involves the rapid boiling of water to produce a vapor bubble which expels the ink. Water has a very high heat capacity, and must be superheated in thermal inkjet applications. This leads to an efficiency of around 0.02%, from electricity input to drop momentum (and increased surface area) out.

[0138] The most significant problem with piezoelectric inkjet is size and cost. Piezoelectric crystals have a very small deflection at reasonable drive voltages, and therefore require a large area for each nozzle. Also, each piezoelectric actuator must be connected to its drive circuit on a separate substrate. This is not a significant problem at the current limit of around 300 nozzles per print head, but is a major impediment to the fabrication of pagewide print heads with 19,200 nozzles.

[0139] Ideally, the inkjet technologies used meet the stringent requirements of in-camera digital color printing and other high quality, high speed, low cost printing applications. To meet the requirements of digital photography, new inkjet technologies have been created. The target features include:

[0140] low power (less than 10 Watts)

[0141] high resolution capability (1,600 dpi or more)

[0142] photographic quality output

[0143] low manufacturing cost

[0144] small size (pagewidth times minimum cross section)

[0145] high speed (<2 seconds per page).

[0146] All of these features can be met or exceeded by the inkjet systems described below with differing levels of difficulty. 45 different inkjet technologies have been developed by the Assignee to give a wide range of choices for high volume manufacture. These technologies form part of separate applications assigned to the present Assignee as set out in the table below.

[0147] The inkjet designs shown here are suitable for a wide range of digital printing systems, from battery powered one-time use digital cameras, through to desktop and network printers, and through to commercial printing systems.

[0148] For ease of manufacture using standard process equipment, the print head is designed to be a monolithic 0.5 micron CMOS integrated circuit with MEMS post processing. For color photographic applications, the print head is 100 mm long, with a width which depends upon the inklet type. The smallest print head designed is IJ38, which is 0.35 mm wide, giving a integrated circuit area of 35 square mm. The print heads each contain 19,200 nozzles plus data and control circuitry.

[0149] Ink is supplied to the back of the print head by injection molded plastic ink channels. The molding requires 50 micron features, which can be created using a lithographically micromachined insert in a standard injection molding tool. Ink flows through holes etched through the wafer to the nozzle chambers fabricated on the front surface of the wafer. The print head is connected to the camera circuitry by tape automated bonding.

Cross-Referenced Applications

[0150] The following table is a guide to cross-referenced patent applications filed concurrently herewith and discussed hereinafter with the reference being utilized in subsequent tables when referring to a particular case:

[0151]

Docket
No.	Reference	Title
IJ01US	IJ01	Radiant Plunger Ink Jet Printer
IJ02US	IJ02	Electrostatic Ink Jet Printer
IJ03US	IJ03	Planar Thermoelastic Bend Actuator Ink Jet
IJ04US	IJ04	Stacked Electrostatic Ink Jet Printer
IJ05US	IJ05	Reverse Spring Lever Ink Jet Printer
IJ06US	IJ06	Paddle Type Ink Jet Printer
IJ07US	IJ07	Permanent Magnet Electromagnetic Ink Jet Printer
IJ08US	IJ08	Planar Swing Grill Electromagnetic Ink Jet Printer
IJ09US	IJ09	Pump Action Refill Ink Jet Printer
IJ10US	IJ10	Pulsed Magnetic Field Ink Jet Printer
IJ11US	IJ11	Two Plate Reverse Firing Electromagnetic Ink Jet Printer
IJ12US	IJ12	Linear Stepper Actuator Ink Jet Printer
IJ13US	IJ13	Gear Driven Shutter Ink Jet Printer
IJ14US	IJ14	Tapered Magnetic Pole Electromagnetic Ink Jet Printer
IJ15US	IJ15	Linear Spring Electromagnetic Grill Ink Jet Printer
IJ16US	IJ16	Lorenz Diaphragm Electromagnetic Ink Jet Printer
IJ17US	IJ17	PTFE Surface Shooting Shuttered Oscillating Pressure Ink Jet Printer
IJ18US	IJ18	Buckle Grip Oscillating Pressure Ink Jet Printer
IJ19US	IJ19	Shutter Based Ink Jet Printer
IJ20US	IJ20	Curling Calyx Thermoelastic Ink Jet Printer
IJ21US	IJ21	Thermal Actuated Ink Jet Printer
IJ22US	IJ22	Iris Motion Ink Jet Printer
IJ23US	IJ23	Direct Firing Thermal Bend Actuator Ink Jet Printer
IJ24US	IJ24	Conductive PTFE Ben Activator Vented Ink Jet Printer
IJ25US	IJ25	Magnetostrictive Ink Jet Printer
IJ26US	IJ26	Shape Memory Alloy Ink Jet Printer
IJ27US	IJ27	Buckle Plate Ink Jet Printer
IJ28US	IJ28	Thermal Elastic Rotary Impeller Ink Jet Printer
IJ29US	IJ29	Thermoelastic Bend Actuator Ink Jet Printer
IJ30US	IJ30	Thermoelastic Bend Actuator Using PTFE and Corrugated Copper Ink Jet
		Printer
IJ31US	IJ31	Bend Actuator Direct Ink Supply Ink Jet Printer
IJ32US	IJ32	A High Young's Modulus Thermoelastic Ink Jet Printer
IJ33US	IJ33	Thermally actuated slotted chamber wall ink jet printer
IJ34US	IJ34	Ink Jet Printer having a thermal actuator comprising an external coiled
		spring
IJ35US	IJ35	Trough Container Ink Jet Printer
IJ36US	IJ36	Dual Chamber Single Vertical Actuator Ink Jet
IJ37US	IJ37	Dual Nozzle Single Horizontal Fulcrum Actuator Ink Jet
IJ38US	IJ38	Dual Nozzle Single Horizontal Actuator Ink Jet
IJ39US	IJ39	A single bend actuator cupped paddle ink jet printing device
IJ40US	IJ40	A thermally actuated ink jet printer having a series of thermal actuator units
IJ41US	IJ41	A thermally actuated ink jet printer including a tapered heater element
IJ42US	IJ42	Radial Back-Curling Thermoelastic Ink Jet
IJ43US	IJ43	Inverted Radial Back-Curling Thermoelastic Ink Jet
IJ44US	IJ44	Surface bend actuator vented ink supply ink jet printer
IJ45US	IJ45	Coil Acutuated Magnetic Plate Ink Jet Printer

Tables of Drop-on-Demand Inkjets

[0152] Eleven important characteristics of the fundamental operation of individual inkjet nozzles have been identified. These characteristics are largely orthogonal, and so can be elucidated as an eleven dimensional matrix. Most of the eleven axes of this matrix include entries developed by the present assignee.

[0153] The following tables form the axes of an eleven dimensional table of inkjet types.

[0154] Actuator mechanism (18 types)

[0155] Basic operation mode (7 types)

[0156] Auxiliary mechanism (8 types)

[0157] Actuator amplification or modification method (17 types)

[0158] Actuator motion (19 types)

[0159] Nozzle refill method (4 types)

[0160] Method of restricting back-flow through inlet (10 types)

[0161] Nozzle clearing method (9 types)

[0162] Nozzle plate construction (9 types)

[0163] Drop ejection direction (5 types)

[0164] Ink type (7 types)

[0165] The complete eleven dimensional table represented by these axes contains 36.9 billion possible configurations of inkjet nozzle. While not all of the possible combinations result in a viable inkjet technology, many million configurations are viable. It is clearly impractical to elucidate all of the possible configurations. Instead, certain inkjet types have been investigated in detail. These are designated IJ01 to IJ45 above.

[0166] Other inkjet configurations can readily be derived from these 45 examples by substituting alternative configurations along one or more of the 1 axes. Most of the IJ01 to IJ45 examples can be made into inkjet print heads with characteristics superior to any currently available inkjet technology.

[0167] Where there are prior art examples known to the inventor, one or more of these examples are listed in the examples column of the tables below. The IJ01 to IJ45 series are also listed in the examples column. In some cases, a printer may be listed more than once in a table, where it shares characteristics with more than one entry.

[0168] Suitable applications include: Home printers, Office network printers, Short run digital printers, Commercial print systems, Fabric printers, Pocket printers, Internet WWW printers, Video printers, Medical imaging, Wide format printers, Notebook PC printers, Fax machines, Industrial printing systems, Photocopiers, Photographic minilabs etc.

[0169] The information associated with the aforementioned 11 dimensional matrix are set out in the following tables.

