Project News

updated Thursday, April 6th, 2000

pstewart@gwi.net

New on the web site

Julie Porter gets Cybermaxx LCD up and running April 4th

2 Designs Underway
Kristian's Design An RGB processor design to go with it is available here
A Fully Integrated Design Pictures and PC board layouts available here
Groundwork
Data Julie S. Porter Has Collected
The Care and Feeding of the Cybermaxx LCDs

New on the web site I've been spending some time revising these pages, recently, with some interesting additions to report here. More Design Test Details Improved 'Care and Feeding' ('HowTo') Document New Cybermaxx Links Added Index Page Added Information Source Page Added Site Redesign Planned Improvements More details on our successful tests. First is the expanded information from Julie Porter's successful test of our LCD driver board design. I'll be improving the descriptions and organization of these results over the next few weeks, but for now, you can see pictures of the working board apparatus, a brief explanation of the adjustments that made the test a success, and some oscilloscope pictures from various system test points. I will be working to provide thumbnail images and other amenities, in addition to passing on technical details, as time allows. Improved 'Care and Feeding of the Cybermaxx LCD' page. I've put a lot of time and effort into updating the 'Care and Feeding of the Cybermaxx LCD' page, clarifying and expanding its coverage to include the results of our design tests. In some cases I've made amendments to improve the page's technical accuracy. Since I think of it as the main summary page of our LCD reverse-engineering effort, I have added a much more thorough introduction, making it more self-sufficient as an explanation of the project, and improved the organization of the part where I explain our reverse-engineering method. As time allows, I will be making further improvements to the page; a second version of it, in bite-sized hyperlinked chunks, will be available, hopefully soon. Graphic improvements to the page, including redesigned diagrams, and new explanatory diagrams where they will be useful, are planned; as graphics-capable computers avail themselves for me to work on, you will see the new diagrams on the 'Care and Feeding' page. New Cybermaxx Links I have added some links which will be of interest to people working with the Cybermaxx. I've located a company on the web that did product development for Victormaxx, and which has several pages devoted to a 'case study' of their development efforts, showing conceptual sketches, early prototypes, and other interesting information on the Cybermaxx's development. More links will be going up on the 'Related Links' page, as time allows. A link to Eric Smith's page will be among these. He worked on the Cybermaxx head-tracking software, and was of great help when some questions came in about the Cybermaxx navigation system. 'Index' page added I've finally added the long-missing 'index' page, an amenity which should help navigation. 'Information Sources' page added There is now an 'Information Sources' page which collects, all in one page, a compendium of the sources we use in our study. This is nearly exhaustively filled in now. If something is missing from it that you can supply, do please send off an email to me about it. All of the sources we have have taken some work to find. Aesthetic and Organizational Changes Design Philosophy: Up until about a month ago, I let two principles guide my design of these pages: 1. The pages should load quickly. This meant making most of the site text-only, with links to graphics but very few graphics displayed 'in-line' as integral parts of the design. I restricted graphics to the LCD Project banners at the top of the page. 2. Visitors should be able to download pages whole, without having to surf through a procession of hyperlinks to compile a document that would be usable offline. As page size increased, this design guideline conflicted with the 'quick loading' requirement. Since internet bandwidth has improved greatly over the last several years, I have decided to relax Design Principle 1. This means I will be redesigning the site to make it more attractive (hopefully) and more useful (with diagrams next to text that refers to them), making extensive use of in-line images. Given many people's preference for short (max. 2 or 3 screen) pages, linked together heirarchically, I plan revisions of the longer pages to make them more 'hypertextual'-- to relax Design Principle 2. I will give the user the option to load the large one-piece pages, in which I rely heavily on cross-links and internal hyperlinks (anchor tags) for navigation, as well. This latter design is the one you will find as of this update. I apologize in advance to those of you who prefer shorter pages. Organizational changes: Some visitors to the site will notice subtle, and not-so-subtle, organizational changes. Some of these changes are detailed below. As our systems designs have matured, it made sense, for example, to place 'System Designs' first on our Resources page. Planned improvements In the future, if time and web space allow, I will be adding information sources for folks working with other subsystems of the Cybermaxx, including the head-tracking system. I have some blurry but probably readable-enough schematics of the head tracker, for example, and a parts list for a version of the Cybermaxx 2.0, provided by Arthur Zwern of the General Reality Corporation. These could be of use for people hoping to copy or improve upon the Cybermaxx VR system.

