Are these vintage SMD type resistors?

[gmc]

I've got a old logic board I'm trying to repair from around 1984. It's a PCB from a IBM 5160 Power supply and its a comparitor board to check for specific voltages.

At first I thought it was a simple PCB with SMD Chips on but after tyring to beep out the connections I didn't see much continuity.

Looking more closely it looks like a dual (or maybe triple layer) board with resistors (I think) built into the layers.  Is this possible?

It's the dark grey squares on the PCB which have specific goudges/scratches on them (Is this for getting a specific resistance?)

I've measured a few of them and they have resistances from 1k-100k - but then it could be the resistances through the chips I'm measuring?

Are these resistors, or maybe capacitors (My cap meter doesn't register anything)

I've not seen anything like this before and it has me a bit stumped.





 

[Haenk]

I'd say those are thick film resistors.
Usually those were used for mass produced items to reduce production costs (think "Casio calculator"). Never seen that on a computer PCB.

[gmc]

Interesting. So I was on the right track.

Thanks

[TimFox]

The first project I worked on in industry (1982) used ceramic-substrate chip-and-wire hybrid technology, where the semiconductors and capacitors were "chiplike" and wirebonded to pads on the substrate, while the resistors were deposited (by silkscreen) directly on the metalized tracks on the ceramic. 
I'm not sure how that would work on a composite substrate

[mikeselectricstuff]

This construction used to be fairly common. if you look carefully, you can see lines where the resistors have been laser-trimmed.

[Gyro]

It's not a "PCB" in the conventional sense, it's a ceramic thick film hybrid (don't try to flex it!). Deposited thick film resistive elements like that need a stable and rigid substrate, as does the laser trimming.

[Warhawk]

This brings my memories. We also did some hybrids at my alma mater.

[wraper]

It's not a "PCB" in the conventional sense, it's a ceramic thick film hybrid (don't try to flex it!). Deposited thick film resistive elements like that need a stable and rigid substrate, as does the laser trimming.

I've seen plenty of them on regular FR-4. Carbon pots are basically the same thing on phenolic PCB and with mechanic wear added on top.

[Gyro]

Maybe I should have added the word 'accurate'. The hybrid in the OP's photo is a ceramic one, you can see from the white border and pin attachments.


Edit: I think I've seen some Alps audio step attenuator style pots that do also use laser trimming though.

[gmc]

You are correct - I've checked the PCB and it's ceramic.  Attached is a photo of the back .

[T3sl4co1l]

Here's another classic example:
https://www.elektronaut.at/nes-gamepad-dimensions-pictures/624
Note the carbon pads for pushbuttons as well as pull-up resistors.  (Also note the complete lack of a bypass capacitor on the logic chip -- CD4000 at 5V is painfully slow, and it's not very surprising this worked even with the long (10ft?) cables on these!)

Don't know if they can be laser trimmed on PCBs; maybe needs a fast pulse type to ablate/atomize rather than burn material away?

Hybrids can also use metallization, usually vacuum deposited on the substrate, then etched, then laser trimmed.

Also note the multiple layers, like the crossings with an extra blob of glaze (insulating glass) between them.  A kind of multilayer PCB no less.

AFAIK, hybrids never really got cheaper?  They've always been something of a mass-production item, hard to prototype.  Or, really, I suppose they're easy to prototype in that the materials are readily available and the techniques are easily executed, but just that the amount of hand work put into the preparation, printing and finishing of them is much more than you'd expect, compared to PCBs?  Also not like wirebonding ever really became a small-shop thing, but plenty of hybrids are made with solder or conductive epoxy so that's not a problem.

You can actually get ceramic PCBs from a number of online suppliers nowadays, though I haven't quoted any to see what kind of cost, quantity and lead time they offer.  I suppose it might be an interesting option for stuff like LEDs and small power modules; or multilayer metal-core PCBs, to the same end (which are also fairly reasonably available).  I mean, assuming the ceramic is reasonably conductive -- if it's 99% alumina it should be alright for heatsinking purposes, not great but alright.

Tim

[TimFox]

It's not a "PCB" in the conventional sense, it's a ceramic thick film hybrid (don't try to flex it!). Deposited thick film resistive elements like that need a stable and rigid substrate, as does the laser trimming.

Yes, ceramic hybrid construction required a high temperature process to fire the traces and deposited resistance inks on the substrate, followed by wire bonding (compression welding) to connect the parts (e.g., semiconductors) that were not compatible with high temperatures.
Thick film inks were suspensions of metal or resistive powder particles in an organic or similar binder that pyrolized in the processing oven, leaving the metal or resistive material adhering to the ceramic.