Go to previous article.
Go to next article.
Return to 1998 Conference Table of Contents
Max O. Lange
Phone +49 6421 8020
Fax +49 6421 80214
PRINT stands for "Non-Impact Printer/Plotter for Braille/Moon and Tactile Graphics". Going through those terms one by one may help explaining what the aims of PRINT are. "Non- impact" means PRINT is about an actual tactile printer, rather than an embosser. In fact the device under developments works much like a conventional ink-jet printer. "Plotter" is, therefore, a bit of a misnomer inasmuch as PRINT does not involve an X-Y steering system to draw dots and continuous lines, but rather prints all features line by line. "Braille" is probably obvious but "Moon" maybe less so. this has nothing to do with either Astronomy or religious groups, but is in fact a tactile font developed by one Mr. Moon, in Great Britain. I'll come back to Moon in detail later. Finally, "Tactile graphics" as such is pretty clear; what is innovative in PRINT, however, is the ease and sharpness with which tactile graphics can be produced - though, it must be added, with certain restrictions in relief height.
The project involves three small enterprises, two polytechnic colleges, and four end user organizations: * Blista-Brailletec
Three main reasons initially led to the development of PRINT: Firstly, Brailletec's wish to develop a noiseless braille printer; Secondly, the desire of certain organizations of the Blind to improve the access to tactile graphics information; and thirdly, the European Union's aim to integrate braille readers better into the modern communications environment. First consider noise reduction. The current generation of braille embossers, which almost invariably use magnetically actuated pins to make braille dots in paper, all make so much noise that their usefulness e.g. in classrooms or normal offices is severely restricted. At best, when fitted with an acoustic cabinet they are still a nuisance; at worst, they make work impossible while running. This situation led some Brailletec engineers to search for a completely different way to create braille as early as 1994. A comprehensive study was performed with the engineering college of Fachhochschule Augsburg, in southern Germany, to find suitable technologies. Among the candidates considered were rapid prototyping, vacuum forming, foam paint plotting, and many more. The outcome was somewhat surprising, as it appeared the solid ink printing technology, used normally in high-end color printers, had the greatest potential for braille considering cost, printing speed, and development effort.
Solid ink printing is a dot matrix printing technology, meaning that features are composed of tiny dots made by drops of color ejected at a high rate from a moving printhead. Matrix printing is commonly used in desktop printers, because it is cheap and allows printing high-quality graphics. Then, of course, introducing a matrix printing technology made it possible to increase the scope of the project because tactile graphics and Moon could then be produced as well.
Some explanation is needed here concerning Moon. You all know that in the developed countries, most blind are over 65 when they lose their sight. A that age, it is difficult for them to learn braille, not least because the sensitivity of the fingers is often degraded. However, most of them could perfectly well read capital Latin characters when they were sighted. That led to the development of Moon, which is based on greatly simplified capitals rather than dots. Most elderly blind and also many with degraded learning skills find Moon easier to learn and read than braille. However, as it involves lines, it is difficult to emboss without destroying the paper. In fact most Moon is currently produced as thermoform copies from originals made with a stereo lithography device normally used for rapid prototyping. A Moon printer would greatly reduce the cost of production, and also make it possible to use normal paper, rather than thermoform. That is why there was an ongoing requirement within the RNIB, the world's leading promoter of Moon, for a new, Moon-capable, tactile printer.
Now it is obvious that similar cost benefits would apply to tactile graphics production with solid ink. Furthermore, such a system would for the first time make it possible to print faxes in a tactile way. So, a proposal was written and accepted within the European union's Telematics Initiative for the Disabled and Elderly, or TIDE, program, for a tactile printer capable of printing braille, Moon, and tactile graphics. Upon a suggestion from the EU, a capability to produce colored output was also included. This is because many severely visually impaired benefit from the combination of tactile and visual output, and also because it would make it possible to combine tactual and purely visual information on the same page. However, with all these capabilities the project had evolved from a relatively straightforward improvement of an existing technology to a rather more ambitious innovation. Therefore, several user groups were invited to participate, and a thorough market study and user needs analysis preceded the main development phases.
There are five main phases in any research and development project: * Market analysis
PRINT is currently in the third phase, design and prototyping. In fact an engineering prototype has been made which is able to demonstrate the basic feasibility of the concept, and a series of advanced prototypes or zero-series models is being designed. But let us look at the phases one by one. At the beginning of the project, there was no real alternative to PRINT, so the size and quality of the market was not quite obvious. Relief printing is not altogether new. In fact Howtek produced a relief color printer, the Pixelmaster, for a time, though this was a very expensive machine, delicate to operate and very unreliable, and predictably failed in the market. There were also projects such as Graphtact at Quebec University, or Relief at Vienna University,and I understand there is currently an effort underway in the NBL to get funding for a project that would use a similar technology to PRINT. But all currently available tactile printers still are embossers, and none offer satisfactory raised graphics. Swell paper and thermoform copies exist, of course, but in both cases, the cost of production are high and the output quality is just marginally adequate. Several other, proprietary technologies are used in braille publishing, none of which has been, or aspired to be, successful in the market. In order to create a solid database, therefore, over 100 blind and sighted users in four European countries were interviewed and asked to fill in a lengthy questionnaire to determine their requirements for a future relief printer. Options ranged from a small, portable device to a printer designed for high-rate production use with complex paper handling capabilities. Noise, health, and environmental aspects were considered, as were usability and maintenance. The outcome of this study was a requirements catalog that formed the basis of the functional printer specifications.