Julie Porter Gets Cybermaxx LCD Up and Running Julie S. Porter got her LCDs working over the weekend, sending email on Monday, April 4th that she had gotten a test pattern displayed stably on the screens. You can see her results at MN83803AL close-up working LCD close-up! working LCD & test rig (including several revisions of the driver circuit, and test pattern generator) working LCD & test rig, showing whole bench setup Here is a good view of the board layout (rotated 90 degrees counterclockwise from the orientation of the ExpressPCB layout): My Revision L board, mostly finished, March 2000 More details on Julie's test, including oscilloscope pictures for various test points Julie and I had been tinkering with Revision L for a while. This past month, we worked out various details in the design and began testing the boards, finally installing the LCDs on them and hooking up video sources to the inputs. In both our test rigs, voltage was sagging on the RGB input to the LCD. It's supposed to be offset by 7 volts from DC, but I was unable to get it to go above 6 volts by adjusting the DC voltage on the IR3y05y's Pin 28. On my board, I could not get the RGB DC offset above 6 volts, even after shorting across the positive-most resistor of the voltage divider feeding pin 28 of the IR3y05y, which controls the offset. What Julie did, confronting this problem, was to lower the LCCOM voltage (called 'COM' in the MCL0712A03 data sheets) to something like the 1.7 volts lower than RGB center voltage, as the data sheets specify. Adjusting the COM voltage, rather than keeping it fixed as in our original, camcorder-based design, worked. Presto: Video appeared! We're not sure what the long-term effects of running the LCD below its rated video center voltage will be, but one thing seems certain: We should get better dynamic range from the LCDs if we can drive them across their full input range-- and setting the center voltage to 7 with VCC2 near 14 volts is one very good way to do this (see our 'Care and Feeding of the Cybermaxx LCD' section (3.2), video center voltage (DC offset) control for details on this approach). Julie's success in getting video to show on the display validates a number of the design assumptions we've been making, showing that the camcorder schematics can be used as the basis for a working driver; showing that the MN83803AL chip will drive the LCDs; and finally, proving that the LCDs work. Her finding that the LCDs will work on less than their rated RGB center voltage is surprising and useful. We will be working out the details of the test setup in the next few weeks, working towards an optimal design. A full report of Julie's test setup and results will be available, as well. :)

2 Designs Underway
Right now we have two related design projects underway. One uses Kristian Bognaes's PC board design, modeled on the schematic for the Panasonic PV-D705 camcorder electronic viewfinder; the other, spearheaded by Julie S. Porter, uses the PV-750D camcorder EVF as the model for a fully integrated design.

Kristian's Design
Kristian has built a board (detailed under 'PC Boards' on this web site), and kindly provided me with a copy of the PC board itself, which, with Julie Porter's help and advice, I've built my own version of this spring. I have designed RGB inverter circuitry, and made a slight modification to the voltage divider for the LCD's Vcom (aka LCCOM), to prepare the board for testing. When I finish the inverter construction (a matter of ordering the resistors and getting home from vacation to finish it), I will test the board and post results here. Initially I will use the Nuts & Volts video pattern generator to supply video to this circuit.

I've provided a circuit idea for an RGB signal conditioner to go with Kristian's board. It's only partially tested, but will give you an idea of what to do even if it isn't the best circuit for the task. You can find it on our schematics page.

A Fully Integrated Design
Julie's design makes use of the Sharp IR3y05 video processing chip to provide all the treatment of NTSC composite video that is needed to make the LCD display a TV picture. This design can be readily modified to display computer RGB such as suggested by Tom Cumming, since the Sharp chip can also perform this function (having RGB inputs as well).

Revision 'L' of our design has almost all of the mistakes taken out of it, so it's available for download. Check it out on our pc boards page. The design is in ExpressPCB format, and is ready to be manufactured (we've made five of them). All you need to view it is to download the freeware ExpressPCB CAD program. Revision 'M,' a more mistake-free design, will be available on this site soon as well.

Groundwork
Julie's previous design employs an MC44011 video processor controlled by the MC68HC11EC microcontroller. Her extensive professional experience with embedded microcontroller applications design makes this a potentially viable route for future development.

Data Julie S. Porter Has Collected
Julie has experimented with the MN83803AK chip and gotten crucial data on how it works and what it does, supplementing the data sheets we have for it. Click for a summary of Julie's results. You may wish to look at the 'care and feeding of the Cybermaxx LCD' discussion below, and the camcorder schematics provided elsewhere on this web site, in addition to the MN83803A(k) data sheets, to fully understand what Julie finds (though of course I try to give the clearest summary of it possible).

...which leads us to the
The Care & Feeding of the CyberMaxx LCD. Here we summarize what the LCD needs to work. A block diagram simplifies the presentation. Almost all the information not presented on this page is indexed from it (e.g. inferences we draw from the schematics, discussed more thoroughly on the schematics page.

Click here to return to my home page.
Click here to return to the top of the present page.
Click here to return to the main LCD project page.