Hi In the previous video, I did a tear down of a 1986 vintage Pon 2 PDA and uh, on a quick scan of the board, there were these black uh marks on here across component pads and uh, at first glance I thought they were just, um, a no place marker for, you know, for not actually placing the components on the board. but um, several people pointed out that they're most likely um, carbon printed resistors directly on the board and uh, I think that's probably the case on Second Glance here now I sort of, uh, casually. uh, dismissed this uh thought at the time. uh, when I was briefly looking over the board, uh, doing the tear down because it didn't really seem to make sense that you would carbon print resistors on here when you could have just placed a, you know, a cheapest chips.

they almost free these resistors uh, plac and they're so darn uh cheap, it just didn't seem to make sense cuz they weren't using um, carbon printing anywhere else on the board like for example, the Uh buttons uh, which is a fairly uh, common technique. You'll get uh, carbon printed on the Um button surfaces, but they didn't do that so it would have been an extra step. An extra cost in the manufacturing process to actually print these carbon resistors on here, but you can see the physical. It looks like they are because you can see the physical.

The thickness of these ones is much thicker than these ones here, for example. So they're obviously using a different value of resistance for these ones, which are clearly are pull-ups or pull down resistors there I'm not sure what rail that is I haven't actually measured it yet, but these ones are physically thinner. And of course, the resistance of these carbon. These printed carbon tracks is going to depend um upon the Uh length and the width and the surface area of these things and the Uh thickness of the coding.

Of course, they're actually, um, quite variable. These things they're typically like in the order of like 30% or so, so they're They're very crude resistors. uh, but good enough for pull up and pull down resistors like this. But these ones over here.

they've obviously got some other uh pull-ups or pull Downs Probably on this look like something going into the uh, uh, the ROM devices here or somewhere under there. but uh, they does look like they have printed on there. So let's actually measure these things and see what we get now. Uh, you can't actually measure these uh in circuit CU You're not actually going to get a reliable reading, but these pull down resistors here and they are pull down.

This side here is actually ground. I've checked that and uh, these pull Downs just go to these. um, 0.1 in expansion um headers here. so where are? So they're not connected to anything else in circuit, so we are actually able to measure the value of these things.

So let's get in there and let's do that. and we can get in there and measure this individual pad. There we go: 65k, 72k, so there's already 76k. There's already quite a significant 7 4 Very significant spread on these values.

We've got a few in a row there that are quite, uh, quite close to each other. But there you go. those large thick ones up there in the corner. They're you know, around about that 75k value.

So these are actually carbon printed resistors. Now, while it's not great to measure these other ones in circuit because they're going to active devices, Um, but at least we can get sort of. you know, um, a, uh, a ballpark figure And this one down here, for example, is 213. K So these are physically thinner tracks and there you go.

It looks like there are a couple hundred K And of course, the voltage from the multimeter is, uh, usually not um, high enough on the resistance range to turn on any active Junctions within these devices here. But you know it's if you really want to get an accurate reading, you of course, should be getting in there and actually breaking the PCB traces. But of course, uh, it, it can't. It's not going to give us a higher reading unless there's uh, uh, voltages present in the circuit and stuff like that.

So it's not a bad Ballpark And they are all you know Circa like 200k. So they are effectively at least double those thicker resistor values there. and that's exact. exactly what you'd expect.

Uh, just based on the size and shape of them. and if you have a closeup view of these again, um, that's pretty much precisely what you'd expect, you'd expect sort of these thinner ones to be. You know, roughly double the value of these thicker ones here. And uh, so they definitely are printed resistors.

The million dooll question is why they've gone to the effort of doing that. Let's do a quick test here to see if we can modify the value of one of these resistors. I'll do the one on the end there, right on the very end. so it's 65.2k and I'll get my knife out and let's uh, give it a bit of a scrape here and see if we can change this value.

Actually, what I'll do here is I'll show you something on the camera here. I've currently got this to um Auto uh aperture and it's got an aperture value of f uh 2.8 on my My Lens here. I'm using my Opteka Time 10 macro lens and you'll notice that uh, in here right in the center is in focus and because my camera is at an angle like this, maybe a 60 angle or something. Uh, up here because it's a very low aperture value of uh F 2.8 You get blurring right at the back here, so back's not in Focus Center's in focus and this one's not and now I've I haven't move the camera and I've gone into aperture priority mode now I've set it to F 4.8 and you can see.

It's gotten a bit better, but the image has gotten darker of course and a bit more grainy. Um, because of the relatively low light in here and I'll increase that even further. And uh, see if we can get it all in focus. and there you go.

I've gone up to F8 and now you can see all of it is in Focus right at the back and right at the front. but this is as high as it goes and it's very grainy, very dark. Um, that's just an interesting uh effect of when you shoot things at an angle like this. If you're using a low aperture value like that and you've got a good camera, um, things aren't in Focus uh, the you know, the end of the board.

um, and the front of the board is not in focus and that's just a interesting side effect of, uh, shooting stuff like this. So anyway, I'll scrape away some of that. So what was it? 6 2.5 k so we'll scrape off. hopefully I won't I shouldn't cut it.

These are fairly rigid fairly. uh, solid things so, but that will definitely I'm sure that would have, uh, changed the value, so let's measure it. now. There you go: 74.5k it's gone up very significantly, chip away a little bit more there.

Let's see what we get now. Hey, there we go. 186k Beauty Now you've got to remember that this is 1986 we're talking about here. but uh, because of the date code, this one was actually manufactured in 1989.

But we're still talking. You know, 23 years ago or greater. So you know, um, just the pick and place uh, manufacturing technology. and the cost of the individual uh component resistors was much different back then we.

