Searching \ for '[EE]: Killed another PIC! Why!' 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=pic
Search entire site for: 'Killed another PIC! Why!'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Killed another PIC! Why!'
2001\10\14@084427 by Dennis Noordsij

flavicon
face
Heips,

This is the second time the following happened, it's getting kinda
frustrating!

I have a circuit on a breadboard, a PIC(16F877), a MAX232 and support
components, some LED's, etc.

The power comes from a 7805 regulator with a source of 12 volts, this is the
only component on a veroboard rather than a breadboard. The regulator can go
up to 1.5 amps if properly heatsinked.

For some reason (just my luck I guess!) every veroboard I use manages to
attract mysterious particles from the air and create a short between 2
tracks, even though the solder doesn't even take up the whole width of the
track. I end up butchering the groove until it is wider. (thats why I
switched to breadboards, saved me lots of time! :)

Anyways, in this case it created a short between the 5V and GND tracks. The
regulator is heatsinked (well isolated though) and got hot quite fast :-) It
only took a second before I had disconnected it again, but at that point the
PIC was burning hot at both VDD-VSS pin locations, and consequently the
resistance when measured between VDD and VSS pins is ZERO. I might as well
use a paper clip in the circuit rather than the PIC after this "incident" :(

But I'm wondering what it is exactly that kills the PIC in this case.
Shouldn't the current simply go through the short (which is at the beginning
of the circuit, way before the power lines reach the other components)? Why
did the LED's and MAX232 survive fine but the PIC died?

And more importantly, what is the easiest way to build protection against
this into the circuit? (other than fuses which I guess would do the trick :)

Just wondering :)
Regards,
Dennis

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@094430 by Lasse Madsen

flavicon
face
As you probertly know the 7805 is destined to get wery_hot as you are
dissipating ALOT of wattage

lets do a simple calculation
you are drawing 500mA in your circuit ...

12V - 5V  * 500mA = 3,5W - That's why you DO need a heatsink and why the
7805 can (and properly will) get wery hot.

Maybe 500mA is a bit "over calculated" but it shows the effect that the
power loss dissipation has on the regulator.

You could protect your circuit from over-voltage with a zener diode and a
resistor. (which shouldn't happen because the regulator has a shutdown unit
but  - you cant be to safe :)

You could also protect your circuit from over-current with a transistor.

Try measuring with your scope (if you have one) the Ripple voltage on the
output of the regulator when its pretty hot. try adding a large electrolytic
capacitor to reduce the ripple ... ripple can kill a pic ...

You could also use a Driver IC to supply the PIC if anything is killed it
would be the Driver IC

You could also add a 1K resistor to protect against overcurrent.

CHECK if you have reversed the supply pins ! ! - THIS WILL KILL THE PIC !

Hope you will get some ideas from this.

Best Regards
lasse madsen

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@094442 by Lasse Madsen

flavicon
face
Just a quick question:

Have you remembered to power the pic from both its supply rails

Vdd = 11 , 32
Vss = 12, 31

Where Vss is Minus
and Vdd is Plus

Best regards
lasse madsen

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@095057 by Dennis Noordsij

flavicon
face
On Sunday 14 October 2001 16:43, you wrote:
> Just a quick question:
>
> Have you remembered to power the pic from both its supply rails
>
> Vdd = 11 , 32
> Vss = 12, 31
>
> Where Vss is Minus
> and Vdd is Plus
>

Yup. And the circuit has been working fine UNTIL the short :-) Regulator does
have 0.1 uF caps against ripple. Supply pins are the right way around, and
both VDD and VSS pins are connected.

But, even though I understand the issues involved with the regulator etc, I
am wondering why it killed the PIC but not the other components. Even though
lots of current flows through the short, it shouldn't flow through the PIC ?
What kills it?

Dennis

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@100548 by Roman Black

flavicon
face
Dennis Noordsij wrote:

> But, even though I understand the issues involved with the regulator etc, I
> am wondering why it killed the PIC but not the other components. Even though
> lots of current flows through the short, it shouldn't flow through the PIC ?
> What kills it?


Most probably because you wired it wrong on
the breadboard and the current went a different
path to what you thought.

Absolute best way for "quickie" circuits is
to use a 220ohm resistor between 12v and the
7805 "input" pin. This will allow about 20mA to
25mA to power your circuit, and even in the
event of really faulty wiring will limit max
current to about 50mA which is fine into any
PIC pin for short periods. You should never
blow another PIC from faulty wiring. :o)
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@101816 by Dennis Noordsij

flavicon
face
On Sunday 14 October 2001 17:01, you wrote:
> Dennis Noordsij wrote:
> > But, even though I understand the issues involved with the regulator etc,
> > I am wondering why it killed the PIC but not the other components. Even
> > though lots of current flows through the short, it shouldn't flow through
> > the PIC ? What kills it?
>
> Most probably because you wired it wrong on
> the breadboard and the current went a different
> path to what you thought.

But I didn't wire it wrong. The circuit on the breadboard uses the 5V and GND
rails, which are connected from the board to the veroboard where the
regulator is. Everything was fine until 2 tracks on the veroboard (the 5V and
GND ones, not the 12V which doesn't go past the regulator) shorted and THEN
the PIC died. Everything was fine until things blew up, without anyone
touching anything on either board.

So what you're saying is that this is basically impossible and maybe the 12 V
shorted to GND causing totally different effects (+12 on GND and +5 on well,
+5 :) I'm saying that from what I've seen all that happens is a short between
+5 and GND, but if you tell me that's not enough to kill the PIC and
something else HAD to happen then I am not going to argue that :)

> Absolute best way for "quickie" circuits is
> to use a 220ohm resistor between 12v and the
> 7805 "input" pin. This will allow about 20mA to
> 25mA to power your circuit, and even in the
> event of really faulty wiring will limit max
> current to about 50mA which is fine into any
> PIC pin for short periods. You should never
> blow another PIC from faulty wiring. :o)
> -Roman

Will most definately do :-)

Regards,
Dennis

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@172849 by Andy N1YEW

picon face
I know I know!

You blew out the internal aluminum 'wires' that connect power to the chip =\

been there seen that =T
andy
{Original Message removed}

2001\10\14@174116 by Tim Hamel

picon face
In a message dated 10/14/2001 2:29:40 PM Pacific Daylight Time,
spam_OUTn1yewTakeThisOuTspamSOFTHOME.NET writes:


{Quote hidden}

This happened to me once too. I used alligator clips to connect my
7805-on-a-perfboard regulator to my breadboard. One time, the clips shorted
(5V-GND) and the PIC died. From then on, if I powered it up, it got REAL hot.
The PIC obviously doesn't like it's rails tied together.

-Tim

P.S.
If anyone cares, I'll switch my subscription to my other email so you guys
don't have to put up with AOL's HTML crap.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@180015 by David VanHorn

flavicon
face
>
>This happened to me once too. I used alligator clips to connect my
>7805-on-a-perfboard regulator to my breadboard. One time, the clips shorted
>(5V-GND) and the PIC died. From then on, if I powered it up, it got REAL hot.
>The PIC obviously doesn't like it's rails tied together.

Crowbarring the supply rails, by itself, isn't going to cause any problems.
However, if you have charged caps connected to I/O pins, those will
discharge back through the protection diodes, and put the chip into SCR
latchup, although the latchup should clear when the chip cools.

Point being, you are allowed to crowbar the supply, but other things you
have hung on the chip may cause problems in that case.
--
Dave's Engineering Page: http://www.dvanhorn.org

Got a need to read Bar codes?  http://www.barcodechip.com
Bi-directional read of UPC-A, UPC-E, EAN-8, EAN-13, JAN, and Bookland, with
two or five digit supplemental codes, in an 8 pin chip, with NO external parts.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@184421 by Tony Nixon

flavicon
picon face
Dennis Noordsij wrote:
>
> Anyways, in this case it created a short between the 5V and GND tracks.

> But I'm wondering what it is exactly that kills the PIC in this case.

The high current is flowing around the circuit somewhere. Maybe the
PIC's VCC to VDD rails or IO pins just happen to be in, or become part
of, that shorted path.

--
Best regards

Tony

mICros
http://www.bubblesoftonline.com
.....salesKILLspamspam@spam@bubblesoftonline.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@201733 by Alexandre Domingos F. Souza

flavicon
face
       Dear bin laden of PICworld, ;o)

>Anyways, in this case it created a short between the 5V and GND tracks. The
>regulator is heatsinked (well isolated though) and got hot quite fast :-) It
>only took a second before I had disconnected it again, but at that point the
>PIC was burning hot at both VDD-VSS pin locations, and consequently the
>resistance when measured between VDD and VSS pins is ZERO. I might as well
>use a paper clip in the circuit rather than the PIC after this "incident" :(

       Wow! You got a real cheap way to pic something, eh! ;o)

>But I'm wondering what it is exactly that kills the PIC in this case.
>Shouldn't the current simply go through the short (which is at the beginning
>of the circuit, way before the power lines reach the other components)? Why
>did the LED's and MAX232 survive fine but the PIC died?

       Very interesting question. I think what fried your PIC was not the short, but the fast rise and fall of the +vcc rail, in a short circuit. 7805 has a small period of no-regulation, that's small enough to rise vcc to 7, 8 volts and not fry the circuit. Since your veroboard is getting particles "from the air", it creates an intermitent short, that makes the regulator turn on and off very fast. Strange but true

       Solutions:
       1 - change your supplier of veroboard. Humidty in the board can cause it. Is it fiberglass or fenolite?
       2 - build a better Power Supply.
       3 - What you have in your PS along the 7805? Can we see schematics of what you did?
       4 - Call that small woman from "poltergeist" to "decontaminate" your protoboard and say "this protoboard...is clean!"
       5 - pray

       Anyway, keep trying, everyone has fried a pic before during learning ;o)


---8<---Corte aqui---8<----

Alexandre Souza
taitospamKILLspamterra.com.br
http://planeta.terra.com.br/lazer/pinball/

---8<---Corte aqui---8<----

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\14@210616 by Olin Lathrop

face picon face
> Yup. And the circuit has been working fine UNTIL the short :-) Regulator
does
> have 0.1 uF caps against ripple. Supply pins are the right way around, and
> both VDD and VSS pins are connected.
>
> But, even though I understand the issues involved with the regulator etc,
I
> am wondering why it killed the PIC but not the other components. Even
though
> lots of current flows through the short, it shouldn't flow through the PIC
?
> What kills it?

Are there any capacitors to ground on PIC pins?  If so, these could have
dumped significant current into their pins if the PIC supply got suddenly
shorted.  That could kill the PIC.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, .....olinKILLspamspam.....embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\15@015133 by Lasse Madsen

flavicon
face
> Are there any capacitors to ground on PIC pins?  If so, these could have
> dumped significant current into their pins if the PIC supply got suddenly
> shorted.  That could kill the PIC.

Lets not forget that Decoupling capacitors at the Pic's supply lines are
most nessary if the powersupply doesnt have alot of filters.

Ripple can also kill the pic and therefore its nessary to Decoupling the Pic
close at its supply rails,
with a 100nF capacitor.

What i think happened was probertly a faulty wiring - hey i've done it alot
of times on breadboards ..


best regards
lasse madsen

--
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\10\15@022743 by Vasile Surducan

flavicon
face
Dennis sorry for this crude ( but true ) answer:
You have killed another PIC because you're not carefully.
Once you have a short anywhere on the board you'll not rush to insert
another PIC on the board before you solve the short circuit problems.
I've guess you have a reverse voltage on one pair of powering pins.
Vasile


On Sun, 14 Oct 2001, Dennis Noordsij wrote:

{Quote hidden}

--
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\10\15@052244 by Dennis Noordsij

flavicon
face
On Monday 15 October 2001 08:23, you wrote:
> Dennis sorry for this crude ( but true ) answer:
> You have killed another PIC because you're not carefully.
> Once you have a short anywhere on the board you'll not rush to insert
> another PIC on the board before you solve the short circuit problems.
> I've guess you have a reverse voltage on one pair of powering pins.
> Vasile

Hi Vasile,

Actually my first prototype was on a veroboard, and I had lots of tracks
shorting (though soldering did not cover the whole width of a track, I still
don't know why it's so fragile :-)) which I had to fix until finally the
5V-GND went and everything went dead :)

I didn't rush out to stick a new PIC on the board, I reassembled the circuit
more carefully on a breadboard, and everything was going fine until the power
supply part (the 05 regulator, the ONLY thing on a veroboard! Ohh the irony!
:) had to some more work due to an external device and the tracks shorted out.

