[TSP #125] Etching PCB with laser

[NiHaoMike]

At 53:38, they show a laser etching a PCB, as in the laser etching the copper directly. How do they vaporize the copper without scorching the PCB material under it?

[T3sl4co1l]

Carefully.

Probably ablation, so that a very fast, very hot pulse comes in (Q-switched Ti-sapphire, say?), heats the surface material (top, say, 100nm or so), that vaporizes, maybe melting the underlying few 100nm but otherwise not doing much.

I don't have a good intuition for how much depth should be affected.  Do the usual thermal conductivity rules apply?  If the spot is hot for, say, 1us, and the material is so-and-so thermally conductive, then it should be heated to decomposition temperature* for this-and-that depth?

Also, if the affected depth is ~100nm, it will barely be visible, just a thin film of affected material.  For example, the dark layer left before it did the finishing pass might be charred resin, which is then blasted away on the final pass using a different pulse setting.

*Note that decomposition needs to be rapid in this case, so this will be several times the nominal decomposition temperature.  For FR-4, maybe not 300C, but 400C or more.  One might then (rightfully) worry about HAZ (heat affected zone) beneath the etched area, if that can become a stress raiser or what.  Though, given the composite structure of most PCB materials, that's probably not a big deal.

Probably, porous materials would be adversely affected -- the ablation zone is probably a small explosion (supersonic expansion of hot gas).  That will force material in all directions, including through pores a much greater distance than even charring of a soft (but nonporous, and reasonably dense**) material would produce.  The residue would probably tend to be conductive as well.  That would really only be a counterindication against poor ceramic substrates, but dense alumina (and others) isn't hard to come by as far as I know, so that probably isn't a concern.

**That is, enough mass, strength and elastic modulus, that it doesn't bounce or deform out of the way, too.

Tim

[Marco]

Wavelength of Ti-Sapphire is still a bit too long, probably UV.

PS. oh he said it was UV.

[Brumby]

How do they vaporize the copper without scorching the PCB material under it?

My guess would be they have a very well defined focus point.  Anything closer or further from that point would be an unfocussed spot, instead of a sharply defined point.  Power density (and thus vaporisation ability) would vary immensely.