You know it's a totally different world today, which is why you would essentially never, ever see a carbon printed resistor like this in a bit of Modern Gear Uh, you know, actually in the last? Well, I I Haven't seen these for like the last, oh, probably 15 years maybe or something like that as a rough, uh, ballpark that you can actually, uh, do them in. you know, more exotic uh products for various, uh, exotic uh, and Niche reasons. But as far as a general purpose product goes, it's just much cheaper to just place an an 0805 0603, 0402 resistor because they cost virtually nothing. You know, 0.00001 cents each or something like that.

When you're manufacturing hundreds of thousands of items like this, they're practically free. And the pick and play machines these days are so fast that really, you know it's going to be cheaper to place a physical resistor than it is to do the carbon printing process on the PCB because a lot of Uh cost in a product will be barebo PCB manufacturer especially if it's multi-layer and you know, high density and all that sort of stuff. So adding the extra step on there to you know, imprint uh to print those carbon resistors on there is um, you know, pretty much unheard of these days. But what were they thinking back in 198, 6 or 1989 here when they manufactured, uh, this board I presume they did it on the original board back in the first board back in 1986 as well.

So what were they thinking? there were they? There would be a couple of reasons. One would be that the machine that they were using at whatever Factory they were using to assemble these the pick and place machine didn't have enough uh, real spaces available for all the different types of components. But um, you know one of these pick and place machines might typically have say for example, 50 uh, feeders, real feeders on it. And if you want to have a look at these things, you can go have a look at my Um video of a typical Um modern anyway.

um uh PCB pick and place assembly line and I'll paste the link in here for that if you want to take a look I Highly recommend it if you haven't seen it. Um, so you know if you exceed that, Uh, you know that maximum number of reels of components say you've got 50 different values of resist sister on this board in 50 different values of caps? Um, then you need that number of feeders to manufacture this board in a single pass? Um, and of course the machines these days. Um, you know they're double-sided. Um, not double-sided uh board.

but uh, double sided is in. They have reels on both sides of the machines and they can have hundreds of feeders or 100 plus feeders on them. So really. But I don't remember back in 198 Six I actually.

uh, wasn't you know that was a um year or two before I actually started working in the industry? Um I was still doing my uh hobby stuff back then, but I certainly wasn't involved in the industry. So maybe somebody a viewer out there who was uh, in the PCB assembly, uh, business back in ' 86 or ' 89 can actually, uh, clue Us in on that. But anyway, that would be a reason um for going for these carbon printer resistors and they would have done it. The key reason would have been cost.

It would be cheaper to do that cuz there's no way you'll go to the effort to print these carbon resistors here and pay more for it because there's no Advantage they they're just freaking pull down resistors. That's all they are. So um, you know you don't care about the value. So um, let's look at the number of components on this board.

You know we're talking one two. You know there's I Assume that these are all the same value. Uh, cap here. and uh, so they only take one feeder each and there's only a couple of resistors on here.

You know there's a couple of caps down there, but really, there's not that many. So, um, couple up here. So even if they're all different values and of course, there's nothing on the bottom, it's just the uh uh uh board itself. and you'll notice that these are goldplated and not carbon plated tracers.

So it's not like you get those carbon printed resistors for free. It's an expensive extra step. How expensive it was back in ' 86? I Don't know, but you can bet your bottom dollar they did it because it was cheaper and we've just got to figure out why. So really, um, they wouldn't have exceeded the number of feeders on there, even in a basic machine back in? Uh 86? I'm sure.

So maybe it was the physical speed of the machine and the price of the resistors themselves. Maybe they were much more expensive back then than they are. Now you know they practically give them away now. but back then, hey, maybe 0805 surface mount resistors cost? you know, a couple of cents each and they and that takes x amount of machine time to actually Place each one.

The head's got to go back. it's got to fly back. pick up the component, move over boom, drop it down. If it's a multiple head one, it can do it a bit more efficiently.

pick up three five at a time or something like that, and drop them in. but it still takes time. so there's that time. machine assembly time, cost there, which you'll pick typically pay, you know, cost per minute or something like that of the machine time plus the cost of the resistors.

So yeah, they've decided Well it's going to be cheaper to carbon print these things I wonder how much it would have cost them per bare board to get that those carbon resistors. and uh, in terms of the feeders, these um, uh, so packages here, they may have came, come on, uh, tapes and that, maybe the small, um, uh, quad flat pack here may have, uh, come on tape, but these larger ones probably weren't on tape. they might have been in trays or something like that. so they're a different part of the pick and place machine, but really, there's very few components on there.

so I don't think it was I'd be incredibly surprised if it was the limit of the feeders um, uh, you know, and and the requirement for a second pass through in that case now I can't actually see uh, any panelization breakout marks on the side of this PCB here like there's no no breakout tabs or anything. So I think this is a fully routed, um, a fully routed PCB So they would have assembled this in like a custom holder as it went through the pick and place machine. So it's not like that they got. so they probably assembled this one here differently.

This one looks fully routed as well. So they probably assembled the two boards uh, differently as two different Uh processes. So it's not like they sort of. you know, panelized.

Uh, you know the board like this and they put both boards through the pick and place machine at the same time rolling through like that. Gee, I don't know. clearly done those carbon printed resistors. So if you got any, uh, better, uh, insight into that back in the uh, late mid to late 80s, then, uh, leave it in the comments or jump on over to the Eev log.

Forum So I hope you found that interesting. It's not something that you see very often these today's carbon printed resistors. Uh, if you like the video, please give it a thumbs up. Catch you next time.