Nestamicky and others in this group want to know how I've successfully
reflowed the ball grid array (BGA) solder to solve video chipset problems in
G3 and G4 iBooks and some G4 PowerBooks. I've also used the same technique
to reflow the solder and fix the G4 iBook U28 chip problem, and a loose G5
iMac video chip.
First, despite the fact there are a couple of companies that advertise
reflows on eBay, successfully reflowing BGA solder is an art, not a science.
Not even a megabuck reflow station with flux flushing and X-ray capabilities
can successfully revive a computer or X-Box or other device with a
BGA-installed video chipset that has one or more broken joints. I know, I've
paid hard cash to learn that a professional reflow cannot perform a miracle,
though it succeeds in enough instances to make it appear that way.

That's where art comes in. And the art derives from experience and learning
everything you can about a particular logic board in a particular machine.
If you pay close attention to an iBook, say, and patiently examine it, test
it, stress it, and run a myriad of trial and error procedures, sooner or
later you'll arrive at an "aha" moment where you understand whether it can
be repaired successfully using the reflow method, or not.

That's about all I can say. If an iBook makes a chime sound at startup, that
is a sign of a successful power-on self-test (POST) that indicates the basic
machine is OK. (But I've successfully revived iBooks that refused to chime,
too, just as I've failed to revive iBooks that did chime.) The next thing to
do is run the model-specific Apple Hardware Test. That will tell if the
logic board has a video or other problem, assuming the board will support
complete booting and then operation. If the board chimes and then crashes,
it could be anything. But bad video will shut down a laptop every time, even
if it chimes and begins the boot sequence. Beyond all this, the only way to
figure out what's wrong is to play around with a completely assembled iBook
and gather clues.

One technique I use is to remove the iBook's bottom case and inner shield so
I can position a screwdriver handle directly under the video chipset. Then I
attempt to start the iBook and boot into Open Firmware, which has much less
demanding video requirements than an OS boot. If the OF screen doesn't come
up when I hold down Command, Option, O and F, I try it again, putting gentle
downward pressure on the left palm rest above the video chipset. (I'm
careful to not exert too much pressure because the fragile hard drive is
directly between the palm rest and the video chipset.) Usually, this gets
the OF screen to appear.

Then I clear any corruption caused by previous aborted boots or bad-video
shutdowns during starts by typing these three lines, hitting the
enter/return key after each line is typed:

set-defaults
reset-nvram
reset-all

Quite often, the iBook will chime, start, boot and run perfectly when I
finish this sequence and hit enter/return the last time. Sometimes I have to
keep pressure on the chipset with one hand while I'm typing with the other.
Sometimes I have to play around with this process until the right amount of
pressure elicits the OF screen and subsequent successful boots. Sometimes I
have to use lots of pressure, just short of making the hard drive do
zing-zing noises.

Once I get an iBook running and it reaches a stable temperature, I usually
can remove the pressure as the heat has closed the broken solder gap(s). I
then let it run for hours while I run the Apple Hardware Test and/or the
appropriate Apple Service Diagnostic disk (G4 and G5 models), and watch for
screen artifacts and syptoms. If arcing caused by broken joints has caused
internal video chipset damage, it will show as an error during the AHT and
the test will stop at that error. That error usually is not fixable with a
reflow. So I stop right there and don't waste any more time on that logic
board, except to try one reflow in case the error was caused by a broken
solder joint and not irreparable damage to a video trace. (Some people have
had the chipset replaced with a known-good one, but that's way beyond my
capabilities and financial resources.)

I also manipulate the iBook during testing and while it's running the AHT.
Sometimes all it takes is a slight lifting on one corner to cause video
artifacts to appear or invoke a shutdown. But what happens is part of the
clue-gathering process, which can go on for hours until I'm satisfied the
iBook has told me all it can.