[0170]

ACTUATOR MECHANISM (APPLIED ONLY TO SELECTED INK DROPS)
Actuator Mechanism	Description	Advantages	Disadvantages	Examples
Thermal bubble	An electrothermal heater heats the	1)	Large force generated	6)	High power	16)	Canon
	ink to above boiling point,	2)	Simple construction	7)	Ink carrier limited to water		Bubblejet 1979 Endo
	transferring significant heat to the	3)	No moving parts	8)	Low efficiency		et al GB patent
	aqueous ink. A bubble nucleates and	4)	Fast operation	9)	High temperatures required		2,007,162
	quickly forms, expelling the ink.	5)	Small integrated circuit area required for	10)	High mechanical stress	17)	Xerox heater-in-
	The efficiency of the process is low,		actuator	11)	Unusual materials required		pit 1990 Hawkins et
	with typically less than 0.05% of the			12)	Large drive transistors		al U.S. Pat. No. 4,899,181
	electrical energy being transformed			13)	Cavitation causes actuator failure	18)	Hewlett-
	into kinetic energy of the drop.			14)	Kogation reduces bubble		Packard TIJ 1982
					formation		Vaught et al U.S. Pat. No.
				15)	Large print heads are difficult to		4,490,728
					fabricate
Piezoelectric	A piezoelectric crystal such as lead	19)	Low power	23)	Very large area required for	28)	Kyser et al U.S. Pat. No.
	lanthanum zirconate (PZT) is		consumption		actuator		3,946,398
	electrically activated, and either	20)	Many ink types can be	24)	Difficult to integrate with	29)	Zoltan U.S. Pat. No.
	expands, shears, or bends to apply		used		electronics		3,683,212
	pressure to the ink, ejecting drops.	21)	Fast operation	25)	High voltage drive transistors	30)	1973 Stemme
		22)	High efficiency		required		U.S. Pat. No. 3,747,120
				26)	Full pagewidth print heads	31)	Epson Stylus
					impractical due to actuator size	32)	Tektronix
				27)	Requires electrical poling in high	33)	IJ04
					field strengths during manufacture
Electro-strictive	An electric field is used to activate	34)	Low power	39)	Low maximum strain (approx.	44)	Seiko Epson,
	electrostriction in relaxor materials		consumption		0.01%)		Usui et all JP
	such as lead lanthanum zirconate	35)	Many ink types can be	40)	Large area required for actuator		253401/96
	titanate (PLZT) or lead magnesium		used		due to low strain	45)	IJ04
	niobate (PMN).	36)	Low thermal	41)	Response speed is marginal (~10 μs)
			expansion	42)	High voltage drive transistors
		37)	Electric field strength		required
			required (approx. 3.5 V/μm)	43)	Full pagewidth print heads
			can be generated without		impractical due to actuator size
			difficulty
		38)	Does not require
			electrical poling
Ferroelectric	An electric field is used to induce a	46)	Low power	52)	Difficult to integrate with	55)	IJ04
	phase transition between the		consumption		electronics
	antiferroelectric (AFE) and	47)	Many ink types can be	53)	Unusual materials such as
	ferroelectric (FE) phase. Perovskite		used		PLZSnT are required
	materials such as tin modified lead	48)	Fast operation (<1 μs)	54)	Actuators require a large area
	lanthanum zirconate titanate	49)	Relatively high
	(PLZSnT) exhibit large strains of up		longitudinal strain
	to 1% associated with the AFE to FE	50)	High efficiency
	phase transition.	51)	Electric field strength
			of around 3 V/μm can be
			readily provided
Electrostatic	Conductive plates are separated by a	56)	Low power	59)	Difficult to operate electrostatic	64)	IJ02, IJ04
plates	compressible or fluid dielectric		consumption		devices in an aqueous environment
	(usually air). Upon application of a	57)	Many ink types can be	60)	The electrostatic actuator will
	voltage, the plates attract each other		used		normally need to be separated from the
	and displace ink, causing drop	58)	Fast operation		ink
	ejection. The conductive plates may			61)	Very large area required to
	be in a comb or honeycomb				achieve high forces
	structure, or stacked to increase the			62)	High voltage drive transistors
	surface area and therefore the force.				may be required
				63)	Full pagewidth print heads are not
					competitive due to actuator size
Electrostatic	A strong electric field is applied to	65)	Low current	67)	High voltage required	72)	1989 Saito et al,
pull on ink	the ink, whereupon electrostatic		consumption	68)	May be damaged by sparks due to		U.S. Pat. No. 4,799,068
	attraction accelerates the ink towards	66)	Low temperature		air breakdown	73)	1989 Miura et
	the print medium.			69)	Required field strength increases		al, U.S. Pat. No. 4,810,954
					as the drop size decreases	74)	Tone-jet
				70)	High voltage drive transistors
					required
				71)	Electrostatic field attracts dust
Permanent	An electromagnet directly attracts a	75)	Low power	80)	Complex fabrication	86)	IJ07, IJ10
magnet electromagnetic	permanent magnet, displacing ink		consumption	81)	Permanent magnetic material
	and causing drop ejection. Rare earth	76)	Many ink types can be		such as Neodymium Iron Boron
	magnets with a field strength around		used		(NdFeB) required.
	1 Tesla can be used. Examples are:	77)	Fast operation	82)	High local currents required
	Samarium Cobalt (SaCo) and	78)	High efficiency	83)	Copper metalization should be
	magnetic materials in the neodymium	79)	Easy extension from		used for long electromigration lifetime
	iron boron family (NdFeB,		single nozzles to pagewidth		and low resistivity
	NdDyFeBNb, NdDyFeB, etc)		print heads	84)	Pigmented inks are usually
					infeasible
				85)	Operating temperature limited to
					the Curie temperature (around 540 K)
Soft magnetic	A solenoid induced a magnetic field	87)	Low power	92)	Complex fabrication	98)	IJ01, IJ05, IJ08,
core electromagnetic	in a soft magnetic core or yoke		consumption	93)	Materials not usually present in a		IJ10
	fabricated from a ferrous material	88)	Many ink types can be		CMOS fab such as NiFe, CoNiFe, or	99)	IJ12, IJ14, IJ15,
	such as electroplated iron alloys such		used		CoFe are required		IJ17
	as CoNiFe [1], CoFe, or NiFe alloys.	89)	Fast operation	94)	High local currents required
	Typically, the soft magnetic material	90)	High efficiency	95)	Copper metalization should be
	is in two parts, which are normally	91)	Easy extension from		used for long electromigration lifetime
	held apart by a spring. When the		single nozzles to pagewidth		and low resistivity
	solenoid is actuated, the two parts		print heads	96)	Electroplating is required
	attract, displacing the ink.			97)	High saturation flux density is
					required (2.0–2.1 T is achievable with
					CoNiFe [1])
Magnetic	The Lorenz force acting on a current	100)	Low power	105)	Force acts as a twisting motion	110)	IJ06, IJ11, IJ13,
Lorenz force	carrying wire in a magnetic field is		consumption	106)	Typically, only a quarter of the		IJ16
	utilized.	101)	Many ink types can be		solenoid length provides force in a
	This allows the magnetic field to be		used		useful direction
	supplied externally to the print head,	102)	Fast operation	107)	High local currents required
	for example with rare earth	103)	High efficiency	108)	Copper metalization should be
	permanent magnets.	104)	Easy extension from		used for long electromigration lifetime
	Only the current carrying wire need		single nozzles to pagewidth		and low resistivity
	be fabricated on the print-head,		print heads	109)	Pigmented inks are usually
	simplifying materials requirements.				infeasible
Magnetostriction	The actuator uses the giant	111)	Many ink types can be	115)	Force acts as a twisting motion	120)	Fischenbeck,
	magnetostrictive effect of materials		used	116)	Unusual materials such as		U.S. Pat. No. 4,032,929
