Searching \ for '16F84 survives reverse polarity' 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/devices.htm?key=16F
Search entire site for: '16F84 survives reverse polarity'.

Truncated match.
PICList Thread
'16F84 survives reverse polarity'
1998\07\22@050504 by org Hager

flavicon
face
Hi all!

Yestarday I've made an experience that I just have to share. In a state of
being tremendously overtired, I decided to work on a little serial LCD
interface project that had been lying around for a couple of days. To
check if everything was still OK, I hooked it up to my lab power supply
and nothing happened (there was supposed to be a message on the display).
My heart stopped for a moment when I discovered that I had connected the
power supply with the wrong polarity. Afer disconnecting, I noticed that
the PIC had gotten *really* hot. The LCD showed no sign of impression.

Amazingly, the thing worked OK after I connected the supply with the
correct polarity again. I always thought that reverse polarity was poison
for LSIs (and not only for them)... FYI: The power supply uses a 5V, 5A
regulator (78HGASC), of which I don't know the short-cicuit current, but
it's surely in the several 100mA range. The PIC as well as the LCD module
seem to be all right again.

Might this have reduced the expected lifetime, EEPROM reliability etc. of
the parts? I have since sucessfully reprogrammed the PIC several times.

Georg.

1998\07\22@115414 by Reginald Neale

flavicon
face
Georg said:

>My heart stopped for a moment when I discovered that I had connected the
>power supply with the wrong polarity. Afer disconnecting, I noticed that
>the PIC had gotten *really* hot. The LCD showed no sign of impression.
>
>Amazingly, the thing worked OK after I connected the supply with the
>correct polarity again. I always thought that reverse polarity was poison
>for LSIs (and not only for them)... FYI: The power supply uses a 5V, 5A
>regulator (78HGASC), of which I don't know the short-cicuit current, but
>it's surely in the several 100mA range. The PIC as well as the LCD module
>seem to be all right again.
>
>Might this have reduced the expected lifetime, EEPROM reliability etc. of
>the parts? I have since sucessfully reprogrammed the PIC several times.
>

Georg:

Of course there are no guarantees, but PICs are famous for taking this kind
licking and keeping on ticking. I had a 16C54/JW part that got reverse
connected and I swear I could see a dim red glow coming through the window.
After it cooled down it seemed back to normal!

Reg Neale

1998\07\22@125443 by Peter L. Peres

picon face
> 5V 5A

May I ask, what kind of wires you have used, and how long they were ? I
have troubles with low voltage / medium amp equipment all the time as the
bench PSU has 70 cm wires on it and I have to use 2.5 mm^2 cables to avoid
voltage drop. 'Usual' crocodile wires (the kind bought in bunches) uses
19x0.2 strands and is not even soldered. It is 'safe' to connect this to a
sizeable bench PSU without anything very bad happensing (okay, it smells,
but...).

PS: The part that goes really hot in a low power LSI that is reverse
connected, seems to be the bonding wire. Since I work with CCDs and other
windowed electro-optical devices I can actually see the bonding wires glow
red somtimes (I don't do it on purpose, I'm in the repair business and it
sometimes happens that I get units with weird faults...). I can tell from
experience, that even oldie EPROMs in CERDIP withstand such abuse for
several seconds and show no problems later. The key seems to be the ultra
thin chip bonding wire that starts to glow and thus limits the current
through the bulk diode.

Peter

1998\07\22@161132 by Mike Ghormley

flavicon
face
Georg Hager wrote:

<SNIP>
> I hooked it up to my lab power supply
> and nothing happened (there was supposed to be a message on the display).
> My heart stopped for a moment when I discovered that I had connected the
> power supply with the wrong polarity. <SNIP>
> Amazingly, the thing worked OK after I connected the supply with the
> correct polarity again. I always thought that reverse polarity was poison
> for LSIs (and not only for them)...

I had a similar experience when one of my students hooked up the PS on a protobo
ard with
a 12C509JW.  It was connected backwards for at least a minute -- probably a low
current
from the student's PS, though.  Still, very impressive!

The part has been used over and over again since without the slightest sign of a
ny
problem.

Michael

*************************************************************************
When the way of the Tao is forgotten, kindness and ethics must be taught.
Men must learn to pretend to be wise and good.  --  Lao Tzu
*************************************************************************

1998\07\23@033112 by paulb
flavicon
face
Peter L. Peres wrote:

> I can tell from experience, that even oldie EPROMs in CERDIP withstand
> such abuse for several seconds and show no problems later.

 Sadly, my experience has been to build/ own equipment with well-built
power supplies and busses.  I do have a few EPROMs with the pin 1 (or
28 or something like that - the power wire) nicely interrupted with
little sherical "balls" on each side of the interruption.

--
 Cheers,
       Paul B.