The process is time-consuming, frustrating and physically taxing (scrunched
over for hours on end, dealing with tiny stuff), and I only do as much of it
as needed. What's worse, all too often no amount of time spent can bring
success as there often are multiple problems that I have neither the tools
nor parts to resolve.

That said, anyone who's interested in the G4 iBook U28 chip issue can learn
about one person's "fix" here: <http://www.coreyarnold.org/ibook/> Some
people have fixed the U28 by touching each "leg" with a fine solder gun tip
while using a magnifying glass. I've found that the process I describe below
works just as well, as long as you clean away burned flux and examine the
legs carefully for solder shorts after a heat gun reflow with a loupe or
magnifying glass.

Here's where I learned the basics about how to do successful solder reflows;
some of the critical elements were found at links contained within posts at
this site: <
http://forums.macnn.com/69/mac-notebooks/210232/diy-ibook-dual-usb-logic-board/
>

My tools are simple: $20 heat gun; candy thermometer that reads up to 600
degrees F.; aluminum foil with a chipset cutout for heat shielding; a clock;
4 small rocks; solder with a melting point of 430 degrees F. (put pieces on
chip being heated to determine when reflow begins; bend pieces to prevent
rolling; keep solder away from U28 legs if that's what you're doing); and
lots of patience. My last tool is resignation: I only do this if I have
nothing to lose.

My process: With the iBook upside down and the screen hanging off the edge
of my workbench, I place 4 sheets of aluminum foil over the bottom of the
iBook. I use an Xacto knife to make a cutout for the video chipset (or U28),
being careful to not damage the board. I then use small rocks to weight down
the foil so hot air  isn't blown under the edges.

Next, I take a small piece of solder and bend it sharply so it doesn't roll
during reflow. Then I put the solder in the center of the chip and position
the tip of the candy thermometer in the bottom of the u-shaped solder
opening to help keep the solder from moving. I position my watch/clock so I
can see it clearly.

Then I take my heat gun, turn it on low and lower it to a few inches above
the chip, constantly moving it in a figure-8 pattern to equalize
temperatures on the chip.

1) 1-minute warm up to 300 degrees F.
2) 1 to 2-minute increase in temperature to solder melt at 430 degrees F.
3) 45 seconds to 1 minute dwell time at 400-450 degrees F.
4) Slow cool down to minimum temperature on candy thermometer.

I am very careful to not jar the laptop during this process in order to
avoid moving molten solder in the BGA. I also use a dual-heat heat gun with
the switch set on low (750 degrees F.) during warm-up, then moved to high
(1000 degrees F.) to get to solder melt and during dwell time. I switch to
low and raise the heat gun slowly during cool down.

Usually, I'm successful with the first reflow attempt. If not, I increase
the dwell time about 15-30 seconds during the second attempt. If that
doesn't do it, all bets are off because there likely will be damage caused
by the reflow process during subsequent attempts. Also, if a video chipset
has been reflowed previously, that logic board makes a poor candidate for a
successful reflow. (Been there, done that.)

Finally, as I said at the beginning, successful reflowing of solder is an
acquired art. My success rate was slightly less than 50 percent in the
beginning, but now has climbed to the low 70s over time. I've not had any
failures for a year, but only because I have learned to pick my reflow
candidates carefully and I've discovered what works best considering the
tools I'm using.

Good luck if you try reflowing, and don't blame me if you fail, as you most
certainly will, sooner or later. Just be sure that you're prepared to write
off any failure before you begin. On the other hand, the logic board already
has let you down, so you've got nothing to lose.

-- Jim Scott

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are a member of G-Group, a group for 
those using G3, G4, and G5 desktop Macs - with a particular focus on Power Macs.
The list FAQ is at http://lowendmac.com/lists/g-list.shtml and our netiquette 
guide is at http://www.lowendmac.com/lists/netiquette.shtml
To post to this group, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/g3-5-list
-~----------~----~----~----~------~----~------~--~---

Reply via email to