F-25A (Home-Built) Flight Simulator

The F-25A Flight Simulator was a hobby project that involved the development of a mission simulator for a hypothetical tactical fighter of the mid-1990's. The simulator included two fully functional Pilot-Vehicle Interface (PVI) displays -- a Head-Up Display (HUD) and Multi-Function Display (MFD). A stick, throttle, and rudder pedals were also available. A digital sound system, capable of producing eight sounds simultaneously, could drive three speakers -- one forward and two aft of the simulator.

The HUD was fully functional. A monochrome CRT projected onto a mirror inside of the HUD enclosure, and the reflected image was displayed on the HUD glass. The "pilot" would see the HUD flight data projected onto the glass as well as being able to see through the glass, just like a real HUD.

The MFD included a functional monochrome CRT surrounded by 18 selection-buttons and a power-button. Each of these buttons could be read by Cygnus-1 Single Board Computers, which controled the simulator's I/O. The selection-buttons allowed the pilot to select an item that was displayed on the CRT next to each button location, just like a real MFD.

Development on this project was ongoing for several years, but towards the end of the development phase, little further progress was being achieved, primarily due to the general lack of interest in programming the old 8MHz Amiga computers that served as the sound system and the image/symbol generators for the HUD and MFD CRTs.

F-25A Test Pilot Gallery

Click on the images below for more data about each of these brave test pilots:

Amelia "Millie" Fuesz       Daniel "Hopper" Fuesz

Computing Hardware

The simulator included four Amiga computers. Two of these machines, Amiga 1000's (the original Amiga model), drove the HUD and MFD, and the other two machines (A500's) produced sounds. The primary (host) computer was a Dell 466DV. It communicated with the Amiga computers through RS232 serial interfaces, which were routed through a custom-built parallel computer system, based on Cygnus-1 Single Board Computers (SBCs) (shown below). The Cygnus computers were also used for cockpit I/O.

Cygnus-1 Single Board Computer (SBC)

History

This (hobby) project began in the fall of 1991. The initial concept was to create a simple electronic toy instrument panel with a few LEDs, switches, and TTL ICs. This was a small "father/son" weekend project. We then decided to build an "ejection seat" on which to sit while using the instrument panel. At this point, the instrument panel was simply a piece of cardboard, and there was no "cockpit" of any kind.

The idea was to have a few LEDs that would represent landing gear lights, warning lights, etc. that the "pilot" would operate, while the "instructor" could set other switches that would cause "failures" of some sort. This quickly became mundane, even for a five year old.

Next, we expanded the concept of merely having basic TTL control and developed a parallel port interface to an Amiga computer. This greatly expanded what the "instrument panel" could do. In fact, it was clear that it would now be possible to develop a flight simulator, with many forms of input and output.

However, it was soon clear that the Amiga's 8MHz 68000 processor lacked the compute power necessary to simulate aircraft dynamics and perform I/O. It was decided that a parallel computer, based on the Motorola 68000 processor of the Amiga, should be developed. The initial idea was to create a wire-wrapped, parallel computer that would be fully scalable as the fidelity of the simulator increased. In addition, it was expected that additional processors could be added as the number of threat systems increased. This would allow full scalability in the simulated environment.

Two wire-wrapped prototype 68000 boards were developed. Development and manufacture of these boards was very time consuming. Economic analysis revealed that it was possible to have printed circuit boards (PCBs) manufactured that would be comparable in cost to wire-wrapped boards when produced in quantities of four or more. In addition, there was a tremendous savings of time in assembling (soldering) a PCB as compared with wire-wrapping. Work on the layout of a PCB then began, which resulted in Cygnus-1.

While development of Cygnus-1 and its predecessors was being undertaken, costs of PCs continued to drop, while Cygnus-1 costs held steady. As a result, the cost-effectiveness of Cygnus-1's parallel processing approach was eclipsed by 486 and Pentium machines (and later, other Pentium-class derivatives).

In 1994, I purchased a Dell 466DV (66Mhz 80486), which had more computing power (especially for floating point calculations) than a dozen Cygnus computers. As a result, the 466DV became the host computer, and the Cygnus parallel computers served as an RS-232 router and I/O box.

Epilogue

Development on the F-25A was halted in 1998. Programming the old Amiga computers, which lacked hard drives, was tedious. Additionally, the limited resolution (320x200) and slow processing speed of the Amigas, which were now completely obsolete, diminished any interest in developing the simulator any further. Finally, in the spring of 2001, the simulator was disassembled and scrapped.

Image Gallery

Additional images of the F-25A can be viewed here.

Last update: 27-OCT-2001