Searching \ for '[PIC]: Input protection V-I curve' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/ios.htm?key=input
Search entire site for: 'Input protection V-I curve'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: Input protection V-I curve'
2001\08\06@101254 by Bob Ammerman

picon face
If somebody had a little bit of time to work it up, I'd love to see an V-I
curve for various flavors of PIC inputs.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\08\06@103749 by Spehro Pefhany

picon face
At 10:08 AM 8/6/01 -0400, you wrote:
>If somebody had a little bit of time to work it up, I'd love to see an V-I
>curve for various flavors of PIC inputs.

It should be just a forward biased silicon diode for all but the oddball
ones without protectin diode (Vpp) ... is that what you want to see?  Or
confirm?

Best regards,

Best regards,

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spam_OUTspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\08\06@112732 by Dan Michaels

flavicon
face
Bob.A wrote:
>If somebody had a little bit of time to work it up, I'd love to see an V-I
>curve for various flavors of PIC inputs.
>

Not done this, but in qualification [ie, stress] testing, I have
run in up to 80 mA without popping the internal diodes.

- dan
===========

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\08\06@172209 by Bob Ammerman

picon face
I would like to see a careful measurement of it in the regions of interest
(below Vss and above Vdd):

to see just where the threshold really is,

how hard the knee is,

etc.

Maybe we can learn a little bit more by doing this.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)




{Original Message removed}

2001\08\06@192742 by Spehro Pefhany

picon face
part 1 1321 bytes content-type:text/plain; charset="us-ascii"At 02:28 PM 8/6/01 -0400, you wrote:
>I would like to see a careful measurement of it in the regions of interest
>(below Vss and above Vdd):

Ok, Bob:-

-I did some measurements above Vdd on a single piece of 12C509JW
-Programmed it to run on all internal and to just sit there with all inputs.
-Connected all pins but power and GP0 to Vss through a 500 ohm resistor.
-Vdd 5.00V

- applied voltage above Vdd through a 100R 1% resistor.

Current       Voltage
(mA)          (V, relative to Vdd)
0.020         0.570
0.050         0.601
0.100         0.625
0.200         0.649
0.500         0.677
1.00          0.699
2.00          0.723
5.00          0.762
10.0          0.807
20.0          0.887

I've attached a gif graph of the results.
A typical 1N4148 diode data sheet is at:
http://www.fairchildsemi.com/ds/1N/1N4148.pdf

- At higher currents, the ratio of currents between the data sheet 1N4148
 and the PIC is about 3:1.

This doesn't seem to change the conclusions that either a series resistor or
a Schottky diode is required in order to be very effective- a 1N4148 with no
resistor should improve the minimum current that would cause latch-up by only
about a factor of 3:1. A small resistor (even 100R) in series with the diode
would improve this MUCH more (more than 500mA in this case).

Best regards,



part 2 4836 bytes content-type:image/gif; name="inputchr.gif"; (decode)


part 3 490 bytes content-type:text/plain; charset="us-ascii"

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..."            "The Journey is the reward"
.....speffKILLspamspam@spam@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


part 4 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\08\06@195806 by Bob Ammerman

picon face
----- Original Message -----
From: "Spehro Pefhany" <speffspamKILLspamINTERLOG.COM>
To: <.....PICLISTKILLspamspam.....MITVMA.MIT.EDU>
Sent: Monday, August 06, 2001 7:10 PM
Subject: Re: [PIC]: Input protection V-I curve


> At 02:28 PM 8/6/01 -0400, you wrote:
> >I would like to see a careful measurement of it in the regions of
interest
> >(below Vss and above Vdd):
>
> Ok, Bob:-
>
> -I did some measurements above Vdd on a single piece of 12C509JW
> -Programmed it to run on all internal and to just sit there with all
inputs.
> -Connected all pins but power and GP0 to Vss through a 500 ohm resistor.
> -Vdd 5.00V

I am assuming the voltages recorded were measured at the PIC pin and not on
the other side of the 100R.

It sure looks like a typical Si P-N junction to me.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\08\07@025504 by Roman Black

flavicon
face
Hi Bob, i'll do some tests if nobody else voluteers.
:o)
-Roman

Bob Ammerman wrote:
{Quote hidden}

> {Original Message removed}

2001\08\07@033248 by Roman Black

flavicon
face
Thanks Spehro!! That's absolutely excellent info!

From the results of your test on the PIC clamp diodes
they look VERY fragile. I consider the in4148 to be
too fragile, from the amount of them i've found faulty
in TVs and VCRs even in low voltage low current circuits.

But the PIC clamp diodes are poitively tiny based on
your V/I performance chart... I'll stick with my RCZ
networks on PIC inputs I think. :o)
-Roman

PS. Based on their max 20mA spec, and your measured
0.887v, Microchip seem to be speccing the clamp diodes
at max of 18mW... Death to the infidels. ;o)


Spehro Pefhany wrote:
{Quote hidden}

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


2001\08\07@052321 by Spehro Pefhany

picon face
At 05:32 PM 8/7/01 +1000, you wrote:

>PS. Based on their max 20mA spec, and your measured
>0.887v, Microchip seem to be speccing the clamp diodes
>at max of 18mW... Death to the infidels. ;o)

Remember that the diodes may work perfectly at 20mA and at
100mA, BUT the current injected due to the diode action
can trigger the BIG PARASITIC SUICIDE SCR (tm) that lives
inside every PIC (connected from Vdd to Vss) just
_waiting_ for a "gate" pulse to cause it to TURN ON AND
KILL THE PIC (sufficient power supply current being
available).

