Searching \ for '[EE] Popping Bubbles' 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/index.htm?key=popping+bubbles
Search entire site for: 'Popping Bubbles'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Popping Bubbles'
2008\12\31@001536 by solarwind

picon face
I swear I don't know how this happened, but today one of those
"bubble" ICs on my HD44780 LCD quite literally "popped" and filled my
room with the foul smell of burnt semiconductor. Now my LCD module is
rendered to be pretty much useless. How could this have happened? It
was connected to my parallel port, happily displaying stuff and this
incident happened spontaneously. I did nothing...

--
solarwind

2008\12\31@004255 by Bob Blick

face
flavicon
face
solarwind wrote:
> I swear I don't know how this happened, but today one of those
> "bubble" ICs on my HD44780 LCD quite literally "popped" and filled my
> room with the foul smell of burnt semiconductor. Now my LCD module is
> rendered to be pretty much useless. How could this have happened? It
> was connected to my parallel port, happily displaying stuff and this
> incident happened spontaneously. I did nothing...

It could be "latchup". If one of the inputs to the IC goes more than a
diode drop's voltage above + or below ground, it can trigger a crowbar
type short internally, and then things go from bad to worse quickly
because the short sustains itself once triggered.

Some IC types are more susceptible than others, and since there are lots
of 44780 clones it sounds like yours was one of the more sensitive ones.

Bummer.

Best regards,

Bob

2008\12\31@010434 by Joseph Bento

face
flavicon
face

On Dec 30, 2008, at 10:15 PM, solarwind wrote:

> I swear I don't know how this happened, but today one of those
> "bubble" ICs on my HD44780 LCD quite literally "popped" and filled my
> room with the foul smell of burnt semiconductor. Now my LCD module is
> rendered to be pretty much useless. How could this have happened? It
> was connected to my parallel port, happily displaying stuff and this
> incident happened spontaneously. I did nothing...

You let the magic smoke out!  Welcome to the world of electronics!

I just took a glimpse at the datasheet for the HD44780.  The 44780 is  
a LCD controller chip - not the LCD assembly itself.  Was your supply  
voltage (VCC) regulated to 5v?  Any higher might explain the fireworks.

It's been literally years since I've used a parallel port.  I don't  
remember much of the Centronics spec, but seem to remember it works at  
TTL logic levels.  Were you powering your display directly from the  
parallel port or an external power supply?  (You did remember those  
0.1uf capacitors, right?)

Joe

2008\12\31@024331 by solarwind

picon face
On Wed, Dec 31, 2008 at 1:04 AM, Joseph Bento <spam_OUTjosephTakeThisOuTspamkirtland.com> wrote:
> You let the magic smoke out!  Welcome to the world of electronics!

Haha, can't wait!

> I just took a glimpse at the datasheet for the HD44780.  The 44780 is
> a LCD controller chip - not the LCD assembly itself.  Was your supply
> voltage (VCC) regulated to 5v?  Any higher might explain the fireworks.

Yes, powered from a regulated 5 v 300 mA wall adapter.

> It's been literally years since I've used a parallel port.  I don't
> remember much of the Centronics spec, but seem to remember it works at
> TTL logic levels.  Were you powering your display directly from the
> parallel port or an external power supply?  (You did remember those
> 0.1uf capacitors, right?)

The LCD module was powered from the wall adapter. The ground pins of
the port were connected to the ground of the adapter and the data
lines to the LCD display. It was working perfectly fine for a few
hours. Even had a 370 ohm resistor to limit current coming into the
display and 1 k ohm resistors on the data lines...

> Joe
--
solarwind

2008\12\31@075236 by Forrest Christian

flavicon
face
Welcome to electronics!   You have now learned about the magic smoke...
see http://en.wikipedia.org/wiki/Magic_smoke , or google "magic smoke".

Wait till you find out about electrolytics.....   or much worse, Tantalums.

