Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

BGA chip soldering and desoldering

Status
Not open for further replies.

Vicent Yang

Member level 5
Joined
Dec 31, 1999
Messages
88
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Location
Taiwan
Activity points
697
bga chip

I need a help on how to make a good soldering and desoldering of BGA chips without using high cost tooling
 

bga chips

This shows how someone did exactly that:
**broken link removed**

Text is German, but the pictures illustrate the process clearly.
 

chip soldering

Gorilla said:
This shows how someone did exactly that:
**broken link removed**

Text is German, but the pictures illustrate the process clearly.

Enter this site address in:
h**p://babelfish.altavista.com/

and select language to what translating ....
 

desolder bga

Ufff.. some difficulties even semi high cost systems.

We have in the lab a rework system from metcal APR500

**broken link removed**

and in some cases we have problems with soldering and we need to send to assembly to do X-ray verification.

My experience told me that bga > 130 balls are very difficult :?
 

what is bga chip

Most of my experience has been on 388 pin devices or more and I have seen various levels of success and failure with different rework stations, and especially that they are not able to inspect the quality of the joint itself. Some systems use laser inspection by means of oblique direction laser path & detection between pins to check for shorts within the grid but because of this you cannot check for opens, high impedance or partially collapsed joints. So they are only partially usefull for small pin count parts.

The biggest failure I have seen is actually in high impedance joints or partially collapsed balls which caused me to see on one design, intermittent data errors & corruption within the contents of local SDRAM which on some boards even varied with clock speed and the smallest change in supply voltage. We actually suspected the SDRAM, changed it and also the series termination resistor packs before getting it x-rayed which showed the defect in less than 10 mins on about 5% of the balls. We had been at this for 6 days, replaced 2 BGAs, SDRAM, R packs and cuased many arguments within the design team.

Depending on how the BGA footprint is defined such as NSMD or other will determine rework method as removal method is most important if the BGA footprint is SM defined as once the pad is flowed, the SM may be damaged by cleanup before applying new balls or paste. You should also identify if possible the type of paste, lead or lead free and select your rework paste accordingly as the reflow profile is quite different and you may need to protect peripheral parts surrounding the device and in some cases relowed parts on the underside.

In any event for larger devices or > 1.27mm pitch devices, X-Ray is the only way to be 100% sure.

But I am not sure how you can justify a budget for x-ray unless you can hire the use of one from a local factory.

But there is no reliable way to rework these devices cheap.

When I looked at a proposal for an in house facility for BGA rework I stopped after exceeding $70K and the yearly costs of keeping the x-ray machine with calibration & safety inspection & saftey certificate was completely crazy also. This did not include operator training!

On a good note, I have seen more common now, for sub contract houses to offer replacement of BGA with x-ray inspection & report as well as give a warranty for as little as $80 per device so it may be better to just pay someone else to do it.

:R
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top