Components embedded within PCB

[daqq]

OK, this is getting silly:

http://www.assembleon.com/images/User%20data/Downloads/Whitepapers/Embedding-components-into-PCBs.pdf

I knew of PCBs with embedded caps before, but whole chips embedded inside?

[ElektroQuark]

Unrepairable PCBs? Wow, manufactures will be very happy!

[Kjelt]

Very good for security, no more pin databus sniffing, excellent 
For hobby and tinkering at existing products this sucks 

[Fred27]

Wow. That's impressive stuff. No fun for the hobbyist though.

[grumpydoc]

No fun for the hobbyist though.

No, indeed. In fact no fun for anyone who might need or want to do component level repair on such a board.

Interestingly the article suggests that the embedded components won't be soldered just rely on a copper-copper contact from the device pins to the PCB pad - wonder how that will affect reliability. I guess that each embedded component ends up in a gas tight little tomb so oxidation shouldn't be a problem.

+1 for keeping industrial secrets secret, +1 for built-in obsolescence. -several million for the rest of us.

[electronics man]

No more need to rub names of chips

[grumpydoc]

No more need to rub names of chips

I'm almost surprised that they continue to bother putting them on in the first place.

[mikeselectricstuff]

Unrepairable PCBs? Wow, manufactures will be very happy!

With today's high density fine-pitch boards, with underfilled BGAs, we're already way-past unrepairable.

[grumpydoc]

With today's high density fine-pitch boards, with underfilled BGAs, we're already way-past unrepairable.

It's not just construction techniques - repair takes skilled labour whereas assembly can be automated or farmed out to parts of the world with low labour costs. Therefore only fairly simple work  or board-level swaps is economically viable in the West.

It's increasingly that only hobbyists that have any real interest in repair.

[c4757p]

To be honest, I don't see much of an application for this, though I certainly don't have much industry experience.

A process like this is going to be complex and expensive. The only place where it will make any sense at all is in making high-volume consumer goods, where the sheer number of devices produced can offset the equipment cost. But the trend there has been different - put more stuff into integrated, application-specific chips. How long will it be before a cell phone is one chip with an antenna, a battery, a display and maybe a few inductors and caps hung off it? Then this sort of high-density PCB technology will be pointless.

I doubt it would ever be affordable for low-volume electronics in niche industries, unless it caught on in high-volume industries.

Perhaps it will be useful for a couple years in the interim.

To me, this seems like Rev. B of the hybrid integrated circuit. Sure, they have their applications, but you don't see them much anymore, do you? And they're also almost completely unrepairable, and really haven't killed hobby repair.

[grumpydoc]

To me, this seems like Rev. B of the hybrid integrated circuit. Sure, they have their applications, but you don't see them much anymore, do you? And they're also almost completely unrepairable, and really haven't killed hobby repair.

No, 1980's and 1990's equipment with hybrids remains repairable but only when it's not the hybrid. OK based on the fact that the hybrids are often pretty reliable and if they're used sparingly the odds favour any particular fault being "somewhere else" but I would think that quite a few Marconi 2022's went to the scrap yard because their OM345's failed (someone's engineered a replacement for those now).

[madires]

Until Altium and Co. list "buried components" as feature I wouldn't be concerned too much.

[daqq]

Until Altium and Co. list "buried components" as feature I wouldn't be concerned too much.

What were you saying? I didn't hear you over the sound of the video.

[madires]

What were you saying? I didn't hear you over the sound of the video.

You got me, now I'm concerned ;-)

[Kjelt]

I am not concerned, it would cost a lot more to design it this way with routing the spaces for each component, glueing multiple finished board layers together etc. etc. this will become a very expensive product. The only products those costs are valid are tiny portable electronics where each mm³ counts like phones, watches, GPS etc. then you can get a payback.
For normal products where the mm³ are not critical it would be much cheaper and easier to piggyback pcb's on eachother like shields or fold them with isolation in between which is used in numerous products these days.

[dexters_lab]

impressive, though i would expect, at least for the time being it would only be used on the types of product you wouldn't consider to have repairable pcbs anyway

phones, tablets etc

[XOIIO]

Next we need smaller components we can put in the components, so you can put components in components in your circuit board.



This is some pretty insane stuff though.

[SArepairman]

lol

[free_electron]

embedded components is nothing … the bees-knees , and ramping up very quickly, is resistive inner layers. they 'etch' things like series and termination and pull up pull down resistors on inner layers.
Apple has been doing that for a while with the phones and pads. almost no resistors to be found. they are all part of an inner layer. some kind of iron powder alloy is used.  it is a foiljust like copperfoil. apply it, soldermask , expose , etch , add next layer of dielectric and copper and keep going.

this technology is gaining traction very fast for mainstream

[calexanian]

Burried components got a bad wrap in the 80's and for good reason. They just made things difficult and tolerances were terrible. Boards used to be a lot more expensive too. It was a shame to throw out a board because an internal resistor was out of spec, when a plain old resistor was and is still a sub cent item. Nowadays things I would imagine are more reliable. I don't have any plans to design with them, but I would not fault somebody else if it lent improvement to their design.