-forrest

solarwind wrote:
> I swear I don't know how this happened, but today one of those
> "bubble" ICs on my HD44780 LCD quite literally "popped" and filled my
> room with the foul smell of burnt semiconductor. Now my LCD module is
> rendered to be pretty much useless. How could this have happened? It
> was connected to my parallel port, happily displaying stuff and this
> incident happened spontaneously. I did nothing...
>
>  

2008\12\31@100504 by M. Adam Davis

face picon face
Probably ground issues.  When you have two seperate power supplies
(the PC supply and the wall adaptor) with circuitry inbetween you are
going to run into the possibility of voltage potentials between the
ground lines.  Is everything soldered well between the two?  Are both
power supplies properly grounded (if they have ground pins) in the
same outlet or power strip?

Two fun problems to ponder:
1. If the wiring between the two devices is not excellent (say you
twisted the wires together and taped them instead of soldering) then
an intermittent ground can develop, and when the ground is floating a
potential can develop that will bridge the ground gap, cause a burst
of current, and fry all sorts of things.
2. If the ground pin is missing on any plugs, or the ground wiring
between the outlet and the cricuit breaker box is broken/corroded/etc
then the ground of the computer may float hundreds of volts above
'real' local ground.  Given the right circumstances this will damage
circuitry and people.  (example:  A highschool teacher would generally
clip off the ground pin for A/V carts.  With the plug turned around
backwards, they would connect to the building's coax, which had poor
grounding.  Another teacher would receive a zap from the ground of the
coax when they attempted to connect their TV)

Isolation is one way to fix this.  Powering both circuits from the
same supply is another way (use the USB port power).  These should be
first on your list to try.

Good luck!

-Adam

On Wed, Dec 31, 2008 at 12:15 AM, solarwind <.....x.solarwind.xKILLspamspam@spam@gmail.com> wrote:
> I swear I don't know how this happened, but today one of those
> "bubble" ICs on my HD44780 LCD quite literally "popped" and filled my
> room with the foul smell of burnt semiconductor. Now my LCD module is
> rendered to be pretty much useless. How could this have happened? It
> was connected to my parallel port, happily displaying stuff and this
> incident happened spontaneously. I did nothing...
>
> --
> solarwind
> -

2008\12\31@105453 by William Bross

picon face
solarwind wrote:

>I swear I don't know how this happened, but today one of those
>"bubble" ICs on my HD44780 LCD quite literally "popped" and filled my
>room with the foul smell of burnt semiconductor. Now my LCD module is
>rendered to be pretty much useless. How could this have happened? It
>was connected to my parallel port, happily displaying stuff and this
>incident happened spontaneously. I did nothing...
>
>  
>
Since you had it working it sounds like it's all connected up properly.  
This sounds like a power problem.  You said your wall-wart is 5V @  
300mA right?  What do you get out of the supply without any load
attached to it?  If it's right around the 5V mark then it's a regulated
supply and the rest of my post is meaningless.  If you get a much higher
voltage, say 7 - 9V then it's an unregulated wall wart and probably
consists of just a transformer, diodes and a filter cap.  If this is the
case then the rating plate means with a 300mA load you should get no
lower than 5V.  Since the LCD module - without a backlight - only draws
a few mA, your voltage was probably too high.  If you don't have a
regulator on your test board you should.  The few pennies extra for a
series diode, low drop out regulator and a couple of caps will save you
from that nasty smell.

This might not be your problem right now, but as a budding young
hobbiest you'll be getting wall-warts from all your friends with all
types of internal configurations and ratings.  Use them with the old
cold war adage in mind -- trust, but verify.

Bill

2008\12\31@105833 by Sean Breheny

face picon face
Hi Solarwind,

Are you SURE the wall adapter is regulated? Most are not, and even the
regulated ones often have a much larger tolerance than typical
regulators do (i.e., it is typical for a 5V regulated wall wart to
output 5.3V).

Note, too, that electrostatic discharge could have caused this, too,
since parts can be thrown into latchup by an ESD event.