	such as Terfenol-D (an alloy of	112)	Fast operation		Terfenol-D are required	121)	IJ25
	terbium, dysprosium and iron	113)	Easy extension from	117)	High local currents required
	developed at the Naval Ordnance		single nozzles to pagewidth	118)	Copper metalization should be
	Laboratory, hence Ter-Fe-NOL). For		print heads		used for long electromigration lifetime
	best efficiency, the actuator should	114)	High force is available		and low resistivity
	be pre-stressed to approx. 8 MPa.			119)	Pre-stressing may be required
Surface tension	Ink under positive pressure is held in	122)	Low power	127)	Requires supplementary force to	130)	Silverbrook, EP
reduction	a nozzle by surface tension. The		consumption		effect drop separation		0771 658 A2 and
	surface tension of the ink is reduced	123)	Simple construction	128)	Requires special ink surfactants		related patent
	below the bubble threshold, causing	124)	No unusual materials	129)	Speed may be limited by		applications
	the ink to egress from the nozzle.		required in fabrication		surfactant properties
		125)	High efficiency
		126)	Easy extension from
			single nozzles to pagewidth
			print heads
Viscosity	The ink viscosity is locally reduced	131)	Simple construction	134)	Requires supplementary force to	139)	Silverbrook, EP
reduction	to select which drops are to be	132)	No unusual materials		effect drop separation		0771 658 A2 and
	ejected. A viscosity reduction can be		required in fabrication	135)	Requires special ink viscosity		related patent
	achieved electrothermally with most	133)	Easy extension from		properties		applications
	inks, but special inks can be		single nozzles to pagewidth	136)	High speed is difficult to achieve
	engineered for a 100:1 viscosity		print heads	137)	Requires oscillating ink pressure
	reduction.			138)	A high temperature difference
					(typically 80 degrees) is required
Acoustic	An acoustic wave is generated and	140)	Can operate without a	141)	Complex drive circuitry	146)	1993
	focussed upon the drop ejection		nozzle plate	142)	Complex fabrication		Hadimioglu et al,
	region.			143)	Low efficiency		EUP 550,192
				144)	Poor control of drop position	147)	1993 Elrod et al,
				145)	Poor control of drop volume		EUP 572,220
Thermoelastic	An actuator which relies upon	148)	Low power	157)	Efficient aqueous operation	160)	IJ03, IJ09, IJ17,
bend actuator	differential thermal expansion upon		consumption		requires a thermal insulator on the hot		IJ18
	Joule heating is used.	149)	Many ink types can be		side	161)	IJ19, IJ20, IJ21,
			used	158)	Corrosion prevention can be		IJ22
		150)	Simple planar		difficult	162)	IJ23, IJ24, IJ27,
			fabrication	159)	Pigmented inks may be infeasible,		IJ28
		151)	Small integrated		as pigment particles may be jam the bend	163)	IJ29, IJ30, IJ31,
			circuit area required for		actuator		IJ32
			each actuator			164)	IJ33, IJ34, IJ35,
		152)	Fast operation				IJ36
		153)	High efficiency			165)	IJ37, IJ38, IJ39,
		154)	CMOS compatible				IJ40
			voltages and currents			166)	IJ41
		155)	Standard MEMS
			processes can be used
		156)	Easy extension from
			single nozzles to pagewidth
			print heads
High CTE	A material with a very high	167)	High force can be	177)	Requires special material (e.g.	181)	IJ09, IJ17, IJ18,
thermoelastic	coefficient of thermal expansion		generated		PTFE)		IJ20
actuator	(CTE) such as	168)	PTFE is a candidate	178)	Requires a PTFE deposition	182)	IJ21, IJ22, IJ23,
	polytetrafluoroethylene (PTFE) is		for low dielectric constant		process, which is not yet standard in		IJ24
	used. As high CTE materials are		insulation in ULSI		ULSI fabs	183)	IJ27, IJ28, IJ29,
	usually non-conductive, a heater	169)	Very low power	179)	PTFE deposition cannot be		IJ30
	fabricated from a conductive material		consumption		followed with high temperature (above	184)	IJ31, IJ42, IJ43,
	is incorporated. A 50 μm long PTFE	170)	Many ink types can be		350° C.) processing		IJ44
	bend actuator with polysilicon heater		used	180)	Pigmented inks may be infeasible,
	and 15 mW power input can provide	171)	Simple planar		as pigment particles may jam the bend
	180 μN force and 10 μm deflection.		fabrication		actuator
	Actuator motions include:	172)	Small integrated
	Bend		circuit area required for
	Push		each actuator
	Buckle	173)	Fast operation
	Rotate	174)	High efficiency
		175)	CMOS compatible
			voltages and currents
		176)	Easy extension from
			single nozzles to pagewidth
			print heads
Conductive	A polymer with a high coefficient of	185)	High force can be	194)	Requires special materials	199)	IJ24
polymer	thermal expansion (such as PTFE) is		generated		development (High CTE conductive
thermoelastic	doped with conducting substances to	186)	Very low power		polymer)
actuator	increase its conductivity to about 3		consumption	195)	Requires a PTFE deposition
	orders of magnitude below that of	187)	Many ink types can be		process, which is not yet standard in
	copper. The conducting polymer		used		ULSI fabs
	expands when resistively heated.	188)	Simple planar	196)	PTFE deposition cannot be
	Examples of conducting dopants		fabrication		followed with high temperature (above
	include:	189)	Small integrated		350° C.) processing
	Carbon nanotubes		circuit area required for	197)	Evaporation and CVD deposition
	Metal fibers		each actuator		techniques cannot be used
	Conductive polymers such as doped	190)	Fast operation	198)	Pigmented inks may be infeasible,
	polythiophene	191)	High efficiency		as pigment particles may jam the bend
	Carbon granules	192)	CMOS compatible		actuator
			voltages and currents
		193)	Easy extension from
			single nozzles to pagewidth
			print heads
Shape memory	A shape memory alloy such as TiNi	200)	High force is available	206)	Fatigue limits maximum number	213)	IJ26
alloy	(also known as Nitinol - Nickel		(stresses of hundreds of		of cycles
	Titanium alloy developed at the		MPa)	207)	Low strain (1%) is required to
	Naval Ordnance Laboratory) is	201)	Large strain is		extend fatigue resistance
	thermally switched between its weak		available (more than 3%)	208)	Cycle rate limited by heat
	martensitic state and its high stiffness	202)	High corrosion		removal
	austenic state. The shape of the		resistance	209)	Requires unusual materials (TiNi)
	actuator in its martensitic state is	203)	Simple construction	210)	The latent heat of transformation
	deformed relative to the austenic	204)	Easy extension from		must be provided
	shape. The shape change causes		single nozzles to pagewidth	211)	High current operation
	ejection of a drop.		print heads	212)	Requires pre-stressing to distort
		205)	Low voltage operation		the martensitic state
Linear Magnetic	Linear magnetic actuators include the	214)	Linear Magnetic	218)	Requires unusual semiconductor	222)	IJ12
Actuator	Linear Induction Actuator (LIA),		actuators can be constructed		materials such as soft magnetic alloys
	Linear Permanent Magnet		with high thrust, long travel,		(e.g. CoNiFe [1])
	Synchronous Actuator (LPMSA),		and high efficiency using	219)	Some varieties also require
	Linear Reluctance Synchronous		planar semiconductor		permanent magnetic materials such as
	Actuator (LRSA), Linear Switched		fabrication techniques		Neodymium iron boron (NdFeB)
	Reluctance Actuator (LSRA), and the	215)	Long actuator travel is	220)	Requires complex multi-phase
	Linear Stepper Actuator (LSA).		available		drive circuitry
		216)	Medium force is	221)	High current operation
			available
		217)	Low voltage operation

