Hi, it's dumpster teardown time and if you're not following me on my second channel eevblog2, you damn well should be. It's linked up here somewhere and at the end or down below. If you haven't seen it, this is where I put a lot of my like dumpster diving videos and just other miscellaneous videos and you would have if you subscribe. you would have seen this the other day.

This was a dumpster finder along with an excellent uh, 24 inch Dell monitor as well. This is actually a touch screen, uh monitor and you can see that it's um, it looks like one of these, uh, all-in-one Pc things, but it's you know it's got this stand which, uh, flips out like this and that allows it just to give like a, you know, a nice angled surface where you can actually do all touchy feely stuff. And of course it plugs into a Pc via Usb and it just acts as like a Windows touch tablet uh, interface and a normal Hdmi input screen as well. But the interesting part about this is that it's not that capacitive touch rubbish.

or none of that resistive touch rubbish. This is actually an optical touch screen. and apparently I believe, like a lot of people are saying, these are relatively, uh, rare. You might find them like industrial applications and things like that, but uh, in a consumer monitor like this.

Uh, Dell one? Fairly rare. So this was a really amazing dumpster. Fine, it's going to be really useful in the lab here, but I thought it would be interesting to have a tear down of this and see what technology it uses for the optical touch system. Now of course there are basically three more, pretty much four kinds of uh touch screen, uh displays.

You've got your resistive touch which uh, basically has like a two uh, transparent conductive surfaces and when you actually uh, touch it, it actually forms a resistive path and then it can figure out the xy location and they're probably the most common type you'll find. But they have disadvantages in that you can only touch it in one location so you can't do a gesture. the two fingered gesture thing which you can with a capacitive touch and with uh, optical touch screens like this as well. and of course the capacitive touch ones.

They've got basically uh, two, uh, glass plates on them and then it detects um x y location based on capacitance of where your finger actually is and the third one is a bit of a masculine one. It uses a surface acoustic wave or saw uh screen so it basically uh cut you know, the transmits from the side and it basically and an ultrasonic uh signal goes across the uh surface of the screen and based on that where your finger is, it interrupts that and it can detect it. And I've done like surface acoustic wave video I might have to link in uh because that like surface acoustic wave uh delay lines for example and they're quite fascinating. but the other major type is used in a lot of industrial applications and that is an optical system.

So around the edge of this you're going to usually they're going to have an infrared so you can't see it. They're going to have an infrared transmitter and infrared receiver and it somehow detects where your finger is. Now there's actually two. Well, at least I know of two ways to actually do this.

Uh, and this was like used back in the 1980s. Uh, Hewlett-packard had one of the first Uh touch screen computers back in in the world, the Hp 150 I believe. it was uh, back in the 80s that used an infrared touch system. It had a bunch of infrared uh transmitters and like you know, transmitters all down here, infrared receivers on the other side and based on you know all those it could detect where your finger interrupts the uh, the transmission of the infrared path.

Now, uh, unfortunately you need like for a big uh, you know, large, like what is this 22 inch screen or something and you need a decent amount of resolution these days. Like you have to get like, if you can see that, that's pretty fine resolution going on there as I move my finger so you can't really do that with I get that fine resolution with well, maybe you can, but you'd have to have an awful lot of transmitters and receivers around the edge of this thing to actually detect where your finger actually is on this thing. So yeah, that's a bit tricky. You can see, it can just detect the fine point of my little poker here.

And really, this does have phenomenal resolution like this. So yeah, it's I think something else is going on here and one of the other methods I know of that you can actually, uh, do this. I still don't know what's name, what this monitor uses. it just says it's optical.

The other method is to have a optical transmitter and receiver in there, but basically you only have like a single one or a couple of them and the rest around the outside surface. Like this, you will have a retro reflecting surface so that when the infrared lead transmits and it'll bounce back off the retroreflective surface. Because retroreflective, it doesn't matter which angle you come in at. If you come in at this angle like that, it'll actually reflect it exactly back at the same angle.

And of course, this is what they famously used to measure the distance to the moon. The Apollo 11 mission, they left a retro reflector on the moon and it doesn't matter which angle you shoot the laser at it from, it comes back and and this is what they use on like safety vests and you know, uh, signs at night and things like that. So I suspect that's what might be happening here. Is this Around the outside might be a retroreflective surface, and they might just use, uh, just use triangulation to try and figure out where your finger or fingers is.

In this particular case, you know we've got two fingers here. and of course, if I just do a single finger, it's not going to do anything. but I use 2 and it's going to do that. So obviously it's capable of getting both in the location of both fingers like that.

So and if we have a look in here and zoom right in, you can see that it's really quite thick. This bezel. six or seven millimeters, something like that in there. So they've got some sort of uh thing around and goes right around the edge of this thing.

So anyway, this could be really interesting. Let's tear this puppy down and see how in this particular case, an optical touch screen works. Because you don't often see these puppies, especially on consumer products because they're more like industrial. They're very good for, like, you know, out on a big kiosk, consumer kiosk out in the wild or something like that.

And uh, you know you can use them with gloves and all sorts of you know, things like that. So unfortunately, uh, the optical ones. They can actually get like dirt. If they get dirt around the outside and things like that, you have to clean it.

Maybe that's why somebody dumped it. Just touchscreen wasn't working anymore, didn't thin. They didn't know how it works, so they probably when I got it, it was filled up with a ton of gunk in there. So I just got a nicer, proper wipe and just wiped around there.

