Hi,

 The "Window Pane" pattern would, I think, be real bad for entrapping air and flux under a QFN. Something I thought about and is fairly simple is a pattern of 5 round dots.

See: www.kondner.com/files/Stencil_5_Dot_Math.pdf

   www.kondner.com/files/Stencil_Processing.pdf

The idea was that melting paste and flux pushes air out better than a window pane. I have no idea if it really works.

Bob K.


-----Original Message-----
From: TechNet <[log in to unmask]> On Behalf Of Robert Kondner
Sent: Friday, August 2, 2019 5:23 PM
To: [log in to unmask]
Subject: Re: [TN] Jumping QFNs

Hi,

 I would vote that the flux is popping them  with bubbels. You might try reflowing a few on a hot plate. Mke sure the hot plat is hot so the temp rise is faster than that in the oven.

Bob K.

-----Original Message-----
From: TechNet <[log in to unmask]> On Behalf Of Guy Ramsey
Sent: Friday, August 2, 2019 2:32 PM
To: [log in to unmask]
Subject: [TN] Jumping QFNs

We are experiencing a bout of QFNs that jump off the pads, even off the assembly during reflow.
The parts are handled in accordance with J-STD-033.
The boards are complex multilayer sequential lamination with thousands of vias, blind, buried, and through hole. All exposed vias are filled and plated,with ENIG finish.
We bake the boards 10 hrs at 105C.
Solder paste is FCT WS159 . . . I know, I know.  The Peregrine parts don't solder well unless the flux is crazy active.
Any ideas on why about 2-5 percent of the parts jump off the pads or assembly completely?
Can this be just a solderability problem?