Wednesday, 1 March 2017

The Sexy World of Reverse Engineering

Mention 'reverse engineering' to the average layman, and it conjures up images of secretive high tech labs dismantling high tech state of the art state secret equipment, or of the Russians in the early 1950s taking apart, bolt by bolt, a captured USAAF B-29, and exactly recreating it as the Tupulov Tu-4.

So the world of reverse engineering is one of high tech, super sexy James Bond like secrets and super spies..., actually, no. In reality its a long, drawn out, painstaking process of careful accurate measurements. And here is what it looks like
The tedium of reverse engineering
Luckily, this isnt a B-29 Superfortress! What it is, is one of the LED display panels from the Spectrum 4160C message display board!

I have been trying everything I can think of to get information, data, firmware etc, for this unit. But so far all to no avail. Sadly the response from Alpha-american was far from helpful, it seems they are only interested in selling me a firmware EPROM (which might not even be the problem!). Spectrum corporation have been more helpful, with a few hints as to possible issues etc.

I suppose that I cant expect too much from a big company on the other side of the Atlantic. I cant prove to them that my intentions are charitable, and of course they do have to consider their profits and protect the Intellectual Property, which I respect.

So it looks very unlikely that I will be able to get the display working under its own computers control! My only option then would be to directly address the displays.

Heres a close up of one of the display panels. As far as ive discovered so far, all five panels in the display are identical, and work by cascading serial data along them from the controller.

In many ways the panels are quite straightforward. Two rows of five 8x8 common anode arrays of bicolour LEDs, with Darlington drivers and high current PNP transistors. Where it gets complicated, is the various shift registers involved with selecting the appropriate  rows/columns, etc etc, this is what I am in the process of finding out!

Probably not easy to see, but the above is the sketched out schematic of the shift registers and darlington sink drivers that handle the left hand column pair of arrays. I have papers like this for most of the sections of the board now. The big task will be to consolidate them all, and expand this one to cover the other four columns of arrays. Oh, and then to work out what the data format is to address the LEDs!

No comments: