반도체 이야기 : 패키지의 시작, 2편
(Continued from the last issue.) Unlike the leadframe, a significant advantage of the PCB substrate is it can have I/O pins in the whole bottom area in a BGA (Ball Grid Array) package. With the entire area packed with connections, increasingly smaller BGA pitch (the distance between balls) means an increasing number of usable I/O sockets.
However, we can’t prefer PCB all the time. Because, naturally it is more expensive than leadframe. The wiring layer can grow without bounds, like 2, 4, etc, but it is accordingly hard to make. Metaphorically speaking.. You wouldn’t dress up fit for a wedding when all you want to do is buy some snacks and drinks across the street from your house. Same for a situation the other way around. Depending on the property of the substrate, some characteristics can be good but some can be bad. So choosing one depends on the required performance and price of the product.
Package Performance Evaluation and Design
Testing is done beforehand or at the same time as the substrate design to see if the package product meets the performance requirements of the client. One of them is thermal characteristics evaluation. The previous author introduced topics of thermal performance and evaluation, and we have to evaluate so that the product can show the desired thermal performance inside the actual equipment. If not satisfied, do the evaluation again using different material for the package or different size and thickness. If it still doesn’t work, the structure has to be modified, or try suggesting the client lowers their requirements.
And the second is the electrical characteristics. One of the most typical evaluations is characteristics impedance. When the package is working, you get some unwanted resistance, resulting in loss of signals. To adjust the characteristics impedance and minimizing the loss, the evaluation helps determining the suitable design value for the package product.
The third is deformation and damage prediction evaluation. Deformation of the package is a big worry while making or mounting one. Temperature change brings thermal expansion, and so thermal deformation cannot be prevented in a package with different materials. It will be ok if it stops at deformation, but it is a big loss if it causes internal breakage and disconnects. It can be predicted using computers, and deformation can be limited within allowed range by employing different materials or modifying the structure.
When I first started this work a senior employee told me that way, way back in time, he was so out of time he made a blueprint by drawing lines on the drafting board using a ruler, instead of computer CAD. It’s an episode from very long ago. Now technology came a long way since then and we are worrying over many variables in the design phase that were not part of the considerations in the past. Of course there are many more design conditions I could not explain yet. The design is complete after the systematic design phase and input from many experienced engineers. When the design is over for a product and I hear it was manufactured all right, I feel proud even if no one else recognizes it.
Now, the worry is which topic to continue on next. If you have any topic that you want to know more about, please leave it in the reply. I will take it into consideration. Then, see you in the next issue 😀
WRITTEN BY Gyuik Jeong
It is already 10 years since my ambitions lead me to Amkor. They say 10 years is enough to change the face of the earth, but I wish for my mind to stay curious and amazed at each and every work like a new recruit, so the work is always enjoyable.