PC/Laptop Motherboard Reflow for a Newbie

[houndazs]

Hey everyone,

I own a PC Repair shop, and often replace motherboards.  I have had success with the butane torch "Reflow" method of laptop GPU's but never, EVER charge for this for my clients.  I know that it is the wrong way to reflow.  I've also had good luck with the oven bake "method" for motherboard relows, again I NEVER charge for it, and I always tell my clients that it's a cheap/ghetto fix.

Question:

I want to start doing this the correct way, and was wondering if this awesome forum can point me in the right direction?

Background about me:

Basic Electronics principal theory knowledge, and very handy with a soldering iron.  I used to repair aircraft avonics at the component level in the Air Force.  I often replace bad capacitors, and other "easier" components of LCD's and Motherboards.

Thank you for the replies!!!

[wraper]

There is no correct reflow. All who do reflow / reball as a repair are either clueless or con men.

[houndazs]

There is no correct reflow. All who do reflow / reball as a repair are either clueless or con men.

Like I said, I never charge for it.  I was hoping to get into repairing them correctly.   I can operate a MM and an O-Scope, and understand the theory, and workings.  Ill check out the video, thanks

[wraper]

And yeah, if you want to do it correctly - replace the chip, there is no easy way to get one. Most of what is sold is a crap from Chinese sewers. They take E-waste, remove/reball the chips, sell as new/refub/used, whatever. The problem is, those chips for which demand is the highest, also fail most often. Therefore those E-waste boards usually failed because of faulty GPU/chipset BGA chip. So in the end you get the same faulty crap (resurrected by heating) as what you want to replace. Also removing old marking and applying new one, is very often. The same die is used in many chip modifications, sometimes remarked chips work, sometimes not, often depends on the actual motherboard. All I can say, doing this kind business with conscience and earning something at the process is nearly impossible.

[houndazs]

Wraper,

I've about watched the whole video, and it makes 110% sense!  Thank you for pointing it out to me!  I also get the whole "attachment" thing to a laptop most people have, and now understand why both the "oven bake" and "butane torch" method "fixed" the issue with the GPU's.  Thank you so much for pointing it out to me!  Now I can tell my customers why it's only a temp "fix" with more information.  Which, might just end up in a new PC build for them

[lacek]

Sorry for digging the old thread, but are there any good references on this problem?
Somehow I see lots of arbitrary unsourced information that basically falls in three categories:
1. No clue why, but it worked/did not work
2. Logically flawed arguments why it should work and their easy rebuttals
3. Louis Rossmann type information - with some arguments, but far from complete explanation

Which brings me to my question:
is there a published "scientific" study that describes why heating up these chips works temporarily? I would expect microscopic description, identifying the cause of the failure of these chips (e.g. precise definition of some failure in manufacturing process), and why heating up the broken chip helps for some time.