[0171]

BASIC OPERATION MODE
Operational mode	Description	Advantages	Disadvantages	Examples
Actuator	This is the simplest mode of	223)	Simple operation	227)	Drop repetition rate is usually	230)	Thermal inkjet
directly pushes ink	operation: the actuator directly	224)	No external fields		limited to less than 10 KHz. However,	231)	Piezoelectric
	supplies sufficient kinetic energy to		required		this is not fundamental to the method,		inkjet
	expel the drop. The drop must have a	225)	Satellite drops can be avoided		but is related to the refill method normally	232)	IJ01, IJ02, IJ03,
	sufficient velocity to overcome the surface		if drop velocity is less than 4 m/s		used		IJ04
	tension.			228)	All of the drop kinetic energy	233)	IJ05, IJ06, IJ07,
		226)	Can be efficient, depending		must be provided by the actuator		IJ09
			upon the actuator used	229)	Satellite drops usually form if	234)	IJ11, IJ12, IJ14,
					drop velocity is greater than 4.5 m/s		IJ16
						235)	IJ20, IJ22, IJ23,
							IJ24
						236)	IJ25, IJ26, IJ27,
							IJ28
						237)	IJ29, IJ30, IJ31,
							IJ32
						238)	IJ33, IJ34, IJ35,
							IJ36
						239)	IJ37, IJ38, IJ39,
							IJ40
						240)	IJ41, IJ42, IJ43,
							IJ44
Proximity	The drops to be printed are selected	241)	Very simple print head	243)	Requires close proximity between	246)	Silverbrook, EP
	by some manner (e.g. thermally		fabrication can be used		the print head and the print media or		0771 658 A2 and
	induced surface tension reduction of	242)	The drop selection		transfer roller		related patent
	pressurized ink). Selected drops are		means does not need to	244)	May require two print heads		applications
	separated from the ink in the nozzle		provide the energy required		printing alternate rows of the image
	by contact with the print medium or a		to separate the drop from	245)	Monolithic color print heads are
	transfer roller.		the nozzle		difficult
Electrostatic pull on ink	The drops to be printed are selected	247)	Very simple print head	249)	Requires very high electrostatic	252)	Silverbrook, EP
	by some manner (e.g. thermally		fabrication can be used		field		0771 658 A2 and
	induced surface tension reduction of	248)	The drop selection	250)	Electrostatic field for small nozzle		related patent
	pressurized ink). Selected drops are		means does not need to		sizes is above air breakdown		applications
	separated from the ink in the nozzle		provide the energy required	251)	Electrostatic field may attract dust	253)	Tone-Jet
	by a strong electric field.		to separate the drop from
			the nozzle
Magnetic pull on	The drops to be printed are selected	254)	Very simple print head	256)	Requires magnetic ink	259)	Silverbrook, EP
ink	by some manner (e.g. thermally		fabrication can be used	257)	Ink colors other than black are		0771 658 A2 and
	induced surface tension reduction of	255)	The drop selection		difficult		related patent applications
	pressurized ink). Selected drops are		means does not need to	258)	Requires very high magnetic
	separated from the ink in the nozzle		provide the energy required		fields
	by a strong magnetic field acting on		to separate the drop from
	the magnetic ink.		the nozzle
Shutter	The actuator moves a shutter to block	260)	High speed (>50 KHz)	263)	Moving parts are required	267)	IJ13, IJ17, IJ21
	ink flow to the nozzle. The ink		operation can be achieved	264)	Requires ink pressure modulator
	pressure is pulsed at a multiple of the		due to reduced refill time	265)	Friction and wear must be
	drop ejection frequency.	261)	Drop timing can be		considered
			very accurate	266)	Stiction is possible
		262)	The actuator energy
			can be very low
Shuttered grill	The actuator moves a shutter to block	268)	Actuators with small	271)	Moving parts are required	275)	IJ08, IJ15, IJ18,
	ink flow through a grill to the nozzle.		travel can be used	272)	Requires ink pressure modulator		IJ19
	The shutter movement need only be	269)	Actuators with small	273)	Friction and wear must be
	equal to the width of the grill holes.		force can be used		considered
		270)	High speed (>50 KHz)	274)	Stiction is possible
			operation can be achieved
Pulsed magnetic	A pulsed magnetic field attracts an	276)	Extremely low energy	278)	Requires an external pulsed	281)	IJ10
pull on ink	‘ink pusher’ at the drop ejection		operation is possible		magnetic field
pusher	frequency. An actuator controls a	277)	No heat dissipation	279)	Requires special materials for
	catch, which prevents the ink pusher		problems		both the actuator and the ink pusher
	from moving when a drop is not to be			280)	Complex construction
	ejected.

[0172]

AUXILIARY MECHANISM (APPLIED TO ALL NOZZLES)
Auxiliary
Mechanism	Description	Advantages	Disadvantages	Examples
None	The actuator directly fires the ink	282)	Simplicity of	285)	Drop ejection energy must be	286)	Most inkjets,
	drop, and there is no external field or other mechanism		construction		supplied by individual nozzle actuator		including piezoelectric and
	required.	283)	Simplicity of operation				thermal bubble.
		284)	Small physical size
						287)	IJ01–IJ07, IJ09,
							IJ11
						288)	IJ12, IJ14, IJ20,
							IJ22
						289)	IJ23–IJ45
Oscillating ink	The ink pressure oscillates, providing	290)	Oscillating ink	293)	Requires external ink pressure	296)	Silverbrook, EP
pressure (including	much of the drop ejection energy.		pressure can provide a refill		oscillator		0771 658 A2 and related patent
acoustic stimulation)	The actuator selects which drops are		pulse, allowing higher operating speed	294)	Ink pressure phase and amplitude		applications
	to be fired by selectively blocking or				must be carefully controlled
	enabling nozzles. The ink pressure	291)	The actuators may	295)	Acoustic reflections in the ink	297)	IJ08, IJ13, IJ15,
	oscillation may be achieved by		operate with much lower		chamber must be designed for		IJ17
	vibrating the print head, or preferably		energy			298)	IJ18, IJ19, IJ21
	by an actuator in the ink supply.	292)	Acoustic lenses can be
			used to focus the sound on
			the nozzles
Media proximity	The print head is placed in close	299)	Low power	302)	Precision assembly required	305)	Silverbrook, EP
	proximity to the print medium.	300)	High accuracy	303)	Paper fibers may cause problems		0771 658 A2 and
	Selected drops protrude from the	301)	Simple print head	304)	Cannot print on rough substrates		related patent
	print head further than unselected		construction				applications
	drops, and contact the print medium.
	The drop soaks into the medium fast
	enough to cause drop separation.
Transfer roller	Drops are printed to a transfer roller	306)	High accuracy	309)	Bulky	312)	Silverbrook, EP
	instead of straight to the print	307)	Wide range of print	310)	Expensive		0771 658 A2 and
	medium. A transfer roller can also be		substrates can be used	311)	Complex construction		related patent
	used for proximity drop separation.	308)	Ink can be dried on the				applications
			transfer roller			313)	Tektronix hot
							melt piezoelectric
							inkjet
						314)	Any of the IJ
							series
Electrostatic	An electric field is used to accelerate	315)	Low power	317)	Field strength required for	318)	Silverbrook, EP
	selected drops towards the print	316)	Simple print head		separation of small drops is near or		0771 658 A2 and
	medium.		construction		above air breakdown		related patent
							applications
						319)	Tone-Jet
Direct magnetic	A magnetic field is used to accelerate	320)	Low power	322)	Requires magnetic ink	324)	Silverbrook, EP
field	selected drops of magnetic ink	321)	Simple print head	323)	Requires strong magnetic field		0771 658 A2 and
	towards the print medium.		construction				related patent
							applications
Cross magnetic	The print head is placed in a constant	325)	Does not require	326)	Requires external magnet	328)	IJ06, IJ16
field	magnetic field. The Lorenz force in a		magnetic materials to be	327)	Current densities may be high,
	current carrying wire is used to move		integrated in the print head		resulting in electromigration problems
	the actuator.		manufacturing process
Pulsed magnetic	A pulsed magnetic field is used to	329)	Very low power	331)	Complex print head construction	333)	IJ10
field	cyclically attract a paddle, which		operation is possible	332)	Magnetic materials required in
	pushes on the ink. A small actuator	330)	Small print head size		print head
	moves a catch, which selectively
	prevents the paddle from moving.

[0173]