Also, is there any chance that you removed the 5V supply and still had
the parallel port attached? Applying voltage to the IO pins of a
device without Vcc supplied can also cause this type of damage.

Sean

On Wed, Dec 31, 2008 at 2:43 AM, solarwind <x.solarwind.xspamKILLspamgmail.com> wrote:
> Yes, powered from a regulated 5 v 300 mA wall adapter.
>

2008\12\31@130925 by solarwind

picon face
On Wed, Dec 31, 2008 at 8:52 AM, Sean Breheny <.....shb7KILLspamspam.....cornell.edu> wrote:
> Hi Solarwind,
>
> Are you SURE the wall adapter is regulated? Most are not, and even the
> regulated ones often have a much larger tolerance than typical
> regulators do (i.e., it is typical for a 5V regulated wall wart to
> output 5.3V).
>
> Note, too, that electrostatic discharge could have caused this, too,
> since parts can be thrown into latchup by an ESD event.
>
> Also, is there any chance that you removed the 5V supply and still had
> the parallel port attached? Applying voltage to the IO pins of a
> device without Vcc supplied can also cause this type of damage.
>
> Sean
>

Yes, it's regulated - 5.1 V with no load.

--
solarwind

2008\12\31@211548 by Ariel Rocholl

flavicon
face
I would point out that I have been using rechargable NiMh batteries very
successfully to feed uC low current demanding circuits, and using them more
and more. It is not only a fact that floating ground problems dissapear with
this approach, but you get the additional benefit of very limited damage to
other electronics if your circuit is doing something fun.
For a LCD circuit like the one you are using, I would definitely recommend
using batteries for testing.

2008/12/31 M. Adam Davis <EraseMEstienmanspam_OUTspamTakeThisOuTgmail.com>

{Quote hidden}

2008\12\31@231847 by solarwind

picon face
On Wed, Dec 31, 2008 at 10:41 AM, Ariel Rocholl <@spam@forosKILLspamspamarocholl.com> wrote:
> I would point out that I have been using rechargable NiMh batteries very
> successfully to feed uC low current demanding circuits, and using them more
> and more. It is not only a fact that floating ground problems dissapear with
> this approach, but you get the additional benefit of very limited damage to
> other electronics if your circuit is doing something fun.
> For a LCD circuit like the one you are using, I would definitely recommend
> using batteries for testing.

Advice taken. I'll be making a multiple voltage battery system -
attaching voltage outputs at useful intervals as well as common
voltage regulators, current limiting resistors and so on.

What battery type do you recommend, and what mA . h rating?

--
solarwind


'[EE] Popping Bubbles'
2009\01\01@150111 by Ariel Rocholl
flavicon
face
BTW if you buy the Sanyo Eneloop brand you can forget about funny NiMh
self-discharge effect.


2009/1/1, Ariel Rocholl <KILLspamforosKILLspamspamarocholl.com>:
{Quote hidden}

>> --

2009\01\01@150211 by Ariel Rocholl

flavicon
face
For what you are using so far, standard 2000mAh AA x5 in series then a
LDO regulator for 5V would be my bet.


2009/1/1, solarwind <TakeThisOuTx.solarwind.xEraseMEspamspam_OUTgmail.com>:
{Quote hidden}

> -

2009\01\01@161826 by solarwind

picon face
On Thu, Jan 1, 2009 at 3:00 PM, Ariel Rocholl <forosEraseMEspam.....arocholl.com> wrote:
> BTW if you buy the Sanyo Eneloop brand you can forget about funny NiMh
> self-discharge effect.
> --
> Ariel Rocholl
> Madrid, Spain

Will do boss. I'm planning to stick a 5 v LDO, 3.3 v LDO, 6 v
boost/buck and 12 v boost/buck on the system once I buy the batteries
- to make a useful power supply station.

--
solarwind

More... (looser matching)
- Last day of these posts
- In 2009 , 2010 only
- Today
- New search...