Showing posts with label screen. Show all posts
Showing posts with label screen. Show all posts

Sunday, June 24, 2012

Reader Control

There was, of course, the problem of control. A touchscreen display would've been great, but the laptop wasn't that advanced.

As you probably noticed from the Introduction, a touchpad was the chosen controller. Several pointing and control devices had been considered in addition to a touchpad, including some sort of keypad, a set of scroll wheels, a pointing stick, and a trackball. The touchpad won out based on flexibility and ease of integration with the display. A different choice may work better for you.

A pointing stick was high on the list based on size. It'd be smaller than about anything, but I couldn't find one that could be plugged in and used like a mouse. It would have had to be integrated into the display or some housing would have had to have been contrived, and arrangements would have been needed to wire it into the mouse interface on the base. On the other hand, USB and PS/2 touchpads were readily available and one could be attached to an edge of the display housing. That left the problem of figuring out a good way to attach it.

The Velcro came from a local fabric store. A nice older lady there asked what I was making with it, to which I smiled and said, “A computer.” She responded with silence and strange look...
In the end, I chose Velcro. Permanent attachment to the housing would have been easy enough, but temporary attachment would permit moving the pad to various locations as desired. Black-colored, soft, loop Velcro was attached around the three flat sides of the display frame and the matching rough, hook Velcro was attached on the back of the touchpad. The result provided a highly flexible solution with a fairly clean look.

I figured that, if something better came along, the touchpad could easily be replaced with a different controller of comparable size that could be plugged into an available USB or PS/2 port. But later in the project the touchpad turned out to be a very good choice with a few unanticipated advantages over other controllers.

Saturday, June 23, 2012

Assembling the Display

With the proof-of-concept trial done, it was now necessary to make things durable enough to be usable and finished enough to not distract from reading. Between the base and the display, the display was the more important in this respect since it would be the more jostled and bumped around of the two and would be the one that I'd be spending time looking at.

Different displays with different available space, materials, wiring location, and etc. will require different solutions to this problem.  Depending on your situation, you may need to be pretty clever and crafty to come up with a workable arrangement.  In my case, it took a bit of rigging to get things to work out...

The size of the display connector and the space in the display housing had already been considered.  I was sure the connector could be made to fit one way or another.  So, the first step was to cut an opening through which it would extend. This I accomplished by carving out parts of the top and bottom of the housing with a Dremel.

The intent was to bolt the connector directly to the housing, but was a bit of a tight squeeze and it turned out that there wouldn't be a way to do it securely. To get around those problems, I cut a metal picture hanger* in half and bolted each half to one side of the connector to form mounting extensions. Then I used the Dremel to cut the housing in places to fit the extensions as solidly as possible. Finally, I was able to fit the housing together and had a fairly decent connector added to the display.

Unfortunately, several of the wires were ripped out of their fragile connections in the process of fitting the connector. After a lot of tedious resoldering, the display was reassembled and tested. It worked pretty well at first. But the connector was a bit loose, and I found it possible to wiggle it enough to short something and cause the computer to reboot.

Reopening the display, I wedged a couple of small nuts between the housing and extensions, and then applied cyanoacrylate adhesive (i.e. “super glue”) to hold things in place. I also applied some of the glue to the connector extension nut/bolt connections to keep them from loosening.

The display was again reassembled and tested after everything had dried. Finally, I had a good, solid connector that was rugged enough for normal use and didn't seem to look too bad.

* If you're going to mod stuff, it helps to have a lot of junk lying around. :)

Wednesday, April 25, 2012

Connection to the Display

Next was to decide how to make the wired connections from the base to the display. I wanted to avoid hardwiring if possible to facilitate transport if needed, and it was clear from the service manual that the base and display were joined by a plug and socket connection. While the part numbers naturally weren't identified, I had the urge to make a cable that could be directly joined with the existing connectors. That would require tracking down the correct connectors.

I carefully examined the connectors after detaching the display from the base, which went fairly smoothly according to the directions in the service manual. Digging through part sites while paying particular attention to connector shapes finally turned up the necessary parts: SHD Connectors manufactured by JST. The parts were easily verified by comparing various dimensional measurements of the parts on the laptop to the part dimensions noted on the parts' spec sheets.

With the parts found, it turned out that a specialized hand tool was needed to properly make the electrical connections to the plug... And it cost, like, $1000! As that was just a bit outside my budget, I considered making the needed crimp connections using a pair of really small needle nose pliers. But that'd be a pain. And who wants to take on that level of tedium?

Display and base detached, display disassembled, and plug cut off from display.The connection was wired rather than being a flexible plastic circuit, so the decision was finally made to just wire in some other readily available connectors and use a readily available cable. The number of wires in the connection, the available space in the laptop display, and the stack of unused cables in my junk closet led to the choice... Parallel printer connectors. I cut off the plug from the display and was ready to go. 

Yes, it's backward.
Since a number of the wires were in twisted pairs, it seemed prudent to to use associated lines in the cable, if possible. Some searching turned up a table with complete parallel port cable pin assignments (see pp.9-10) and a connector pinout diagram with numbers adjacent pins for easy reference. Selecting one of the pins associated with a different signal ground pin for each of the twisted pairs and other pins for the remaining individual wires, the pinout was marked up according to the wire colors.

Then I noticed my mistake... The connections needed to be made to the female connector rather than to the male connector shown in the pinout diagram. Fortunately, this was a minor bump in the road. A couple of mild oaths and a bit of image processing later, I had an inverted image from which to work. Careful soldering to make the actual connections finished the initial work at the display end.

While it felt good to complete this first bit, a worry hung over the project... Would transmission line effects over a six-foot (182-cm) cable muck up the display? Would I need to put in extra impedance matching circuits? Instead of doing the wise thing and calculating it out, which would've required recalling forgotten college stuff from way back before the turn of the century, I had already decided to just try it. But I'd have to wait until the base connection was complete before seeing what would happen...