Dason,
I apologize if some of the information sounds redundant but I don't think I explained my point very well (and the standard can get really confusing at times). I definitely agree with using C-SAM to look for internal defects. My only advise is to be careful not to draw general conclusions based on a small sample of components. It is possible that you will not see any defects in your experiment but that does not mean that all components will always be safe using this same process.
The actual drying effect of storage at any temperature and RH level varies significantly based on the physical properties of each component, as well as the actual duration and conditions of exposure and subsequent dry storage. There are years of research and thousands of samples that went into the creation and ongoing revisions to J-STD-033. In order to provide safe guidelines, all simplifications must be based on a worse case condition. Otherwise you will jeopardize the reliability of some of your components.
The standard specifies that you must insure that your moisture-sensitive components are still within their specified floor life, or to completely bake them prior to any reflow cycle (such as localized hot air rework). Of course this is only important if you plan to re-use the components. The standard provides a complete bake table that takes into account all important parameters described above. The upcoming revision includes bake cycles at three different temperatures, namely 40C, 90C and 125C. The 90C was added specifically for drying populated boards that cannot withstand 125C. The actual duration of the bake cycle varies based on the specific MS level and body thickness of the component that must be baked. If you want to simplify your internal procedure you should use the maximum default values. (i.e. 48 hours at 125C, or 10 days at 90C, or 68 days at 40C).
In order to determine the minimum drying duration at 55C, and to be consistent with J-STD-033, you should use the moisture diffusion models identified in Dr.Shook's paper and calculate the worse case scenario. If you do this, you will end up with a duration somewhere between 10 days and 68 days, consistent with the bake duration at 90C and 40C. My suggestion would be to increase your temperature at 90C if possible, to reduce your cycle time and to be in line with the industry standard. Ideally you would also track the exposure time of your components on the boards up to the rework process. This would avoid baking parts that are still within their specified floor life.
I have no clue where the 6 hours at 125C comes from but it is much too short to be safe and certainly not in line with the present industry standard. You might want to ask your customer where they got that number from.
I am not sure if I understand exactly what kind of additional data you are looking for but if you want to contact me offline I might be able to provide some relevant information.
Rgds, Francois Tel : 450-534-2644 fmonette@cogiscan.com
reply »