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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


DoE on post-reflow PCB - warp

ianchan

#20473

DoE on post-reflow PCB - warp | 25 June, 2002

Hi mates,

has anyone ever done a DoE on the degree of warp PCB, after a SMT Reflow process? what were the findings and conclusion based on such a DoE? which were the critical (DoE) factors that influence the degree of warp PCB? any optimal reflow settings to observe?

Input in greatly appreciated. Cheers!

reply »

#20497

DoE on post-reflow PCB - warp | 26 June, 2002

We have not done a DOE on board warping. Consider: * How close the board gets to its Tg during the reflow cycle * Amount of copper on each side and the balance between the two * Thickness of the board * Overall length x width * Time at temperature * Rate of cool down * Multiple thermal cycles - such as rework - are never a good thing. * MLB construction (homogeneous or unbalanced - too resin rich or poor) * Supplier process capabilities (lamination cycles) * Material selections (usually cheap and dirty), etc.

You sure your �warp� isn�t �twist�?

reply »

ianchan

#20513

DoE on post-reflow PCB - warp | 26 June, 2002

Hi Dave F,

customer calls it "warp", we can call it "twist/bow"..etc.

the "warp" occurs on individual PCB cells of 28per each panel. the overall dimension of the panel is 139mm x 168mm, and each individual PCB cell is 2.4mm x 1.2mm. Thickness is 0.6mm. PCB is 6 layers board with copper laminate makeup of combined 3oz (still waiting for info as to how much is each copper layer %makeup).

Thanks for the pointers, will discuss with my people. Cheers, Dave!

reply »

Daan Terstegge

#20524

DoE on post-reflow PCB - warp | 27 June, 2002

Never did a DOE on that, but I think it would be quite difficult to warp a board inside the oven if it was flat when it entered the machine. If there's no center-board support then it's another story, then the board will bend due to it's own weight. The amount of bending-through is proportional to the third power of the boardwidth, and related to Tg and soldering temperature. If a flat board warps in spite of being supported through the oven, then there must be something wrong in the bareboard. Perhaps the boards are flattened by the vendor, with the enclosed stresses coming out during reflow. Bad practice...

Daan Terstegge http://www.smtinfo.net

reply »

ianchan

#20538

DoE on post-reflow PCB - warp | 1 July, 2002

Daan:

thanks for the tip.

reply »

Peterson

#20539

DoE on post-reflow PCB - warp | 1 July, 2002

What is Tg?

reply »

Daan Terstegge

#20550

DoE on post-reflow PCB - warp | 1 July, 2002

Tg = glass transition temperature, the temperature at which the glass-epoxy starts losing it's properties as a solid material.

daan

reply »

#20551

DoE on post-reflow PCB - warp | 1 July, 2002

Remember! Empirical data will allow you to take advantage of the inherent continuity of quantitative variables.

reply »

reflow oven profiler

Reflow Oven