In article .net,
says...

snippety

I've run the mboard out of the case with just power, and pressed,
squeezed,etc. without any hint of what I'm looking for. I've examined it with
a glass looking for cracked board and/or traces, separating solder joints,
the obvious. It does still fail out of the case, so this isn't a
short-to-case issue.

Any ideas what else this might be? Do ASICS and other ICs fail (internally)
in this mode?

Remember that any modern computer motherboard is going to have a
bare minimum of four layers. Most have six, and I've even heard of a few
that have eight or ten.

It sounds to me like the board you're working with may have
developed a hairline crack somewhere. Short of inspection with high-end
ultrasonic or X-ray equipment (which probably costs more than you, me,
and my wife make together in a year), I really doubt you're going to
find the problem, much less be able to fix it.

Time for a replacement motherboard.

Happy hunting.


--
Dr. Anton T. Squeegee, Director, Dutch Surrealist Plumbing Institute.
(Known to some as Bruce Lane, ARS KC7GR,
kyrrin (a/t) bluefeathertech[d=o=t]calm -- www.bluefeathertech.com
"If Salvador Dali had owned a computer, would it have been equipped
with surreal ports?"

"DaveC" wrote in message
news.net...
Mac PowerBook G4 won't boot (but will give the boot "bong"). If I press on
the case below the motherboard, it boots and, if pressure is maintained,
functions normally.

If pressure is released, video will scramble, and -- sometimes -- system
will
freeze.

I've run the mboard out of the case with just power, and pressed,
squeezed,etc. without any hint of what I'm looking for. I've examined it
with
a glass looking for cracked board and/or traces, separating solder joints,
the obvious. It does still fail out of the case, so this isn't a
short-to-case issue.

Any ideas what else this might be? Do ASICS and other ICs fail
(internally)
in this mode?



It's gotta be a trace, solder joint, or via. Try resoldering anything in the
area, if that doesn't work and it's otherwise garbage, you could try heating
up that area of the board with a heat gun until it's hot enough to melt
solder then let it cool off.

On Thu, 26 May 2005 14:29:27 -0700, DaveC wrote:

Mac PowerBook G4 won't boot (but will give the boot "bong"). If I press on
the case below the motherboard, it boots and, if pressure is maintained,
functions normally.

If pressure is released, video will scramble, and -- sometimes -- system will
freeze.

I've run the mboard out of the case with just power, and pressed,
squeezed,etc. without any hint of what I'm looking for. I've examined it with
a glass looking for cracked board and/or traces, separating solder joints,
the obvious. It does still fail out of the case, so this isn't a
short-to-case issue.

Any ideas what else this might be? Do ASICS and other ICs fail (internally)
in this mode?

Thanks,


It's probably a bad solder connection, but as others have said, it
could be a crack in the board. I've seen a number of laptops with bad
soldering on the DIMM sockets. Try resoldering everything that's easy
to resolder with the equipment you have. On a board like that, there
won't be much that's easy to resolder.

If the problem is on a BGA IC, you're out of luck. I guess you could
try putting the whole board in an oven hot enough to melt all the
solder. It sounds crazy, but that's how it was originally soldered at
the factory.
Andy Cuffe

-- Use this address until 12/31/2005

-- Use this address after 12/31/2005

"DaveC" wrote in message
news.net...
Appropriate temperature for this experiment would be 375F? (Google says
melting point of 60/40 is 356.)

Leave it in for 10 min? 20?
--
Please, no "Go Google this" replies. I wouldn't
ask a question here if I hadn't done that already.



You're probably better off with a heat gun so you can localize it, if you
just put it in the oven the parts will fall off the bottom.

"DaveC" wrote in message
news.net...
On Sat, 28 May 2005 00:15:59 -0700, James Sweet wrote
(in article PUUle.10023$Ri3.5477@trnddc09):

You're probably better off with a heat gun so you can localize it, if
you
just put it in the oven the parts will fall off the bottom.

So how do they stay on for the original wave soldering? That is done with
the
PCB face-down, I believe.
--


They're glued on, so you might have some luck, but I've salvaged components
by heating them until they fell off, so maybe if you don't go quite as hot
the glue won't melt.

On Fri, 27 May 2005 23:24:44 -0700, DaveC wrote:

On Thu, 26 May 2005 21:05:36 -0700, Andy Cuffe wrote
(in article ):


I presume that components are physically attached to the PCB with high-temp
adhesive before soldering. I'm concerned about components on the underside of
the PCB falling off when the solder goes liquid.

If that's not an issue, I'm ready to try it. If I destroy the PCB, I'm not
any further behind; I need a new motherboard anyway...