1998\07\23@033319 by org Hager

flavicon
face
On Wed, 22 Jul 1998, Peter L. Peres wrote:

> > 5V 5A
>
> May I ask, what kind of wires you have used, and how long they were ? I
> have troubles with low voltage / medium amp equipment all the time as the
> bench PSU has 70 cm wires on it and I have to use 2.5 mm^2 cables to avoid
> voltage drop. 'Usual' crocodile wires (the kind bought in bunches) uses
> 19x0.2 strands and is not even soldered. It is 'safe' to connect this to a
> sizeable bench PSU without anything very bad happensing (okay, it smells,
> but...).

I use very short (about 5cm) crocodile wires, but this is not a problem as
the whole circuit needs about 4.5mA only. When the thing is running as I
want it to, I will of course design a small, dedicated supply for it.

Long cables and crocodile connectors are really a problem when connecting
more 'demanding' circuits like euroboards full of stuff which generates
lots of noise on the supply lines. It has been my experience that a large
amount of time when building such things goes into proper power supply
design. Hooking it up to a usual lab supply with crocodile (or similar)
connectors and long wires will often lead to problems.

Georg.

1998\07\23@064729 by Caisson

flavicon
face
> Van: Georg Hager <spam_OUTghagerTakeThisOuTspamtheorie3.physik.uni-erlangen.de>> > Aan: .....PICLISTKILLspamspam@spam@MITVMA.MIT.EDU
> Onderwerp: 16F84 survives reverse polarity
> Datum: woensdag 22 juli 1998 10:52
>
> Hi all!

[Cut]

> My heart stopped for a moment when I discovered that I had connected the
> power supply with the wrong polarity. After disconnecting, I noticed that
> the PIC had gotten *really* hot.

I shifted a 16C74A one pin (was on a bread-board). I noticed the fault
because something was emiting light behind the EPROM-window.  Quite bright
too !  I cut the power, shifted it back to it's correct position & supplied
current again.  The thing worked as if nothing ordinary had happened.  It
still does.  Amazing what beating those things can withstand.

By the way, I reversed polarity on a LCDisplay too,  it became hot, but
seemed not to have suffered any ill effects.  Maybe they design those chips
with people like us in mind :-)

Greetz,
 Rudy Wieser

1998\07\23@095814 by Alberto Smulders

flavicon
You might put a DPDT switch into your proyects for power reversal, and so
come up with a new series of interesting stuff, e.g.:

"Tamagotchi with embedded hand warming device" (especially for places where
winters are severe)

or (remembering the cows a time ago.....)

"Milk machine controller with integrated stable heating device"

The possible applications are only limited by your imagination
.......:))))))))))

Regards,

Alberto Smulders
InSAD - Encarnacion, Paraguay
insadspamKILLspamitacom.com.py

1998\07\23@101558 by myke predko

flavicon
face
Reg Neale wrote:
>Of course there are no guarantees, but PICs are famous for taking this kind
>licking and keeping on ticking. I had a 16C54/JW part that got reverse
>connected and I swear I could see a dim red glow coming through the window.
>After it cooled down it seemed back to normal!

<VBG>  This happened for me with a 2716 when I was working as a student and
told to program a bunch of parts.  The engineer put the parts on a piece of
conductive foam and told me to be very careful to not change the polarity;
he had built the programmer himself and told me not to worry about how it
worked, just follow his instructions.

So, I sorted the parts into two piles, ones where the little light went on
and ones where it didn't (they were about equal).

I was amazed that the NMOS 2716s that were reversed still worked after this.

The PICMicros and pretty amazing beasts, they will take this repeatedly (I
think just about everybody has a story about how they put them in backwards,
sometimes the lettering on top of the package was burned off) and still work.

myke

Twenty nine years ago this week, Neil Armstrong walked on the moon.  Learn
more about it in myke's Book Room,

http://www.myke.com/Book_Room

1998\07\23@111753 by Matt Bonner

flavicon
face
I've also been lucky reversing microprocessors and controllers.  Other
not-so-lucky experiences include:

 1.  4000-series CMOS (can't remember which device).  Went off like a
     firecracker, blowing the die and a chunk of plastic up to the
     ceiling.

 2.  2716 UV EPROM.  I found that this was a fast way to erase them
     (to all FFs).  Unfortunately I could never reprogram them.

--Matt

1998\07\23@122317 by Peter L. Peres

picon face
On Thu, 23 Jul 1998, Caisson wrote:

> with people like us in mind :-)

No, actually, it's a measure of how miserly the power grid connections are
on the silicon. All 'cheap' low power chips use one bonding wire per power
pin, the same as for any other pin. High power chips and high speed and
high quality, all use several bonding wires. Now, if you put THAT in the
wrong way... High speed designs, VFET/HEXFET hybrids etc, use up to 6
bonding wires per power pin. That is guaranteed no to glow in the dark,
the wafer does instead.

Peter

1998\07\25@030502 by Josef Hanzal

flavicon
face
All the time I was curious, what REVERSE engineering means. Now I understand !

Thanks, Matt, myke and others for explanation.

Josef

{Quote hidden}

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...