Why Are Printed Circuit Boards Usually Green? (2017) • Hacker News Board

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#41

Post by zrobotics » Sat, Dec 29, 2018, 8:59 PM UTC

userbinator wrote: post #23

msoad wrote: post #8

Apple PCBs are all black nowadays. They probably went all they to make the black pigmened oil as good as green.

I suspect it's a deliberate choice both as a fashion statement and to discourage repair. It's interesting to see a similar phenomenon with desktop PC motherboards: "gamer" or otherwise "enthusiast" boards are often a non-green colour and come with superfluous design elements (like plastic covers over the actual components, which sort of remind me of the engine covers on newer cars...), while both the ultra-cheap preb…

I blame case windows for this. The last desktop I built, it was a chore finding a case that didn't look stupid, and I eventually gave up on a normal motherboard, as anything with a z170 chip set looked like it belonged in a transformers toy.

And while people definitely geek out over servers, it's nearly all about specs. After all, why make something that gets hidden away in a rack in a DC look pretty?

I suspect that is one reason Apple isn't using green mask, since the connotation for many people is 'cheap'

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#42

Post by aparks517 » Sat, Dec 29, 2018, 8:59 PM UTC

amelius wrote: post #37

robterrell wrote: post #30

Earlier quoted context omitted.

This research is why we were sold computers with green-colored phosphor monitors back in the 80s. I'm not going back, but if you're into it, maybe try: https://marketplace.visualstudio.com/items?itemName=gerane.T...

At some point people started switching to amber. If above research is correct, then why was amber deemed preferable?

My recollection is that it was mostly for brightness. I think the amber phosphor had shorter persistence and was less prone to burn-in as well. But yeah, kinda harsh on the eyes in comparison.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#43

Post by s1mon » Sat, Dec 29, 2018, 9:16 PM UTC

msoad wrote: post #8

Apple PCBs are all black nowadays. They probably went all they to make the black pigmened oil as good as green.

Some flex circuits are black because they use a conductive coating to create ground planes without increasing the stiffness as much as another set of copper/adhesive/coverlay layers.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#44

Post by hwillis » Sat, Dec 29, 2018, 9:16 PM UTC

Sniffnoy wrote: post #38

What this article doesn't answer is, when it talks about the green soldermask being physically superior, is that only because it's had more R&D put into it due to being popular, or would it actually be the best regardless? (Admittedly, I don't know how one would possibly answer that question without, like, actually trying to invent better soldermask of other colors. But it bugs me that the article doesn't even ask th…

Solder dams come down to a bunch of factors, but the bottom line is basically the best dye is the one that lets the resin cure stable and strong.

For instance, white dyes are hard (for a lot of things) because white dye particles are large[1] and you need a lot of them. That means they can interfere with curing.

Green dye can be relatively small amounts since green is so visible to the human eye, but afaik its more down to the fact that theyve been able to tune the dyes down better over time.

[1]: also why you can't dye anodized aluminum white. Most dyes are colored at the molecular level, and are much smaller than the wavelengths they reflect. White dyes need to reflect every wavelength, and normal molecules only reflect quite narrow ranges where they resonate.

In order to reflect white light you need particles at minimum the size of the wavelength. Then you need a range of particles for each wavelength. In practice this judt gives you black, since each particle will also be absorbing light at other wavelengths. You need one size of particle with fine, often fractal-like microstructure that has features at every size. That means white dye particles are hundreds of times larger than colored dyes.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#45

Post by jacquesm » Sat, Dec 29, 2018, 9:30 PM UTC

hwillis wrote: post #44

Sniffnoy wrote: post #38

What this article doesn't answer is, when it talks about the green soldermask being physically superior, is that only because it's had more R&D put into it due to being popular, or would it actually be the best regardless? (Admittedly, I don't know how one would possibly answer that question without, like, actually trying to invent better soldermask of other colors. But it bugs me that the article doesn't even ask th…

Solder dams come down to a bunch of factors, but the bottom line is basically the best dye is the one that lets the resin cure stable and strong. For instance, white dyes are hard (for a lot of things) because white dye particles are large[1] and you need a lot of them. That means they can interfere with curing. Green dye can be relatively small amounts since green is so visible to the human eye, but afaik its more d…

Thank you. That answers a question I once had but could not get a satisfying answer to other than 'can't be done'.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#46

Post by jacquesm » Sat, Dec 29, 2018, 9:32 PM UTC

userbinator wrote: post #25

setquk wrote: post #20

Earlier quoted context omitted.