Some of these functional specifications are: Print at least one page per minute, on paper up to 12" wide, and use cut sheets from a stack. Reliable operation figured high on the users' wish-list. From a technical point of view, requirements include a 200 dpi resolution, standard PCL3 drivers, parallel and USB connectors, and the possibility to load font tables - in other words, compatibility with a modern office environment. Apart from just designing a printer, the project contains some original research aspects that serve to make the product more useful. For example, the question of what sort of material to print with on paper has not, to my knowledge, previously been addressed from the perceptive psychology standpoint. It turns out in practical tests that readers feel a great difference between different materials. There is also a marked discrepancy notably between novice and well-practiced braille readers, specially those who were born blind. These latter tend to prefer a lower dot profile and a slightly soft feel to the dot, which is perceived as comfortable for fast reading, while novices may find it difficult to read, and prefer a sharp, hard dot instead. Both agree, however, that the surface roughness and friction coefficient of the material should be similar to that of the paper, so that the fingers will neither stick nor glide too easily over the dots. Two other important points are that sweat building under the fingers needs to be able to be absorbed into the paper, and that no rub-off from the ink must occur, since it, just as sweat, reduces the tactile sensitivity of a braille reader.
These considerations had important consequences for the development phase of PRINT, as the properties of the material used to create relief print determine the technology used to print with it. For example, transparent hot-melt glue could be dispensed from a needle by pressure. Glue feels fine to a sighted person, but its surface is so even and also so sticky that no satisfactory reading speed can be reached. Therefore, it should not be considered for relief generation. Polyester based plastic is typically perceived as a good material by novice braille readers because it offers a sharp dot profile and a hard edge, but practiced readers tend to dislike it for that very reason. Synthetic waxes, my third example, come with varying surface properties, so that by mixing together different waxes and additional substances, maybe also polymers, a specified surface quality can be maintained and thus the wax can be adapted to the users' needs. For this reason, our current favorite material is a formula that contains several different synthetic waxes and some other ingredients. Before entering the next phase - demonstration and validation - the wax relief ink will still undergo a series of tests to ensure it best meets the users' needs.
We'll skip the validation and exploitation phases here, because they have not yet begun for PRINT.
The first experiments were made using an off-the-shelf color printer, the Phaser 300i from Tektronix (TM). This is a high-quality, phase-change color printer the technology of which is rather close to what was intended for PRINT. With some modifications to the printer software, it was possible to print several layers of color on top of each other, thus creating an acceptable relief height. Experiments showed, though, that the "solid ink" used by Tektronix does not quite meet all the criteria for a good relief material. After some discussions with Tektronix, a number of printheads were obtained and a laboratory printer designed which served to create a great number of sample braille printouts using various waxes and other ingredients designed to make braille dots with an easy-to-read surface, good adhesion to paper, good durability, and the required characteristics of the liquid phase to fit through the Tektronix printhead. This required some creative chemistry, finally resulting in a choice of 15 different solid ink formulations that all technically met the criteria for relief printing such as being non-toxic, having the required durability, friction coefficient, and melting point, and allowing the addition of colorants. A field test was then performed to identify users' preferred relief inks and also learn more about what sort of paper is preferred by braille readers. In the event, four different mixtures came out ahead of the field in the users' preferences, and testing will continue with all of those to further optimize the printing process.
At the same time, due to the high cost of the Tektronix printhead, a novel printing technology is being developed with the University of Ilmenau in Germany, with the aim of increasing printing speed while lowering the cost of the printhead. This also helps avoiding certain intellectual property rights issues, since only proprietary technology is used. Ilmenau is among the world's leading micro system technology developers.
Solid ink is inserted as a block into the printhead, where it melts and is then ejected through a row of nozzles with 200 dpi resolution. The current design allows printing just over 1 standard (A4 or US letter) page per minute, with the printhead moving through each line and back twice to create the necessary dot profile. Paper is inserted and retrieved from the front of the printer and all operating elements - there are only a few - are grouped in an easy-to-locate field, to make operation self-explanatory even for the visually impaired. A language configurable voice output is integrated to announce printer mode settings, control inputs, error messages, or hints.
At the same time, a fax software and hardware is being developed. The fax subsystem will be an option that completely integrates with the printer to make a computer based relief faxing system. It will be able to create meaningful tactile output even from visual input, and offer an interactive fax recognition program.
While the design of the printer is relatively advanced, and all components have been tested using a laboratory setup, a first test run of the entire prototype system is yet several weeks away. After thorough testing, experience gathered with the prototype .will be used in a series of 15 demonstrators to be completed by the end of this year. Only a field test scheduled to begin in January next year will show how the system behaves in everyday life. Until then, the hardware and software components will be further improved and the printing technology and process optimized. The user organizations are accompanying the design at every step, regularly providing feedback to the engineering team. As a backup technology, the Tektronix printhead will also still be retained. The fax software is also continuously being developed (see the paper by Thorsten Puck on TacFax in the Saturday session). Once the demonstrators are available, an extensive validation phase will precede the actual production phase, to enable designers to incorporate the users' experiences and wishes realized while using the demonstrators. Market introduction is then expected for the first quarter 2000.
Go to previous article.
Go to next article.
Return to 1998 Conference Table of Contents
Return to Table of Proceedings