Issue 102 January 1999
In the Face of Medusa
Part 1:Developing Reliable Control

THE CENTRAL SITE

Now that the module’s tasks have been defined, let’s focus on the ’x86 system. Using the 4Q peripheral processors means I can choose a low-cost ’386 system. My choice is the Octagon Systems 4010.

The 4010 is an 80C386CX running at 25 MHz. It has the standard AT-compatible BIOS and uses standard PC-type peripherals like floppy and hard drives. Ideally, the fewer moving parts the better, and the 4010 is equipped perfectly. There’s 512 KB of flash memory and 2 MB of DRAM, which can operate just like their mechanical cousins.

And, just in case they’re needed, the 4010 can also accommodate a 2.5² hard disk or standard 3.5² floppy drive.

The serial ports of the 4010 will be busy handling the peripheral processors, so a means of communicating the collected data, other than the standard serial ports, is necessary. I just happen to have an Octagon 5500 Ethernet card that plugs into the 4010 card cage to transfer data out of the Medusa experiment.

THE COUNTDOWN

We have some one-of-a-kind hardware fresh from the lab, along with some powerful little devices called PicStics. The idea is to allow the PicStics to operate under control of an ’386 embedded-processor complex. Additionally, the PicStics can take over their respective roles if the ’386 system is busy doing other things.

Failure of any component during the experiment time is unacceptable. So, the hardware I chose is robust and reliable. The 4010 has an MTBF of 11 years. I’ve worked with Microchip PICs for many years now and can count on one hand the non-customer-induced failures I’ve seen.

The system concept is described and the jobs are well defined. The next step is integrating the ’386 and the Medusa experiment hardware with the peripheral processors. Once that’s accomplished, the Ethernet conduit must be effected so the data can be put onto a more meaningful platform for the scientists to evaluate (e.g., an Ethernet-capable laptop).

Next time, I’ll show you how to integrate the PicStic-4Qs and the Octagon Systems 4010 into a synchronous system to support the requirements of the Medusa experiment. I’ll also take you through the steps to effect an Ethernet interface between the 4010 and a scientist’s laptop.

I realize this article has been off the beaten path as far as some folks’ definition of embedded is concerned, but I doubt that you’re a typical cookie-cutter professional. The 4Q is a good fit for this application and proves again that it doesn’t have to be complicated to be embedded.