This is the reason not to exceed the data sheet limits
(and de BIG PARASITIC SUICIDE SCR, he like to turn on
even better when it be hot, hot, hot*).

Microchip have done a fine job in guaranteeing the 20mA
minimum current, IMHO, early CMOS circuits would
practically do this if you looked at them sideways too hard.

* Caribana weekend, mon.

Best regards,


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..."            "The Journey is the reward"
speffspamspam_OUTinterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

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


2001\08\07@114138 by Douglas Butler

flavicon
face
       This reminds me of a guy I talked to who was trying to put a Pentium II
in an oceanographic probe.  He insisted that to get a "good" processor
reset it was not enough to ground the /reset pin.  He connected the
/reset pin to a negative current source with a Schottkey clamp to pull
the pin below ground and flood the die with carriers to avoid a
"marginal" processor reset.
       I though he was nuts, but I have never worked with anything bigger than
an 8086.

Sherpa Doug

> {Original Message removed}

2001\08\07@121239 by Jeff DeMaagd

flavicon
face
----- Original Message -----
From: Douglas Butler <@spam@dbutlerKILLspamspamIMETRIX.COM>


>         This reminds me of a guy I talked to who was trying to put a
Pentium II
> in an oceanographic probe.  He insisted that to get a "good" processor
> reset it was not enough to ground the /reset pin.  He connected the
> /reset pin to a negative current source with a Schottkey clamp to pull
> the pin below ground and flood the die with carriers to avoid a
> "marginal" processor reset.

>         I though he was nuts, but I have never worked with anything bigger
than
> an 8086.

I would likely agree with you.  I'd want to see the data sheet before ever
trying anything like that.  While no data sheet is perfect, I'd be more
likely to trust the manufacturer than some guy trying to get a "perfect"
reset that way.

Jeff

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


2001\08\07@122838 by David VanHorn

flavicon
face
Something else that needs to be said here.

Those protection diodes dump to VCC.
Your VCC rail needs to be able to deal with that.
78XX regulators are series pass devices.
If the current in through the protection diodes exceeds the drain of the
rest of the system from the 5V rail, then the 78XX will shut down, and the
5V rail will rise, until the drain equals the input current.
This may take your 5V rail out of spec.

If you're going to dump ESD through these (or any other) diodes, then the
5V rail needs to be able to absorb that transient.  Your 5V rail's
impedance is NOT zero.

I've seen a very similar mistake done on a thermal printer.
The printhead is driven from a darlington array, with protection diodes,
that are tied to VCC.
(Don't ask me why, resistive elements seldom provide much "kick" when
turned off...)
Unfortunately, the printhead is powered from an unregulated supply, usually
about 9-12VDC
The VCC rail usually measures 5.24V on a normal AC line, and the 7805
regulator is cold.
When the printer prints though, it grounds the printhead through the
darlington array, and this makes the 7805 pop on and off during printing.

For some reason, they have a big problem with this printer crashing, and
burning up printheads.
The print head elements can only take a couple mS on-time before smoking.



--
Dave's Engineering Page: http://www.dvanhorn.org

I would have a link to http://www.findu.com/cgi-bin/find.cgi?KC6ETE-9 here
in my signature line, but due to the inability of sysadmins at TELOCITY to
differentiate a signature line from the text of an email, I am forbidden to
have it.

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


2001\08\07@141015 by Spehro Pefhany

picon face
At 11:28 AM 8/7/01 -0500, you wrote:
>Something else that needs to be said here.
>
>Those protection diodes dump to VCC.
>Your VCC rail needs to be able to deal with that.

Good point. Unstated in my described test circuit was a 180 ohm 1/4W
resistor to sink 27mA to allow current to flow into Vdd without
increasing the voltage.

Pretty much *all* regulators behave this way.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany --"it's the network..."            "The Journey is the reward"
KILLspamspeffKILLspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
Contributions invited->The AVR-gcc FAQ is at: http://www.bluecollarlinux.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

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


2001\08\07@151801 by David VanHorn

flavicon
face
At 01:53 PM 8/7/01 -0400, Spehro Pefhany wrote:
>At 11:28 AM 8/7/01 -0500, you wrote:
> >Something else that needs to be said here.
> >
> >Those protection diodes dump to VCC.
> >Your VCC rail needs to be able to deal with that.
>
>Good point. Unstated in my described test circuit was a 180 ohm 1/4W
>resistor to sink 27mA to allow current to flow into Vdd without
>increasing the voltage.
>
>Pretty much *all* regulators behave this way.

I have a lot of fun in analog land, looking at things like that.
Like the fact that all your noise current on VCC ends up on the input lead
of your linear reg, but not on switchers, unless it's significantly below
the switching frequency. The SMPS looks like an integrating A/D converter
(sort of) in the reverse direction.

Once, I needed a two terminal device, with 600 ohms impedance in one
direction, and as close as possible to infinity in the other direction. I
was able to do it.

I can almost remember how. :)  It was a pretty unique situation.

--
Dave's Engineering Page: http://www.dvanhorn.org

I would have a link to http://www.findu.com/cgi-bin/find.cgi?KC6ETE-9 here
in my signature line, but due to the inability of sysadmins at TELOCITY to
differentiate a signature line from the text of an email, I am forbidden to
have it.

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


More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...