Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Anyone facing QFN package crack problem ?

Views: 4335


chs

#52677

Anyone facing QFN package crack problem ? | 27 November, 2007

Anyone can share experience on package crack for QFN package. This package is so sensitive to any external force. Any good recomendation on desoldering / rework procedure.

reply »

#52688

Anyone facing QFN package crack problem ? | 28 November, 2007

I've been using Xilinx TQFP's in the 44 and 144 lead sizes with no trouble. I don't even bake them before reflow and have never had any problem. The Xilinx parts are certainly NOT sensitive, I could not break one with my bare hands if I tried (I have tried it on a defective one, it is amazingly strong for so thin a package).

Ahh, for rework in the higher pin-count=fine pitch varieties, I use an Xacto knife to cut all pins at the package body. I use the knife blade backwards, it scrapes off a metal chip rather than cutting like a knife. Then, I use a soldering iron to just sweep away the old pins like a broom, retouch the solder and resolder manually. I do all the touch-up and resoldering under a stereo microscope. This is for the .65 and .5mm lead pitch.

Jon

reply »


chs

#52691

Anyone facing QFN package crack problem ? | 28 November, 2007

Hi, Thanks for the info. You mentioned about TQFP, how about QFN, which is leadless and one sided package compared to TQFP ? Actually, we are component manufacturer. Recently we have multiple customer come back with issue of electrical failure on particular pins. After customer and our inhouse failure analysis, we observed some units with crack package which further caused broken wire inside the package. At the same time, we also have evidence that some of the crack could happend during desoldering process.

reply »

#52703

Anyone facing QFN package crack problem ? | 29 November, 2007

Desoldering often applies a LOT more heat and thermal stress to the part, and maybe mechanical forces while everything is hot, to get the part free of the board. So, a lot of damage might be done there, making post-mortem harder.

Without the leads, forces from thermal expansion and board flexing are going to be aplied to the package directly, without the compliance of the leads to relieve the stress.

I am working with some 12-pin leadless chip-scale parts right now, but don't have enough experience yet to say much. We are prototyping with them but will go to production next month. They are REAL hard to hand solder, but seem to be pretty robust. I'm guessing your parts are much larger/higher pin count.

Jon

reply »

#52730

Anyone facing QFN package crack problem ? | 30 November, 2007

Hi chs, What type of QFN or MLF package are we talking about? Assuming there is a middle thermal pad? Body Size & no. of terminals & pitch? What about your statement: "This package is so sensitive to any external force". Can you please explain what you mean with this? We use varies size of QFN:s a lot in our RoHS production. So this is interesting. Sincerely, Ps. Sorry for my poor english. Ds.

reply »


chs

#52744

Anyone facing QFN package crack problem ? | 2 December, 2007

Hi, There are QFN 32,40,48 pins with middle pad. Body size are 7x7 and pitch 0.5mm. I means as compared to 2 sided package, QFN are easily damaged by external mechanical / thermal force. Phenomenon observed are mold cap open which further caused wire broken and electrical failure.

reply »

PCB Soldering Tools

HeatShield Gel- thermal PCB shield during reflow