Jump to content
Sign in to follow this  

Reflowing BGA

Recommended Posts

I've been looking a lot at BGA repair, because I picked up a BenQ W1080ST for $50 with an 'exploded globe'. and 'other faults'.
Upon recieving my $30 knockoff globe (and taking special note, to only EVER run it in ECO mode, in case the bulb is rated wrong. Gotta watch that heat boy!), it worked PERFECTLY..... for about 7 minutes.

After that I get about 5 continuous rows of pixels, that decide that the bottom 1/4 of the screen isn't home.
They move up to the top 1/4 of the screen, overlaying the image, and the missing row is full of digital artifact static.

My bet? DLP chip is having a baaaad day. (And, yes, being the MS you know and love, it WILL be getting a RAM-Sink like every other chip in every other thing I own :P)

Now, before I dive head first into this fix, I wanted advice.

I know BGA reflowing using 'stupid tactics' (like towels on Xboxes) isn't going to work. It doesn't reach 235*C, so the solder wont reflow.
As the youtuber Louis Rossman VERY DIRECTLY points out, the most your'e doing is passing that 150*C barrier where silicone layers can shift and transistors all tend to 'reset' their position, and fix something temporarily.
I intend to do this PROPERLY, because there is no reason a reflow shouldn't work. If you've ever done normal through hole soldering, with new flux, you can re-use solder for achieving good connections.


My plan:

-Sit my Rosin flux in some boiled water and get it mostly liquid.
-Spread the Rosin around the edges of the chip in question (or flow it under, if it's thin enough).

- Oven bake the board at 95*C for a few hours to remove moisture (not that I think this will be a problem it's likely not damp, as projectors live in high heat).
- With the flux now melted 'under' the chip (hopefully), add a little more around the edges.
- Place the Circuit Board in question on 2 bricks
- Between the bricks, place a common bar heater
- Heat the card to 150*C or so with the bar heater; or just..... whatever it will reach... that's one hell of a temp for a household room heater.

- Place a steel nut on the chip, for a minor down-force, ensuring a better physical seating of the BGA (because if it was oven set in the factory, it would have been chip-weight only)

- With the chip in question exposed, and the board shielded  using Aluminium Gutter Flashing (0.5mm aluminium sheet; to hell with the weak ass alfoil trick), heat the chip with a heat gun.

- Monitor the temps using an infrared Thermometer

- Once the chip reaches 240*C externally, stop, and move the bar heater away (still ON) to allow it to very slowly cool back to room temperature.

- Reassemble.


Anyone see anything wrong with my plan?

I'm a big believer in reflowing, assuming 2 conditions are met:

1. The user knows the temp required, and knows that mode chips are only rated for 3 miniutes of this. In, and Out. People say heat guns are too hot, but time is of the essence.
2. Rosin based Flux is used, because even if you only get a partial reflow, you will get de-oxidising.

Discuss if you wish, I'll try and get pics, but I'll be focusing on time constraints of the project, so I won't promise.

Share this post

Link to post
Share on other sites

Plan sounds reasonable, things I'd take under consideration:

- Getting precise temperatures with a bar heater is going to be a real trial-and-error affair. Try it with something you don't care about before you operate on your projector! See if you can remove a BGA package IC without damaging the chip or the board, using the method you've planned.

- Go to Jaycar and get a flux pen, rather than messing about with something you have to melt. The stuff in the pen is way way way thinner, like the viscosity of isopro, it'll wick under the chip and between the pins far more easily, and at room temperature. 

- I wouldn't rest anything on the BGA package while reflowing, the packages are designed to be reflow soldered under their own weight. I get the temptation, but also keep in mind that even though these parts are designed for reflow soldering, they are not designed to live at reflow temperatures for long periods of time. Adding something with relatively decent thermal mass like a steel nut will mean the chip will take longer to heat up, and stay hotter for longer. You say 3 minutes, that sounds like a lot to me.


Note the time spent in the critical temperature zone is quite short.

Perhaps more critically, the board would have been assembled with the "right" amount of solder paste applied, pushing down on the package, even just a little bit, and depending on the pin pitch, could mean enough solder gets "squished" out to cause connections between adjacent pins.

- Buy or rent an infrared imaging camera if you can. An IR point-thermometer is OK, but if the IC is biggish, you'll want to make sure it's heated evenly to prevent thermal stresses from appearing.

I guess at the end of the day, you've not paid much for the privilege, so you can afford to be a bit cavalier about it for the learning experience. But I think if you err on the side of caution, you'll have a better chance of success.

  • Yes Sir! Very atomic! 1

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this