The components are glued on, but it's possible that some of the glue
has cracked with age and parts could fall off. The surface tension of
the solder should hold the small parts on, but heavy parts could fall
off if the glue has cracked. Put it in the oven with the large parts
on top.

In the factory, they silk screen a solder and flux paste onto the
solder pads, then glue the components on. Finally, it's pulled
through an oven using a conveyer belt. It's important for the board
to get hot enough to melt all the solder, but not hot enough, or long
enough, to damage any of the components. I'm not sure if you could
get even enough heat in a kitchen oven. I have no idea what
temperature you should use, or for how long.

As others have said, a heat gun might be better if you can narrow down
where the problem is.

What ever you decide to try, I would get a junk board to practice on.
A few seconds too long with the heat gun could burn the board.

Let us know if it works. I've never tried anything like this.
Andy Cuffe

-- Use this address until 12/31/2005

-- Use this address after 12/31/2005

Dave, as a suggestion perahps you could try the Mac groups etc to see if you can
find someone with a similar unit, but crook or broken screen - might be a cheap
way to solve the problem.

Another idea would be to talk to a Mac technician - he may know any common
troublespots on your model board

David - who hates to see a good Mac go down

DaveC wrote:


If that's not an issue, I'm ready to try it. If I destroy the PCB, I'm not
any further behind; I need a new motherboard anyway...
--

On Sat, 28 May 2005 07:15:59 GMT "James Sweet"
wrote:

You're probably better off with a heat gun so you can localize it, if you
just put it in the oven the parts will fall off the bottom.

I've done things like this a few times, and the technique that I found
that seemed to work best was to position the board just above (1-2") a
hot plate, and set the hot plate to just below the solder melting
point. Then I could go into the board with a heat gun or soldering
iron, depending on the task, and apply the small amount of additional
local heat that the job required.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

"quietguy" wrote in
message
...
Dave, as a suggestion perahps you could try the Mac groups etc to see if
you can
find someone with a similar unit, but crook or broken screen - might be a
cheap
way to solve the problem.

Another idea would be to talk to a Mac technician - he may know any common
troublespots on your model board


Hah! A Mac technician? I remember those guys back when I was in highschool
and the labs were filled with Macs. They'd come in and start swapping boards
without doing any sort of troubleshooting at all. I'm sure there's some
hardcore guys out there who attempt component level work but I've never
encountered one.

The idea of finding another one with a busted screen is a good one, ebay is
good for that too.

On Mon, 30 May 2005 13:58:34 -0700 DaveC wrote:

On Sun, 29 May 2005 19:44:19 -0700, Jim Adney wrote
(in article ):

I've done things like this a few times, and the technique that I found
that seemed to work best was to position the board just above (1-2") a
hot plate, and set the hot plate to just below the solder melting
point. Then I could go into the board with a heat gun or soldering
iron, depending on the task, and apply the small amount of additional
local heat that the job required.

Thanks, Jim. This sounds like a good approach.

It brings up a couple of questions: how can I be sure that I don't exceed the
pre-heating temperature (with the hot plate)? I don't want components
dropping off of the underside of the board. And once I start to heat the
topside of the board with the gun, how can I be sure I don't exceed the
melting point of the solder? I understand that getting things too hot will
melt the solder connections within an IC (they use a higher melting point
solder to avoid this problem during manufacture of the PCB, or so I hear.)

You may have to work up on the hot plate temp. The trick is to use it
to heat the under side of the board to just below the melting point,
so that the extra heat from your iron or heat gun will be enough to
melt the parts you're interested in.

I don't think there is any solder in IC connections. The connections
there are spot welded. The temp concern with ICs is that high temps
will cause the doping ions to migrate and this migration may ruin the
careful construction of the IC. This will be a time*temp thing, so you
need to work as fast as you can.

ICs and transistors may have the silicon die soldered to a metal base,
but I'll bet that this solder will be somewhat higher temp than where
you need to work, but that's why you need to be careful not to heat
any hotter than necessary.

Yes, it sure seems touchy, but I've managed to succeed a couple of
times, with no failures so far.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

"DaveC" bravely wrote to "All" (30 May 05 12:33:01)
--- on the heady topic of " Laptop motherboard fixes with pressure"

Da From: DaveC
Da Xref: aeinews comp.sys.mac.hardware.misc:9529
Da sci.electronics.repair:49238
Da On Sun, 29 May 2005 18:14:33 -0700, quietguy wrote
Da (in article
Da ): ...
Da as a suggestion perahps you could try the Mac groups etc to see if you
can
find someone with a similar unit, but crook or broken screen - might be a
cheap way to solve the problem.

Da That's the backup solution. I've got nothing to lose by trying the
Da heat solution first.

