We ran four boards, each with 200 0402 placements (100 E/W and 100 N/S oriented, in a 1cm pitch grid).

Variables we did NOT change this lab session:

* All boards had deliberate thermal imbalance on one side of the part
* All boards used framed 1:1 stencil with no reduction
* All boards used ChipQuik TS391SNL10 paste
* All test components were Stackpole RMCF series 1K ohm