ACTUATOR AMPLIFICATION OR MODIFICATION METHOD
Actuator amplification	Description	Advantages	Disadvantages	Examples
None	No actuator mechanical amplification	334)	Operational simplicity	335)	Many actuator mechanisms have	336)	Thermal Bubble
	is used. The actuator directly drives				insufficient travel, or insufficient force,		Inkjet
	the drop ejection process.				to efficiently drive the drop ejection process	337)	IJ01, IJ02, IJ06,
							IJ07
						338)	IJ16, IJ25, IJ26
Differential expansion bend	An actuator material expands more	339)	Provides greater travel	341)	High stresses are involved	344)	Piezoelectric
actuator	on one side than on the other. The		in a reduced print head area	342)	Care must be taken that the	345)	IJ03, IJ09, IJ17–IJ24
	expansion may be thermal,	340)	The bend actuator		materials do not delaminate	346)	IJ27, IJ29–IJ39,
	piezoelectric, magnetostrictive, or other mechanism.		converts a high force low	343)	Residual bend resulting from high		IJ42,
			travel actuator mechanism		temperature or high stress during	347)	IJ43, IJ44
			to high travel, lower force		formation
			mechanism.
Transient bend actuator	A trilayer bend actuator where the	348)	Very good temperature	351)	High stresses are involved	353)	IJ40, IJ41
	two outside layers are identical. This		stability	352)	Care must be taken that the
	cancels bend due to ambient	349)	High speed, as a new		materials do not delaminate
	temperature and residual stress. The		drop can be fired before
	actuator only responds to transient		heat dissipates
	heating of one side or the other.	350)	Cancels residual stress
			of formation
Actuator stack	A series of thin actuators are stacked.	354)	Increased travel	356)	Increased fabrication complexity	358)	Some
	This can be appropriate where	355)	Reduced drive voltage	357)	Increased possibility of short		piezoelectric ink jets
	actuators require high electric field				circuits due to pinholes	359)	IJ04
	strength, such as electrostatic and
	piezoelectric actuators.
Multiple	Multiple smaller actuators are used	360)	Increases the force	362)	Actuator forces may not add	363)	IJ12, IJ13, IJ18,
actuators	simultaneously to move the ink. Each		available from an actuator		linearly, reducing efficiency		IJ20
	actuator need provide only a portion	361)	Multiple actuators can			364)	IJ22, IJ28, IJ42,
	of the force required.		be positioned to control ink				IJ43
			flow accurately
Linear Spring	A linear spring is used to transform a	365)	Matches low travel	367)	Requires print head area for the	368)	IJ15
	motion with small travel and high		actuator with higher travel		spring
	force into a longer travel, lower force		requirements
	motion.	366)	Non-contact method of
			motion transformation
Reverse spring	The actuator loads a spring. When	369)	Better coupling to the	370)	Fabrication complexity	372)	IJ05, IJ11
	the actuator is turned off, the spring		ink	371)	High stress in the spring
	releases. This can reverse the
	force/distance curve of the actuator
	to make it compatible with the
	force/time requirements of the drop
	ejection.
Coiled actuator	A bend actuator is coiled to provide	373)	Increases travel	376)	Generally restricted to planar	377)	IJ17, IJ21, IJ34,
	greater travel in a reduced integrated	374)	Reduces integrated		implementations due to extreme		IJ35
	circuit area.		circuit area		fabrication difficulty in other
		375)	Planar		orientations.
			implementations are
			relatively easy to fabricate.
Flexure bend	A bend actuator has a small region	378)	Simple means of	379)	Care must be taken not to exceed	382)	IJ10, IJ19, IJ33
actuator	near the fixture point, which flexes		increasing travel of a bend		the elastic limit in the flexure area
	much more readily than the		actuator	380)	Stress distribution is very uneven
	remainder of the actuator. The			381)	Difficult to accurately model with
	actuator flexing is effectively				finite element analysis
	converted from an even coiling to an
	angular bend, resulting in greater
	travel of the actuator tip.
Gears	Gears can be used to increase travel	383)	Low force, low travel	385)	Moving parts are required	390)	IJ13
	at the expense of duration. Circular		actuators can be used	386)	Several actuator cycles are
	gears, rack and pinion, ratchets, and	384)	Can be fabricated		required
	other gearing methods can be used.		using standard surface	387)	More complex drive electronics
			MEMS processes	388)	Complex construction
				389)	Friction, friction, and wear are
					possible
Catch	The actuator controls a small catch.	391)	Very low actuator	393)	Complex construction	396)	IJ10
	The catch either enables or disables		energy	394)	Requires external force
	movement of an ink pusher that is	392)	Very small actuator	395)	Unsuitable for pigmented inks
	controlled in a bulk manner.		size
Buckle plate	A buckle plate can be used to change	397)	Very fast movement	398)	Must stay within elastic limits of	401)	S. Hirata et al,
	a slow actuator into a fast motion. It		achievable		the materials for long device life		“An Ink-jet Head . . . ”,
	can also convert a high force, low			399)	High stresses involved		Proc. IEEE MEMS,
	travel actuator into a high travel,			400)	Generally high power		Feb. 1996, pp 418–423.
	medium force motion.				requirement	402)	IJ18, IJ27
Tapered	A tapered magnetic pole can increase	403)	Linearizes the	404)	Complex construction	405)	IJ14
magnetic pole	travel at the expense of force.		magnetic force/distance
			curve
Lever	A lever and fulcrum is used to	406)	Matches low travel	408)	High stress around the fulcrum	409)	IJ32, IJ36, IJ37
	transform a motion with small travel		actuator with higher travel
	and high force into a motion with		requirements
	longer travel and lower force. The	407)	Fulcrum area has no
	lever can also reverse the direction of		linear movement, and can
	travel.		be used for a fluid seal
Rotary impeller	The actuator is connected to a rotary	410)	High mechanical	412)	Complex construction	414)	IJ28
	impeller. A small angular deflection		advantage	413)	Unsuitable for pigmented inks
	of the actuator results in a rotation of	411)	The ratio of force to
	the impeller vanes, which push the		travel of the actuator can
	ink against stationary vanes and out		be matched to the nozzle
	of the nozzle.		requirements by varying the
			number of impeller vanes
Acoustic lens	A refractive or diffractive (e.g. zone	415)	No moving parts	416)	Large area required	418)	1993
	plate) acoustic lens is used to			417)	Only relevant for acoustic ink jets		Hadimioglu et al,
	concentrate sound waves.						EUP 550,192
						419)	1993 Elrod et al,
							EUP 572,220
Sharp	A sharp point is used to concentrate	420)	Simple construction	421)	Difficult to fabricate using	423)	Tone-jet
conductive point	an electrostatic field.				standard VLSI processes for a surface
					ejecting ink-jet
				422)	Only relevant for electrostatic ink
					jets

[0174]

ACTUATOR MOTION
Actuator motion	Description	Advantages	Disadvantages	Examples
Volume expansion	The volume of the actuator changes, pushing the ink	424)	Simple construction in the case	425)	High energy is typically required	426)	Hewlett-Packard
	in all directions.		of thermal ink jet		to achieve volume expansion. This		Thermal
					leads to thermal stress, cavitation, and		Inkjet
					kogation in thermal ink jet	427)	Canon
					implementations		Bubblejet
Linear, normal to integrated	The actuator moves in a direction	428)	Efficient coupling to	429)	High fabrication complexity may	430)	IJ01, IJ02, IJ04,
circuit surface	normal to the print head surface. The		ink drops ejected normal to		be required to achieve perpendicular		IJ07
	nozzle is typically in the line of		the surface		motion	431)	IJ11, IJ14
	movement.
Linear, parallel	The actuator moves parallel to the	432)	Suitable for planar	433)	Fabrication complexity	436)	IJ12, IJ13, IJ15,
to integrated	print head surface. Drop ejection may		fabrication	434)	Friction		IJ33,
circuit surface	still be normal to the surface.			435)	Stiction	437)	IJ34, IJ35, IJ36
Membrane push	An actuator with a high force but	438)	The effective area of	439)	Fabrication complexity	442)	1982 Howkins
	small area is used to push a stiff		the actuator becomes the	440)	Actuator size		U.S. Pat. No. 4,459,601
	membrane that is in contact with the		membrane area	441)	Difficulty of integration in a
	ink.				VLSI process
Rotary	The actuator causes the rotation of	443)	Rotary levers may be	445)	Device complexity	447)	IJ05, IJ08, IJ13,
	some element, such a grill or		used to increase travel	446)	May have friction at a pivot point		IJ28
	impeller	444)	Small integrated
			circuit area requirements
Bend	The actuator bends when energized.	448)	A very small change in	449)	Requires the actuator to be made	450)	1970 Kyser et al
	This may be due to differential		dimensions can be		from at least two distinct layers, or to		U.S. Pat. No. 3,946,398
	thermal expansion, piezoelectric		converted to a large motion.		have a thermal difference across the	451)	1973 Stemme
	expansion, magnetostriction, or other				actuator		U.S. Pat. No. 3,747,120
	form of relative dimensional change.					452)	IJ03, IJ09, IJ10,
							IJ19
						453)	IJ23, IJ24, IJ25,
							IJ29
						454)	IJ30, IJ31, IJ33,
							IJ34
						455)	IJ35
Swivel	The actuator swivels around a central	456)	Allows operation	458)	Inefficient coupling to the ink	459)	IJ06
	pivot. This motion is suitable where		where the net linear force on		motion
	there are opposite forces applied to		the paddle is zero
	opposite sides of the paddle, e.g.	457)	Small integrated
	Lorenz force.		circuit area requirements
Straighten	The actuator is normally bent, and	460)	Can be used with	461)	Requires careful balance of	462)	IJ26, IJ32
	straightens when energized.		shape memory alloys where		stresses to ensure that the quiescent
			the austenic phase is planar		bend is accurate
Double bend	The actuator bends in one direction	463)	One actuator can be	466)	Difficult to make the drops	468)	IJ36, IJ37, IJ38
	when one element is energized, and		used to power two nozzles.		ejected by both bend directions
	bends the other way when another	464)	Reduced integrated		identical.
	element is energized.		circuit size.	467)	A small efficiency loss compared
		465)	Not sensitive to		to equivalent single bend actuators.
			ambient temperature
Shear	Energizing the actuator causes a	469)	Can increase the	470)	Not readily applicable to other	471)	1985 Fishbeck
	shear motion in the actuator material.		effective travel of		actuator mechanisms		U.S. Pat. No. 4,584,590
			piezoelectric actuators
Radial	The actuator squeezes an ink	472)	Relatively easy to	473)	High force required	476)	1970 Zoltan
constriction	reservoir, forcing ink from a		fabricate single nozzles	474)	Inefficient		U.S. Pat. No. 3,683,212
	constricted nozzle.		from glass tubing as	475)	Difficult to integrate with VLSI
			macroscopic structures		processes
Coil/uncoil	A coiled actuator uncoils or coils	477)	Easy to fabricate as a	479)	Difficult to fabricate for non-	481)	IJ17, IJ21, IJ34,
	more tightly. The motion of the free		planar VLSI process		planar devices		IJ35
	end of the actuator ejects the ink.	478)	Small area required,	480)	Poor out-of-plane stiffness
			therefore low cost
Bow	The actuator bows (or buckles) in the	482)	Can increase the speed	484)	Maximum travel is constrained	486)	IJ16, IJ18, IJ27
	middle when energized.		of travel	485)	High force required
		483)	Mechanically rigid
Push-Pull	Two actuators control a shutter. One	487)	The structure is pinned	488)	Not readily suitable for inkjets	489)	IJ18
	actuator pulls the shutter, and the		at both ends, so has a high		which directly push the ink
	other pushes it.		out-of-plane rigidity
Curl inwards	A set of actuators curl inwards to	490)	Good fluid flow to the	491)	Design complexity	492)	IJ20, IJ42
	reduce the volume of ink that they		region behind the actuator
	enclose.		increases efficiency
Curl outwards	A set of actuators curl outwards,	493)	Relatively simple	494)	Relatively large integrated circuit	495)	IJ43
	pressurizing ink in a chamber		construction		area
	surrounding the actuators, and
	expelling ink from a nozzle in the
	chamber.
Iris	Multiple vanes enclose a volume of	496)	High efficiency	498)	High fabrication complexity	500)	IJ22
	ink. These simultaneously rotate,	497)	Small integrated	499)	Not suitable for pigmented inks
	reducing the volume between the		circuit area
	vanes.
Acoustic	The actuator vibrates at a high	501)	The actuator can be	502)	Large area required for efficient	506)	1993
vibration	frequency.		physically distant from the		operation at useful frequencies		Hadimioglu et al,
			ink	503)	Acoustic coupling and crosstalk		EUP 550,192
				504)	Complex drive circuitry	507)	1993 Elrod et al,
				505)	Poor control of drop volume and		EUP 572,220
					position
None	In various ink jet designs the actuator	508)	No moving parts	509)	Various other tradeoffs are	510)	Silverbrook, EP
	does not move.				required to eliminate moving parts		0771 658 A2 and
							related patent
							applications
						511)	Tone-jet

