At 02:47 PM 2/28/2004 -0500, you wrote:
>I would definitely put resistors in series with the Darlington inputs!
>Note the comment just below the schematic that Dakeng makes apologizing
>for any "magic smoke" because the resistors had been omitted on earlier
>schematics.

The ULNxxx has internal resistors, so that comment does not apply.

The problem happens when there is a negative transient on the
ground of the ULNxxx chip relative to ground on the PIC of more than
1.2V, which means that the parasitic isolation substrate diode in
the ULNxxx will conduct, as will the protection diode in the PIC.
The current available is enough to cause the PIC to latch up (typically
in the 100mA region at room temperature, and far above the absolute
maximum 20mA rating.

Bottom line is that you can be a lot sloppier with layout if there
are resistors in there. To get 100mA through a 1K resistor you'd need
a 100V ground transient, which isn't ever going to happen with the slow-
as-molasses ULNxxx drivers.

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 hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.