Also lots of boards had transparent solder mask after that (Tektronix did this) which was great for debugging! Examples here: http://golddredgervideo.com/kc0wox/tek/7504/detailedpictures... Whole scope is a work of art!

Wow, gold everywhere. Even the ICs have gold-plated pins... that's not something I've seen before.

They're beautiful. Good luck oxidizing any of that in normal environments, those scopes will outlast the people that built them. Old computer hardware used to be just like that, I would buy the boards by the pound in the 70's for parts. Gorgeous stuff.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#47

Post by jacwah » Sat, Dec 29, 2018, 9:47 PM UTC

Eduardo3rd wrote: post #2

Interesting example of a place where a small but meaningful advantage in a manual application (reduced eye strain during visual inspection) leads to a larger, nearly insurmountable advantage in a world where the task has been largely automated thanks to preexisting R&D/economies of scale. I expect we will see more and more of this in the coming decades across multiple industries. Anything else fit this pattern today?

This reminds me of "The Calf-Path": https://www.poets.org/poetsorg/poem/calf-path

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#48

Post by CoolGuySteve » Sat, Dec 29, 2018, 9:55 PM UTC

ludwigvan wrote: post #22

> 1. Green can relieve visual fatigue and aid in inspections > In the early days, due to technological restrictions, quality inspections relied on workers manually checking the boards with their bare eyes. Squinting at tiny circuits all days is tiring work, but neurologists and psychologists agree that the wavelength of green light has relaxing effects on the body and can reduce fatigue. Additionally, they have found…

Human eyes are more sensitive to slight variations in green than other colours. This is why YUV video often encodes green sharper and why 16bit colour give 6 bits to green and 5 bits to red/blue. So on a circuit board with subtle green-on-green variations, it makes sense that they picked green.

But since people have many more rods than cones, black/white contrast is easier to discern and white uses all the subpixels in your display to maximize luminance. And speaking of subpixels, subpixel antialiasing effectively triples your resolution when displaying black/white contrast.

The brain also highlights contours of contrasting black/white. (There's an evolutionary quirk in our blue-sensing cones that allows you to see contrasting blue/yellow better than most combinations too but still not as well as black/white.)

So sharp white on black or black on white are typically better suited to human vision if you're trying to discern outlines like when reading text.

Black/white also avoids the persistence effects that cones have, where when you look away from a green image for example, you'll see an inverted magenta copy of the image for a second or two.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#49

Post by pvarangot » Sat, Dec 29, 2018, 9:59 PM UTC

borgel wrote: post #11

greglindahl wrote: post #6

One of my startups was purchased by QLogic, which made PCI cards in bulk. One of their corporate standards is that prototypes have red PCBs, to prevent them ever being mixed together with production parts. These days I have some friends who have a particular debit card that's the same color of red. Every time I see it, I twitch.

At work we often try to do a different color for each revision so you can tell at a glance what's in a given unit. It's definitely saved me a few times!

We started doing that but I think we ran out of colors pretty fast.

Re: Why Are Printed Circuit Boards Usually Green? (2017)

#50

Post by lambda » Sat, Dec 29, 2018, 10:05 PM UTC

setquk wrote: post #19

lambda wrote: post #17

I love that OSH Park ( https://oshpark.com/ ) uses purple soldermask. It's a fairly unusual soldermask color, but they use only purple. It looks great with gold plating, and it gives a satisfying sense of branding to for open source hardware and prototyping.

Actually OSHpark boards I find are pretty crap. They have mouse bites left over, the mask is crap and the registration poor on some boards. Getting better boards from JLC but you lose HASL and purple.

Yeah, there are a lot more options for easy, small-batch prototyping PCBs these days. OSH Park was one of the first that was really easy and accessible to hobbyists, so even if it's not the best or the cheapest any more, the purple soldermask still has a place in my heart.