On Mon, 03 Apr 2006 17:58:28 +0100, you wrote:

>In message <59b85d8f05969c249e60a120e09bab02@mac.com>
>          William "Chops" Westfield <westfw@mac.com> wrote:
>
>> I wonder if anyone will ever sell an open-source oscilloscope?
>> Don't like the way the SW works?  Fix it yourself, or download
>> the "unstable with enhanced extra knobs" version from your favorite
>> repository.  Since a lot of the high end scopes these days seem to
>> be essentially PC clones as the Display/UI, it's not unthinkable.
>
>I'd love to do something like that. I can do the software and probably the
>digital hardware, but I have absolutely no idea how to go about designing
>high-bandwidth low-noise amplifiers and such for the input circuitry.
>
>A DSO isn't *that* complex - when you boil it down to its essentials, you've
>got:
>  - Front end amplifier (1 per channel)
>  - A/D converter / acquisition circuitry (1 per channel)
>  - Heavily filtered low-noise power supply
>  - High speed RAM (SDRAM these days, though 10nS SRAM would be quite
>    suitable if you could live with the relatively low data density)
>  - Acquisition controller
>  - Trigger circuitry (or just leave the ADC acquiring all the time and
>    implement this in the acq controller)
>  - CPU interface (PCI?)
>  - CPU, display and UI (PC?)
>  
>It would certainly be fun to build something that gave Tek's low- to
>mid-range scopes a run for their money :)
>
>Anyone want to join me? :P

It gets more interesting when you want a more responsive acquisition-to-display interface though
(Tek DPO etc.) Also, SDRAM only gives decent performance in bursts, so a SRAM buffer would also be
necessary. Both of these could be done in a not-too expensive FPGA. 

This would be a good base to start with perhaps : 
http://www.enterpoint.co.uk/moelbryn/raggedstone1.html

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist