Em 8/4/2010 21:47, Harold Hallikainen escreveu:
> I have an SPI clock that leaves a PIC and goes to several devices on a
> board. Part way down the line are some SPI flash chips that can run at
> 75MHz. They really do not like reflections on the clock. We're trying
> various termination methods at the far end of the line with the PIC pin
> driving the line directly. This is a PIC32 which has a pin current limit
> of 4mA. If we terminate with a pull-up/pull-down, each 1k, we'd have 3.3mA
> source or sink. But, the Thevenin resistance seen by the line is 500 ohms.
> Though it can certainly vary, I've thought the characteristic impedance of
> a trace on a multilayer board was in the 100 ohm area. This 500 ohms would
> be a bit high.
>
> We've also tried an RC combination, but, since the SPI clock spends most
> of its time low, the RC is very similar to a termination to ground. The
> PIC has to source current, but not sink it. This requires the termination
> resistance to be even higher to avoid overcurrent on the pin.
>
> So... what are people doing to terminate lines like this?
>
> Thanks!
>
> Harold
>   =



Use series termination on the PIC side of the line (just a resistor
between the PIC's pin and the line), this way you reduce the slew-rate
of the signal and avoid reflections.


Best regards,

isaac

__________________________________________________
Fale com seus amigos  de gra=E7a com o novo Yahoo! Messenger =

http://br.messenger.yahoo.com/ =


-- =

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