At 05:50 PM 4/14/2004 +0000, you wrote:
> > To clarify, you'd have to switch the SDO output of the
> > PIC to high-Z (make it an input) and back during
> > transmission/reception to/from this chip in order to
> > avoid using the extra control pin.
>
>So the device (with the common SDI/SDO line) must be
>accessed "half duplex", as opposed to SPI, which clocks
>data in/out on the same pulse?  Is this really SPI then?

Sort-of.

> > Minimalist solution is a resistor from SDI to SDO on the
> > MCU. Value would depend on worst-case bus timing and
> > capacitive loading, but something like 4.7K..
>
>What does the resistor do in this case exactly?  Does it
>prevent a short-circuit in case both the PIC and device
>accidentally drive the line at the same time?

Not really. With normal SPI devices, the SDO output of the PIC
drives the SDI input of the device and vice versa. The output
overpowers the 1mA or so coming from the other bus line (through the
resistor) if they happen to be in opposite states. It wastes a bit of
current, but otherwise has little effect.

When you're using the "half-duplex" SPI part, when it goes tristate
on its I/O pin, then the PIC drives it to the desired state through
the resistor. When its output is low-Z, it overpowers any possible
different logic level coming out of the PIC (so you don't have to
tri-state the PIC output).

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body