*****************DON'T LOOK AT AN521 !!!!!!!!!!!!!!!!!!!!!

I am very surprised Microchip still put AN521 on their website
without putting a ***ALL CAPITAL LETTER SAFETY DISCLAIM***.=20
The AN was written in 1997 and I am not so sure if the
engineer (Doug Cox) is still with Microchip or not.

It is true that the author mentioned that "The least expensive=20
method to limit current is to use a high value resistor". However
he forget to mention to that "the most expensive method to
destroy a design is to cause safety accident by using a=20
least expensive resistor to connect a PIC pin to main line".

I am also very suppressed that the author dared to mention
"RELIABILITY" on the application note.

The author wrote in the AN521 the following.
"A short is the most unlikely failure mode for a resistor,
and resistors are more reliable than transformers or
capacitors, which are the alternate components for
measuring line parameters. This reliability can be
enhanced even further by using two resistors in series.
Both would have to fail short to cause catastrophic
failure, a very unlikely event.

The second possible failure mode is that excessive
injection of current into the PIC16C5X might cause the
protection diode to open. This would allow the input to
go to power line peak voltage (162V) and short the
input transistor gate oxide, causing device failure. The
maximum continuous injection current into an I/O pin is
specified at =B1 500uA."

So the author does not need to care for "catastrophic
failure" since it is a "very unlikely event"? How
"very unlikely event' is "very unlikely"?

Even if reliability and safety are not of concern
by the author (can he ???), he forgot to mention that
a 5Meg Resistor can be short or open depending on=20
the resistors chosen. A metal film or wire-winding
resistors are less likely to be short when failed.
Ceramic based resistors are more likely to be in any
status (including short) when failed. The creepage
distance and clearance need to be kept as well.

I think there is a similar discussion in the Microchip
Forum recently as well.=20

Regards,
Xiaofan

-----Original Message-----
From: piclist-bounces@mit.edu=20
Sent: Thursday, October 27, 2005 3:37 AM
To: Microcontroller discussion list - Public.
Subject: Re: [EE] Inputs over Vdd / under Vss with
current-limitresistors... ? and "input clamp current" spec?

Thanks Mark!  Just what I needed.
J

Mark Scoville wrote:

> Hi Jesse,
>=20
> See Microchip app note AN521
>=20
> -- Mark

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist