Hacker News Reader

Comment by femto | Hacker News Reader

A reason for more external copper pours might also be that the EDA tools have improved and can now handle the complex shapes. Back in the 80s/90s copper pours were a pain using Protel (later called Altium), as they were built using straight tracks rather than polygons. Eventually the program got actual polygons and life became easier.

View on HN

Piggybacking onto this comment, but another reason for external pours is thermal performance. A copper pour on the surface on the PCB allows heat to convect off the board more easily. The gains aren't massive, but they can help as part of a larger thermal management scheme.

I've also heard, possibly apocryphally, that in the old days when we used harsher chemical etchants, removing all of the copper from unused sections of the PCB would increase the risk of thinning the traces beyond what was intended. So in those cases a copper pour would reduce the time the PCB would need to spend in the etchant bath.

I've worked with several stepper motor driver ICs that feature a ground pad on the bottom of the IC and recommend an unmasked copper pour connected to the ground pad via thermal vias, on the opposite side of the board from the IC, sized at least as big as the IC itself. Like, the manufacturer's suggestion is literally to use the copper as a heat sink. If you wanted, you could then affix dedicated heat transfer features, like a traditional finned heatsink, or a heat pipe to a dedicated cooler.

Thermal pads are wonderful. Most of the time for moderate thermal loads, an exposed pad soldered to an internal ground plane running through the whole board is enough, as the copper layer there spreads out the heat well enough to dissipate. It always amazes me that an outer-layer copper fill is not much better than an inner-layer one, so the larger coverage of the inner layer wins every time.

A great app note on getting started with thermal design from TI: https://www.ti.com/lit/an/snva183b/snva183b.pdf

Going by the inconsistent spacing and angles on that 1984 PCB, I'd almost guarantee that it was routed using black tape on mylar film, not a CAD package to be seen. Trying to create large fills back then would require manual positioning of tape over all of the copper fill areas. Tools is a big part of the reason for the shift, it's easy now, and the results are generally much better.

I think that one was actually done in an early Japanese computerized CAD package. There's a lot of weird crap in that layout, but it's the sort that the old-school computer layout programs made, not the sort that humans did. Take a look around U111 pin 30, or south of R112, or, heck, any of the text. Whereas there's no sloppy-but-OK vias or wobbly text or anything like that.

I actually prefer using plane layers with tracks to split to the messiness of polygon pours. It signals design intent to users and fab houses and doesn't require a ton of rules and calculations. Polygon pours have their place in top/bottom power nets.

Reply on news.ycombinator.com