Another idea would be to talk to a Mac technician - he may know any common
troublespots on your model board

Da Already have, several. They all say "replace the motherboard". Nothing
Da against them; they've learned the modular-replacement method. I'm
Da looking a little closer than that.


Dave,

They are not too far off the mark actually because a modern pc
motherboard is a rather sophisticated piece of technology. To be fair,
I'm assuming you have little repair experience and even less equipment
than is available in a well equiped kitchen. Put it this way, it is as
if you were to attempt brain surgery with stone knives and suture the
wound using bone needles. Please, forgive me if my assumption is wrong.

A*s*i*m*o*v

.... Back when I was a boy, we carved our own ICs out of wood.

Why not spend an hour or so with a high power magnifier and look for
obviously poor soldered components before embarking on this potentially
distructive approach?

DaveC wrote:

On Sun, 29 May 2005 19:44:19 -0700, Jim Adney wrote
(in article ):



I've done things like this a few times, and the technique that I found
that seemed to work best was to position the board just above (1-2") a
hot plate, and set the hot plate to just below the solder melting
point. Then I could go into the board with a heat gun or soldering
iron, depending on the task, and apply the small amount of additional
local heat that the job required.



Thanks, Jim. This sounds like a good approach.

It brings up a couple of questions: how can I be sure that I don't exceed the
pre-heating temperature (with the hot plate)? I don't want components
dropping off of the underside of the board. And once I start to heat the
topside of the board with the gun, how can I be sure I don't exceed the
melting point of the solder? I understand that getting things too hot will
melt the solder connections within an IC (they use a higher melting point
solder to avoid this problem during manufacture of the PCB, or so I hear.)

Thanks,



--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"

The Lost Deep Thoughts By: Jack Handey
Before a mad scientist goes mad, there's probably a time
when he's only partially mad. And this is the time when he's
going to throw his best parties.

"DaveC" wrote in message
news.net...
On Mon, 30 May 2005 21:32:37 -0700, THE-RFI-EMI-GUY** wrote
(in article ):

Why not spend an hour or so with a high power magnifier and look for
obviously poor soldered components before embarking on this potentially
distructive approach?

Thanks, REG. I did that. Yes, it's best to look for the obvious *one*
failure
rather than attack the whole board, but I'm down to that choice, now: heat
it
and hopefully fix it, or pay for a replacement board.

Nothin' to lose, right?


Might be helpful to poke around on the board with a sharper tool, like a
philips screwdriver with electrical tape on the end to insulate it. Then you
may be able to localize down to a more specific component.

Andy Cuffe writes:

If the problem is on a BGA IC, you're out of luck. I guess you could
try putting the whole board in an oven hot enough to melt all the
solder. It sounds crazy, but that's how it was originally soldered at
the factory.

But they ususally do this before the plastic parts are attached.

Try it at home and you may find slag where you once had DIMM sockets.

-- David

"DaveC" wrote in message
news.net...
On Wed, 1 Jun 2005 15:11:49 -0700, David C. wrote
(in article ):

But they ususally do this before the plastic parts are attached.

Do you know this for a fact? The SODIMM sockets are surface-mount solder
tabs
just like all the other components. They look like they are soldered right
along with all the other components...

Try it at home and you may find slag where you once had DIMM sockets.

Maybe. Maybe not. The board's toast, anyway. Nothing to lose.


YMMV, just be very careful not to go too high or too long with the
temperature.

On Tue, 31 May 2005 18:14:38 -0700 DaveC wrote:

On Mon, 30 May 2005 20:21:42 -0700, Jim Adney wrote
(in article ):

ICs and transistors may have the silicon die soldered to a metal base,
but I'll bet that this solder will be somewhat higher temp than where
you need to work, but that's why you need to be careful not to heat
any hotter than necessary.

What about plastics? There are several sockets (for RAM SODIMMs) and headers
(for connector cables from HD and keyboard, etc.). Also some plastic film
that wraps around the edge of the PCB where it mounts to the back of the
laptop's case, presumably to insulate the edges of traces from shorting to
the case.

You'll have to be careful of the plastics. They will soften and may be
ruined. The ones I've dealt with must have been added after soldering.

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------

On Wed, 01 Jun 2005 06:27:03 GMT "James Sweet"
wrote:

Might be helpful to poke around on the board with a sharper tool, like a
philips screwdriver with electrical tape on the end to insulate it. Then you
may be able to localize down to a more specific component.

The localized pressure idea is a good one, but I'd stick with a
totally nonconductive tool. There are many possibilities: popsicle
stick, Bic pen turned around, plastic knitting needle, plastic tuner
alignment tool, etc....

-
-----------------------------------------------
Jim Adney
Madison, WI 53711 USA
-----------------------------------------------