Re: Glitches using 8255?

On Sun, Jun 15, 2003 at 04:37:42PM -0400, William T Wilson wrote:
> I am using a National Instruments PCI-6503 for digital I/O, this is a
> fairly simple PCI based 8255 card.  I am curious about the possibility of
> "glitches" using the digital output.  I have port A (channels 0-7)  
> configured for input and the other 16 channels configured for output.
> 
> I use one half of port C (4 lines) to control an address decoder.  I am
> curious if, using comedi_dio_bitfield, it is possible to have a "glitch"  
> in the output such that the lines are in a state that is neither "before"  
> nor "after".  In other words, can I count on this function (and the 8255
> chip) to make all the requested changes (within a single port)
> simultaneously?  Or do I need to disable my address decoder, put the new
> address on the port and then re-enable the decoder?  I know that disabling
> the address decoder would solve the problem for sure, but I do not want to
> do that because I do not want to use another channel on the I/O board.

The comedi_dio_bitfield() function will change all the bits at the
same time on a particular 8255 port, _in the software sense_.  Which
means that they will typically have a jitter on the order of 10 ns
or so.  I don't know if this is adequate for your application or
not -- usually address decoders have a latch for this reason.



dave...

Received on 2003-06-16Z18:13:20