[0175]

NOZZLE REFILL METHOD
Nozzle
refill
method	Description	Advantages	Disadvantages	Examples
Surface	After the actuator is energized, it	512)	Fabrication	514)	Low speed	517)	Thermal inkjet
tension	typically returns rapidly to its normal		simplicity	515)	Surface tension force relatively	518)	Piezoelectric
	position. This rapid return sucks in	513)	Operational		small compared to actuator force		inkjet
	air through the nozzle opening. The		simplicity	516)	Long refill time usually	519)	IJ01–IJ07, IJ10–IJ14
	ink surface tension at the nozzle then				dominates the total repetition	520)	IJ16, IJ20, IJ22–IJ45
	exerts a small force restoring the				rate
	meniscus to a minimum area.
Shuttered	Ink to the nozzle chamber is provided	521)	High speed	523)	Requires common ink pressure	525)	IJ08, IJ13, IJ15,
oscillating	at a pressure that oscillates at twice	522)	Low actuator		oscillator		IJ17
ink	the drop ejection frequency. When a		energy, as the	524)	May not be suitable for	526)	IJ18, IJ19, IJ21
pressure	drop is to be ejected, the shutter is		actuator need only		pigmented inks
	opened for 3 half cycles: drop		open or close
	ejection, actuator return, and refill.		the shutter, instead
			of ejecting the
			ink drop
Refill	After the main actuator has ejected a	527)	High speed, as the	528)	Requires two independent	529)	IJ09
actuator	drop a second (refill) actuator is		nozzle is actively		actuators per nozzle
	energized. The refill actuator pushes		refilled
	ink into the nozzle chamber. The
	refill actuator returns slowly, to
	prevent its return from emptying the
	chamber again.
Positive	The ink is held a slight positive	530)	High refill rate,	531)	Surface spill must be prevented	533)	Silverbrook, EP
ink	pressure. After the ink drop is		therefore a high	532)	Highly hydrophobic print head		0771 658 A2 and
pressure	ejected, the nozzle chamber fills		drop repetition		surfaces are required		related patent
	quickly as surface tension and ink		rate is possible				applications
	pressure both operate to refill the					534)	Alternative for:
	nozzle.					535)	IJ01–IJ07, IJ10–IJ14
						536)	IJ16, IJ20, IJ22–IJ45

[0176]

METHOD OF RESTRICTING BACK-FLOW THROUGH INLET
Inlet back-flow restriction method	Description	Advantages	Disadvantages	Examples
Long inlet channel	The ink inlet channel to the nozzle	537)	Design simplicity	540)	Restricts refill rate	543)	Thermal inkjet
	chamber is made long and relatively	538)	Operational simplicity	541)	May result in a relatively large	544)	Piezoelectric
	narrow, relying on viscous drag to	539)	Reduces crosstalk		integrated circuit area		inkjet
	reduce inlet back-flow.			542)	Only partially effective	545)	IJ42, IJ43
Positive ink pressure	The ink is under a positive pressure,	546)	Drop selection and separation forces can	548)	Requires a method (such as a	549)	Silverbrook, EP
	so that in the quiescent state some of		be reduced		nozzle rim or effective hydrophobizing,		0771 658 A2 and
	the ink drop already protrudes from	547)	Fast refill time		or both) to prevent flooding of the		related patent
	the nozzle.				ejection surface of the print head.		applications
	This reduces the pressure in the					550)	Possible
	nozzle chamber which is required to						operation of the
	eject a certain volume of ink. The						following:
	reduction in chamber pressure results					551)	IJ01–IJ07, IJ09–IJ12
	in a reduction in ink pushed out					552)	IJ14, IJ16, IJ20,
	through the inlet.						IJ22,
						553)	IJ23–IJ34, IJ36–IJ41
						554)	IJ44
Baffle	One or more baffles are placed in the	555)	The refill rate is not as restricted as the	557)	Design complexity	559)	HP Thermal Ink
	inlet ink flow. When the actuator is		long inlet method.	558)	May increase fabrication		Jet
	energized, the rapid ink movement	556)	Reduces crosstalk		complexity (e.g. Tektronix hot melt	560)	Tektronix
	creates eddies which restrict the flow				Piezoelectric print heads).		piezoelectric ink jet
	through the inlet. The slower refill
	process is unrestricted, and does not
	result in eddies.
Flexible flap restricts inlet	In this method recently disclosed by	561)	Significantly reduces back-flow for edge-	562)	Not applicable to most inkjet	565)	Canon
	Canon, the expanding actuator		shooter thermal ink jet devices		configurations
	(bubble) pushes on a flexible flap			563)	Increased fabrication complexity
	that restricts the inlet.			564)	Inelastic deformation of polymer
					flap results in creep over extended use
Inlet filter	A filter is located between the ink	566)	Additional advantage	568)	Restricts refill rate	570)	IJ04, IJ12, IJ24,
	inlet and the nozzle chamber. The		of ink filtration	569)	May result in complex		IJ27
	filter has a multitude of small holes	567)	Ink filter may be		construction	571)	IJ29, IJ30
	or slots, restricting ink flow. The		fabricated with no
	filter also removes particles which		additional process steps
	may block the nozzle.
Small inlet compared to nozzle	The ink inlet channel to the nozzle	572)	Design simplicity	573)	Restricts refill rate	576)	IJ02, IJ37, IJ44
	chamber has a substantially smaller			574)	May result in a relatively large
	cross section than that of the nozzle,				integrated circuit area
	resulting in easier ink egress out of			575)	Only partially effective
	the nozzle than out of the inlet.
Inlet shutter	A secondary actuator controls the	577)	Increases speed of the ink-jet print head	578)	Requires separate refill actuator	579)	IJ09
	position of a shutter, closing off the		operation		and drive circuit
	ink inlet when the main actuator is
	energized.
The inlet is located behind	The method avoids the problem of	580)	Back-flow problem is eliminated	581)	Requires careful design to	582)	IJ01, IJ03, IJ05,
the ink-pushing surface	inlet back-flow by arranging the ink-				minimize the negative pressure behind		IJ06
	pushing surface of the actuator				the paddle	583)	IJ07, IJ10, IJ11,
	between the inlet and the nozzle.						IJ14
						584)	IJ16, IJ22, IJ23,
							IJ25
						585)	IJ28, IJ31, IJ32,
							IJ33
						586)	IJ34, IJ35, IJ36,
							IJ39
						587)	IJ40, IJ41
Part of the actuator moves to	The actuator and a wall of the ink	588)	Significant reductions in back-flow can	590)	Small increase in fabrication	591)	IJ07, IJ20, IJ26,
shut off the inlet	chamber are arranged so that the		be achieved		complexity		IJ38
	motion of the actuator closes off the
	inlet.	589)	Compact designs possible
Nozzle actuator does not result in	In some configurations of ink jet,	592)	Ink back-flow problem is eliminated	593)	None related to ink back-flow on	594)	Silverbrook, EP
ink back-flow	there is no expansion or movement of				actuation		0771 658 A2 and
	an actuator which may cause ink						related patent
	back-flow through the inlet.						applications
						595)	Valve-jet
						596)	Tone-jet
						597)	IJ08, IJ13, IJ15,
							IJ17
						598)	IJ18, IJ19, IJ21

[0177]

NOZZLE CLEARING METHOD
Nozzle Clearing method	Description	Advantages	Disadvantages	Examples
Normal nozzle firing	All of the nozzles are fired	599)	No added complexity on the print head	600)	May not be sufficient to displace	601)	Most ink jet
	periodically, before the ink has a				dried ink		systems
	chance to dry. When not in use the					602)	IJ01–IJ07, IJ09–IJ12
	nozzles are sealed (capped) against					603)	IJ14, IJ16, IJ20,
	air.						IJ22
	The nozzle firing is usually					604)	IJ23–IJ34, IJ36–IJ45
	performed during a special clearing
	cycle, after first moving the print
	head to a cleaning station.
Extra power to ink heater	In systems which heat the ink, but do	605)	Can be highly effective if the heater is	606)	Requires higher drive voltage for	608)	Silverbrook, EP
	not boil it under normal situations,		adjacent to the nozzle		clearing		0771 658 A2 and
	nozzle clearing can be achieved by			607)	May require larger drive		related patent
	over-powering the heater and boiling				transistors		applications
	ink at the nozzle.
Rapid succession of actuator	The actuator is fired in rapid	609)	Does not require extra drive circuits on the	611)	Effectiveness depends	612)	May be used
pulses	succession. In some configurations,		print head		substantially upon the configuration of		with:
	this may cause heat build-up at the	610)	Can be readily controlled and initiated by		the inkjet nozzle	613)	IJ01–IJ07, IJ09–IJ11
	nozzle which boils the ink, clearing		digital logic			614)	IJ14, IJ16, IJ20,
	the nozzle. In other situations, it may						IJ22
	cause sufficient vibrations to					615)	IJ23–IJ25, IJ27–IJ34
	dislodge clogged nozzles.					616)	IJ36–IJ45
Extra power to ink pushing	Where an actuator is not normally	617)	A simple solution where applicable	618)	Not suitable where there is a hard	619)	May be used
actuator	driven to the limit of its motion,				limit to actuator movement		with:
	nozzle clearing may be assisted by					620)	IJ03, IJ09, IJ16,
	providing an enhanced drive signal to						IJ20
	the actuator.					621)	IJ23, IJ24, IJ25,
							IJ27
						622)	IJ29, IJ30, IJ31,
							IJ32
						623)	IJ39, IJ40, IJ41,
							IJ42
						624)	IJ43, IJ44, IJ45
Acoustic resonance	An ultrasonic wave is applied to the	625)	A high nozzle clearing capability can be achieved	627)	High implementation cost if	628)	IJ08, IJ13, IJ15,
	ink chamber. This wave is of an	626)	May be implemented at very low cost in systems		system does not already include an		IJ17
	appropriate amplitude and frequency		which already include acoustic actuators		acoustic actuator	629)	IJ18, IJ19, IJ21
	to cause sufficient force at the nozzle
	to clear blockages. This is easiest to
	achieve if the ultrasonic wave is at a
	resonant frequency of the ink cavity.
Nozzle clearing plate	A microfabricated plate is pushed	630)	Can clear severely clogged nozzles	631)	Accurate mechanical alignment is	635)	Silverbrook, EP
	against the nozzles. The plate has a				required		0771 658 A2 and
	post for every nozzle. The array of			632)	Moving parts are required		related patent
	posts			633)	There is risk of damage to the		applications
					nozzles
				634)	Accurate fabrication is required
Ink pressure pulse	The pressure of the ink is temporarily	636)	May be effective where other methods cannot	637)	Requires pressure pump or other	640)	May be used
	increased so that ink streams from all		be used		pressure actuator		with all IJ series ink
	of the nozzles. This may be used in			638)	Expensive		jets
	conjunction with actuator energizing.			639)	Wasteful of ink
Print head wiper	A flexible ‘blade’ is wiped across the	641)	Effective for planar print head surfaces	643)	Difficult to use if print head	646)	Many ink jet
	print head surface. The blade is	642)	Low cost		surface is non-planar or very fragile		systems
	usually fabricated from a flexible			644)	Requires mechanical parts
	polymer, e.g. rubber or synthetic			645)	Blade can wear out in high
	elastomer.				volume print systems
Separate ink boiling heater	A separate heater is provided at the	647)	Can be effective where other nozzle clearing methods	649)	Fabrication complexity	650)	Can be used
	nozzle although the normal drop e-		cannot be used				with many IJ series
	ection mechanism does not require it.	648)	Can be implemented at no additional cost in some				ink jets
	The heaters do not require individual		inkjet configurations
	drive circuits, as many nozzles can
	be cleared simultaneously, and no
	imaging is required.

[0178]

NOZZLE PLATE CONSTRUCTION
Nozzle plate construction	Description	Advantages	Disadvantages	Examples
Electroformed nickel	A nozzle plate is separately	651)	Fabrication simplicity	652)	High temperatures and pressures	655)	Hewlett Packard
	fabricated from electroformed nickel,				are required to bond nozzle plate		Thermal Inkjet
	and bonded to the print head			653)	Minimum thickness constraints
	integrated circuit.			654)	Differential thermal expansion
Laser ablated or drilled polymer	Individual nozzle holes are ablated	656)	No masks required	660)	Each hole must be individually	664)	Canon
	by an intense UV laser in a nozzle	657)	Can be quite fast		formed		Bubblejet
	plate, which is typically a polymer	658)	Some control over	661)	Special equipment required	665)	1988 Sercel et
	such as polyimide or polysulphone		nozzle profile is possible	662)	Slow where there are many		al., SPIE, Vol. 998
		659)	Equipment required is		thousands of nozzles per print		Excimer Beam
			relatively low cost		head		Applications, pp. 76–83
				663)	May produce thin burrs at exit	666)	1993 Watanabe
					holes		et al., U.S. Pat. No. 5,208,604
Silicon micromachined	A separate nozzle plate is	667)	High accuracy is	668)	Two part construction	672)	K. Bean, IEEE
	micromachined from single crystal		attainable	669)	High cost		Transactions on
	silicon, and bonded to the print head			670)	Requires precision alignment		Electron Devices,
	wafer.			671)	Nozzles may be clogged by		Vol. ED-25, No. 10,
					adhesive		1978, pp 1185–1195
						673)	Xerox 1990
							Hawkins et al., U.S. Pat. No.
							4,899,181
Glass capillaries	Fine glass capillaries are drawn from	674)	No expensive	676)	Very small nozzle sizes are	678)	1970 Zoltan
	glass tubing. This method has been		equipment required		difficult to form		U.S. Pat. No. 3,683,212
	used for making individual nozzles,	675)	Simple to make single	677)	Not suited for mass production
	but is difficult to use for bulk		nozzles
	manufacturing of print heads with
	thousands of nozzles.
Monolithic, surface micromachined using VLSI lithographic	The nozzle plate is deposited as a	679)	High accuracy (<1 μm)	683)	Requires sacrificial layer under	685)	Silverbrook, EP
processes	layer using standard VLSI deposition	680)	Monolithic		the nozzle plate to form the nozzle		0771 658 A2 and
	techniques. Nozzles are etched in the	681)	Low cost		chamber		related patent
	nozzle plate using VLSI lithography	682)	Existing processes can	684)	Surface may be fragile to the		applications
	and etching.		be used		touch	686)	IJ01, IJ02, IJ04,
							IJ11
						687)	IJ12, IJ17, IJ18,
							IJ20
						688)	IJ22, IJ24, IJ27,
							IJ28
						689)	IJ29, IJ30, IJ31,
							IJ32
						690)	IJ33, IJ34, IJ36,
							IJ37
						691)	IJ38, IJ39, IJ40,
							IJ41
						692)	IJ42, IJ43, IJ44
Monolithic, etched through substrate	The nozzle plate is a buried etch stop	693)	High accuracy (<1 μm)	697)	Requires long etch times	699)	IJ03, IJ05, IJ06,
	in the wafer. Nozzle chambers are	694)	Monolithic	698)	Requires a support wafer		IJ07
	etched in the front of the wafer, and	695)	Low cost			700)	IJ08, IJ09, IJ10,
	the wafer is thinned from the back	696)	No differential				IJ13
	side. Nozzles are then etched in the		expansion			701)	IJ14, IJ15, IJ16,
	etch stop layer.						IJ19
						702)	IJ21, IJ23, IJ25,
							IJ26
No nozzle plate	Various methods have been tried to	703)	No nozzles to become	704)	Difficult to control drop position	706)	Ricoh 1995
	eliminate the nozzles entirely, to		clogged		accurately		Sekiya et al U.S. Pat. No.
	prevent nozzle clogging. These			705)	Crosstalk problems		5,412,413
	include thermal bubble mechanisms					707)	1993
	and acoustic lens mechanisms						Hadimioglu et al EUP
							550,192
						708)	1993 Elrod et al
							EUP 572,220
Trough	Each drop ejector has a trough	709)	Reduced manufacturing	711)	Drop firing direction is sensitive	712)	IJ35
	through which a paddle moves. There		complexity		to wicking.
	is no nozzle plate.	710)	Monolithic
Nozzle slit	The elimination of nozzle holes and	713)	No nozzles to become	714)	Difficult to control drop position	716)	1989 Saito et al
instead of	replacement by a slit encompassing		clogged		accurately		U.S. Pat. No. 4,799,068
individual	many actuator positions reduces			715)	Crosstalk problems
nozzles	nozzle clogging, but increases
	crosstalk due to ink surface waves

