DIY BGA and Solderless Multilayer PCB Tests
October 23, 2020

A copper-wire puncture turned trapped air into a controllable process step inside 0.4 mm pitch conductive-ink vias. Preliminary tests reached less than 1 ohm through a via and more than 100 megaohms between vias; the assembled stack then exposed ink wicking through Avery fibers as the next mechanism to solve.

The Via Needed a Way to Breathe

The solderless interconnect targeted dense BGA geometry at 0.4 mm pitch. Laser-drilled adhesive masks placed CircuitScribe ink over each contact. Capillary action filled the holes, but trapped air interrupted the conductive path.

Stabbing each ink-filled hole with a fine copper wire released the bubble and helped the ink dry. That change produced preliminary readings below 1 ohm through a via and above 100 megaohms between neighboring vias.

Eight Steps Built the Test

  1. Laminate pink Avery Neon 5975 sticker stock for the lower laser-drilled layer.
  2. Cut the BGA hole pattern with a 3 W engraver at 0.050 mm per pixel.
  3. Apply the laminated layer to the BGA contact surface.
  4. Cut the same pattern into non-laminated Avery stock to create the conductive-ink mask.
  5. Align the two hole fields and place the ink mask over the laminated layer.
  6. Fill each opening with CircuitScribe ink while controlling bubbles and capillary flow.
  7. Puncture every filled hole with a fine copper wire while supporting the chip on a firm surface.
  8. Lift the upper mask before the shorted ink film on its surface dries into the assembly.

The Full Stack Revealed the Material Problem

The later assembly failed because the ink permeated Avery paper fibers, spread between the two adhesive layers, and never reached the pads. The experiment therefore separated two mechanisms: copper-wire puncture solved trapped air inside the opening, while porous paper redirected the conductive path away from the contact.

The next iteration can keep the laser alignment and bubble-release steps while changing mask porosity, ink flow, pad contact, and drying. That is how a maker-scale process advances: each electrical measurement turns invisible material behavior into a design variable.

The Missing Maker Fab is the larger invitation to develop processes like this into useful fabrication capability.

Originally posted on LinkedIn

Brian Greenforest · (2020-10-23 01:15:39 UTC)

Open the original LinkedIn post · LinkedIn activity 6725213048711331840

LinkedIn status when archived: Visible to anyone on or off LinkedIn.

The preliminary tests for 0.4 mm pitch #multilayerpcb #DIY #BGA #solderless process finally exceeded all my expectations (less than 1 ohm via resistance, and more than 100 megaohm inter-via), after introducing a silly step (thanks to the reballing video makers, showing how meditative that process is!) Shortly: 1) Laminate pink (for laser!) Avery stickers (Neon 5975). 2) Laser-drill BGA holes in laminated Avery (engraver 3W 0.050mm/pixel). 3) Stick the Avery to your BGA. 4) Laser-drill "conductive ink mask" (the same BGA holes) in another, NON-laminated Avery. 5) (Carefully, precisely aligning the holes) Stick the non-laminated Avery on top of the laminated. 6) Apply CircuitScribe conductive ink on each hole (paying attention to the bubbles of air under it, and to the whole CAPILLARY PHYSICS of the process). 7) (The most important step!) To facilitate the removal of the air bubbles (and to improve drying of the conductive ink!), STAB each ink-filled hole using a SMALL COPPER WIRE. Make sure your chip lays on a firm surface, otherwise the chip can pull the ink between two holes on the adhesive side, short circuiting the vias. 8) (Don't wait too long before this step) Remove the non-laminated Avery (it has short-circuited conductive ink on top of it anyway). PM me for details.

Comments added by Brian Greenforest on LinkedIn

This comment was also preserved verbatim from Brian Greenforest’s LinkedIn data export or the public post page.

Didn't work because of the ink. It permeates the Avery paper fibers and fills up between the two layers, never touching the pads :-/

View the LinkedIn post