Bozos, all I did to control the current was fuse the project fuse the project with a 900mA fuse (I know, it's an uncommon rating) and make sure that all components's maximum current requirements and current used to power an attached device(plus 15% saftey buffer) did not max out past 900mA. The problem I had here was that the circuitry was supposed to only draw 300mA max. I struggled to find ways to reduce components which in turn reduced power consumption.
The transfer rate was a little trickier. I used a PIC18F452 that was run off a 20MHz oscillator (max speed that PIC can handle). The USART that the serial signals went into on the PIC was configured for 19.2k baud--that was the easy part. The USB side required me to create a working protocol that was faster than the transfer rate, which was then delayed in the software so that it performed at 19.2k baud. The problem that I had was keeping my USB protocol efficient enough so that I could get the 19.2k baud transfer rate.
Bungleau, as far as patenting and all that, there are already other (and better) devices out there that do what my project does. Most people who produce this product do not use a PIC to translate the signals because it is too inefficient and slow. I would be hard-pressed to find anyone who would like to buy the rights to this device. If someone would though, it would be happy land for me! [img]smile.gif[/img]
__________________
I am a peon.
|