[0179]

DROP EJECTION DIRECTION
Ejection direction	Description	Advantages	Disadvantages	Examples
Edge (‘edge shooter’)	Ink flow is along the surface of the	717)	Simple construction	722)	Nozzles limited to edge	725)	Canon
	integrated circuit, and ink drops are	718)	No silicon etching required	723)	High resolution is difficult		Bubblejet 1979 Endo
	ejected from the integrated circuit	719)	Good heat sinking via substrate	724)	Fast color printing requires one		et al GB patent
	edge.	720)	Mechanically strong		print head per color		2,007,162
		721)	Ease of integrated circuit			726)	Xerox heater-in-
			handing				pit 1990 Hawkins et
							al U.S. Pat. No. 4,899,181
						727)	Tone-jet
Surface (‘roof shooter’)	Ink flow is along the surface of the	728)	No bulk silicon etching required	731)	Maximum ink flow is severely	732)	Hewlett-
	integrated circuit, and ink drops are	729)	Silicon can make an effective		restricted		Packard TIJ 1982
	ejected from the integrated circuit		heat sink				Vaught et al U.S. Pat. No.
	surface, normal to the plane of the	730)	Mechanical strength				4,490,728
	integrated circuit.					733)	IJ02, IJ11, IJ12,
							IJ20
						734)	IJ22
Through integrated circuit, forward (‘up	Ink flow is through the integrated	735)	High ink flow	738)	Requires bulk silicon etching	739)	Silverbrook, EP
shooter’)	circuit, and ink drops are ejected	736)	Suitable for pagewidth print				0771 658 A2 and
	from the front surface of the						related patent
	integrated circuit.	737)	High nozzle packing density therefore low				applications
			manufacturing cost			740)	IJ04, IJ17, IJ18,
							IJ24
						741)	IJ27–IJ45
Through integrated circuit, reverse (‘down	Ink flow is through the integrated	742)	High ink flow	745)	Requires wafer thinning	747)	IJ01, IJ03, IJ05,
shooter’)	circuit, and ink drops are ejected	743)	Suitable for pagewidth print	746)	Requires special handling during		IJ06
	from the rear surface of the	744)	High nozzle packing		manufacture	748)	IJ07, IJ08, IJ09,
	integrated circuit.		density therefore low				IJ10
			manufacturing cost			749)	IJ13, IJ14, IJ15,
							IJ16
						750)	IJ19, IJ21, IJ23,
							IJ25
						751)	IJ26
Through	Ink flow is through the actuator,	752)	Suitable for piezoelectric print	753)	Pagewidth print heads require	756)	Epson Stylus
actuator	which is not fabricated as part of the		heads		several thousand connections to drive	757)	Tektronix hot
	same substrate as the drive				circuits		melt piezoelectric ink
	transistors.			754)	Cannot be manufactured in		jets
					standard CMOS fabs
				755)	Complex assembly required

[0180]

INK TYPE
Ink type	Description	Advantages	Disadvantages	Examples
Aqueous, dye	Water based ink which typically	758)	Environmentally friendly	760)	Slow drying	765)	Most existing inkjets
	contains: water, dye, surfactant,			761)	Corrosive
	humectant, and biocide.	759)	No odor	762)	Bleeds on paper	766)	All IJ series ink jets
	Modern ink dyes have high water-			763)	May strikethrough
	fastness, light fastness			764)	Cockles paper	767)	Silverbrook, EP 0771 658 A2 and
							related patent applications
Aqueous, pigment	Water based ink which typically	768)	Environmentally friendly	773)	Slow drying	778)	IJ02, IJ04, IJ21, IJ26
	contains: water, pigment, surfactant,	769)	No odor	774)	Corrosive
	humectant, and biocide.	770)	Reduced bleed	775)	Pigment may clog nozzles	779)	IJ27, IJ30
	Pigments have an advantage in	771)	Reduced wicking	776)	Pigment may clog actuator	780)	Silverbrook, EP 0771 658 A2 and
	reduced bleed, wicking and	772)	Reduced strikethrough		mechanisms		related patent applications
	strikethrough.			777)	Cockles paper
						781)	Piezoelectric ink-jets
						782)	Thermal ink jets
							(with significant restrictions)
Methyl Ethyl Ketone (MEK)	MEK is a highly volatile solvent used	783)	Very fast drying	785)	Odorous	787)	All IJ series ink jets
	for industrial printing on difficult	784)	Prints on various	786)	Flammable
	surfaces such as aluminum cans.		substrates such as metals
			and plastics
Alcohol (ethanol, 2- butanol,	Alcohol based inks can be used	788)	Fast drying	792)	Slight odor	794)	All IJ series ink jets
and others)	where the printer must operate at	789)	Operates at sub-freezing	793)	Flammable
	temperatures below the freezing		temperatures
	point of water. An example of this is	790)	Reduced paper cockle
	in-camera consumer photographic	791)	Low cost
	printing.
Phase change (hot melt)	The ink is solid at room temperature,	795)	No drying time- ink	801)	High viscosity	807)	Tektronix hot melt
	and is melted in the print head before		instantly freezes on the print	802)	Printed ink typically has a ‘waxy’		piezoelectric ink jets
	jetting. Hot melt inks are usually wax		medium		feel
	based, with a melting point around	796)	Almost any print medium can be	803)	Printed pages may ‘block’	808)	1989 Nowak
	80° C. After jetting the ink freezes		used	804)	Ink temperature may be above the		U.S. Pat. No. 4,820,346
	almost instantly upon contacting the	797)	No paper cockle occurs		curie point of permanent magnets	809)	All IJ series ink jets
	print medium or a transfer roller.	798)	No wicking occurs	805)	Ink heaters consume power
		799)	No bleed occurs	806)	Long warm-up time
		800)	No strikethrough
			occurs
Oil	Oil based inks are extensively used in	810)	High solubility medium for some	813)	High viscosity: this is a	815)	All IJ series ink jets
	offset printing. They have advantages		dyes		significant limitation for use in inkjets,
	in improved characteristics on paper	811)	Does not cockle paper		which usually require a low viscosity.
	(especially no wicking or cockle). Oil	812)	Does not wick through paper		Some short chain and multi-branched
	soluble dies and pigments are				oils have a sufficiently low viscosity.
	required.			814)	Slow drying
Microemulsion	A microemulsion is a stable, self	816)	Stops ink bleed	820)	Viscosity higher than water	823)	All IJ series ink jets
	forming emulsion of oil, water, and	817)	High dye solubility	821)	Cost is slightly higher than water
	surfactant. The characteristic drop	818)	Water, oil, and amphiphilic soluble dies		based ink
	size is less than 100 nm, and is		can be used	822)	High surfactant concentration
	determined by the preferred				required (around 5%)
	curvature of the surfactant.	819)	Can stabilize pigment suspensions

Ink Jet Printing