From memory (dont have the board in front of me :-) there would be have been
caps going to ground (this is to answer some other people's answers) at CLKIN
and CLKOUT with the 4 MHz crystal. Otherwise a bunch of LED's to test the
bootloader/serial programming modes, and not yet any real other circuitery.

The regulator had caps (though small, like 0.1 uF) between the main +12V and
GND, and another cap between the GND and +5V. Then another cap on the
breadboard where the +5V and GND start.

So a good idea would be to stick a bigger cap there to filter the supply
lines at least for the PIC (100uF?) and maybe diodes to protect against
reverse polarity? (hey just in case :)  Maybe a resistor going to the VDD
which would limit any current to less than the max allowed by the PIC?

Thanks :)
Dennis




{Quote hidden}

--
My cup hath runneth'd over with love.

--
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\10\15@055802 by Vasile Surducan

flavicon
face
Hmm, the regulator looks ok. Whitout any PIC in the socket, ( eventually
use a resistor as an equivalent load ) take a scope and look how is the
transition of VCC after off-on switch. If it's ok ( the output voltage
does not exceed +5V ) then check " the bunch of LEDs connection", again.
Capacitors from clkin and clkout are usual a few pF so don't matter
here. A 10uF from VCC to GND and a 100nF in parallel is also enough. Check
again if pin 11 and 32 are tied to Vcc and 12 and 31 are connected to gnd.
If supply is ok ( even the unregulated +12V input) then you have to
desolder all pic IO pins from the circuit and only then to supply again
the board with pic in the socket.
An anti fool protection ( diodes in series with unregulated supply ) is
a good choice on every protoboard.

Take care and succes, Vasile



On Mon, 15 Oct 2001, Dennis Noordsij wrote:

{Quote hidden}

--
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\10\15@055805 by John Lawton Electronics

flavicon
face
>The regulator had caps (though small, like 0.1 uF) between the main +12V and
>GND, and another cap between the GND and +5V. Then another cap on the
>breadboard where the +5V and GND start.
>
>So a good idea would be to stick a bigger cap there to filter the supply
>lines at least for the PIC (100uF?) and maybe diodes to protect against
>reverse polarity? (hey just in case :)  Maybe a resistor going to the VDD
>which would limit any current to less than the max allowed by the PIC?

I like to give the regulator a minimum load current by using a resistor
between 5V and 0V, say 4k7.
I would suggest that you add an electrolytic cap, say 10uF across 5V and 0V
to absorb any spikes on the line, and to further improve stability of the
regulator. I've seen 7805's oscillate quite merrily without a good load
capacitor (I needed to use  at least 470nF on that particular occasion).
- However, don't add too large a capacitor, since you will read in the data
sheet that shorting the regulator input will cause destructive currents to
flow from the output capacitor into the regulator.

The data sheets also recommend a capacitor 330nF or greater on the
regulator input.

BTW, I've never blown a PIC.

John

-----------------------------------------------
John Lawton Electronics
Custom Electronics Design & Development
EraseMEdesignspam_OUTspamTakeThisOuTjle.co.uk   http://www.jle.co.uk
-----------------------------------------------

--
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\10\15@161836 by uter van ooijen & floortje hanneman

picon face
> >This happened to me once too. I used alligator clips to connect my
> >7805-on-a-perfboard regulator to my breadboard. One time, the clips
shorted
> >(5V-GND) and the PIC died. From then on, if I powered it up, it got REAL
hot.
> >The PIC obviously doesn't like it's rails tied together.
>
> Crowbarring the supply rails, by itself, isn't going to cause any
problems.
> However, if you have charged caps connected to I/O pins, those will
> discharge back through the protection diodes, and put the chip into SCR
> latchup, although the latchup should clear when the chip cools.

If I recall well Microchip specifically warns agains a caps directly on MCLR
(I think a ~100 ohms resistor is recommended).

Wouter van Ooijen

Van Ooijen Technische Informatica: http://www.voti.nl
Jal compiler for PIC uC's:  http://www.xs4all.nl/~wf/wouter/pic/jal

--
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\10\16@153120 by Benjamin Bromilow

flavicon
face
In my experience when voltage regulators give up they usually start
supplying VDD rather than nothing. Immediately you get 12v into the PIC.
Adios muchachos! Sad personal experience speaking there! Zener protection is
not a bad idea....

Ben

--
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...