Or do it with a cotton bud or whatever. And Bob's your uncle and I thought I might be able to see something on the camcorder. Some sort of infrared, uh, transmitter? that's a lead up there? Don't worry about that. Um, but no, I can't see anything.

No, I have to tear it apart. but it's got to be infrared. The model number for those playing along home St. Triple Two, O T C.

All right, getting this apart could be a bit tricky. There were three screws there, but apart from that, they're the only screws I could find on this thing. So I obviously got around here and it was able to get that off. So I'm hoping that the whole this whole top bezel will just come off and we might be able to see something interesting already.

This is really quite tricky business. I really hate these things that are all just put together with plastic clips. Damn it. Yeah, look.

two plastic clips down there. It's going to be all the way around. Ah no. it's all integrated into the frame.

Get the whole screen out and then get the frame out there. you go. It's uh, really solid. I like that.

wow look at the spring mechanism for that stand at the back. That's really quite something. that's really beefy, but you know it needs to be. He needs to take a fair bit of abuse as a touch screen, because you know you get frustrated.

Damn it. I got some tape holding this together and it's an extra screw under here, which is ridiculous. I don't know what they were smoking there. Why couldn't they just have left a little cut out in there like that? I don't know.

It's dumb, but, uh, non-symmetrical Hate. non-symmetry in design. Our T-con driver board. so that's for the Uh.

Lcd of course, made in Korea. How do all my Korean viewers? Uh, so yeah. Dell monitors? Well, at least this one Lg display. so that's clearly our backlight driver.

The you know, cold cathode? uh, backlight on the thing. That's got nothing to do with the optical touch, so we've got to look further. Could another ribbon cable going off there? These are definitely all uh. Lcd flat flex connections.

Got another one going off there and another one going off there. so maybe is that where it's doing the business and for those playing along at home? This was, uh, first released in December 2011. So it's not new. top right corner and there's two there.

I reckon one is a transmitter, one is a receiver and if we go over to the other top corner up here, got to get the right angle to see it. There it is. It's the same thing but the other corners down the bottom here. I'm not seeing the jewel window so I reckon that's just some retro reflective tape there in the corner.

If I hold my hand over here, you can see it's actually reflecting on my hands. This surface along the edge is actually reflective so it doesn't appear to be a retro retro reflect us. And of course, that now makes total sense with these flat flecks here. This is the corner that has that uh, dual element, uh, sensor.

There's that little flat flex going off there and same in the other corner down here, although not sure what that one's doing and that's it. There little six pin slot 23 requires delicate removal of the outer frame here. This is an Ips display as well for those playing along at home. frames separated from the panel.

so the panel's just all the basic column drivers there chip on board. I can show you a bit closer there's our row driver so that's all you know. That's your standard monitor, but what adds the touch magic is the outer frame. Somebody has individually tested that and as you can see, there's nothing in it except aha they got the one.

Hopefully that shows up. It's got like the two split surfaces as we've seen the other one down in this corner. it's got the same and the one in this corner has got the same but this one down here has nothing. so that's interesting.

I suspected either four or two. I didn't expect a three. Hmm. there you have it.

Um, all three sensors do look identical. So are they, as I said, a transmitter and a receiver? Or is one like a transmitter and the others a Cmos line camera or something? I don't know. All right, let's have a look at this thing under the tagano shall we and zoom right in. I love this Beautiful.

Oh look at that. There's two distinct elements down there. They're different shapes, different sizes. That's really interesting, can we? They are certainly very different beasts.

So there you go. So I'd say one's a transmitter and one's I ah see. look looking. see the pattern that looks retro reflective to me? Yep, Bingo.

We got you a bit crusty burger down there. but there's the other end. That's the one that doesn't have anything going to it. There's another one.

I think you're going to find that they're all identical. I still can't get back to that retro reflective. Oh yeah, there it is. There it is.

I reckon that's a retro reflective surface. See the individual elements in there. just need to get that at the right angle. So the laser that was bouncing off.

Just the sort of semi-reflective outer part of that. Geez, You get all sorts of problems when you try and view a frame like this. This is Just insane. So yeah, short of getting that those outer there which is probably going to destroy the functionality of this thing and I don't want to do it because it's a working frame.

I don't want to upset the alignment. The corner piece there looks like it's glued onto the two side pieces and then the sensor is sort of stuck on top of that with the flat flex. There we go. Two, four, six, eight, nine.

so nine conductors. none are bigger than the other that is identical for one in each corner. Three of them, not four. Weird.

There's the main processor board. Didn't expect to find anything interesting on there. There's the main ship that goes to the Uh panel down there. and of course, these uh, sensors touch sensors are connected through to that panel via those uh, little six pin slot 23 chip.

So that's really the only interface there. So really, you know there's no clue how that works. Uh, really. So meh.

All right, let's get the scope out. let's probe some signals. I had to, um, put it basically all back together, including like the front panel so I could like switch it on and stuff. and I've hooked it up to uh, the Pc at the back there and I had to because I was getting no signals out of this thing unless I actually hooked up the Usb.

So that seems to enable uh, the touchscreen. so it's got to you know, do the Usb enumeration and all that, uh, sort of jazz. Anyway, let's probe a signal here. Yes, it is.

uh, mains, Earth reference. It makes no difference whether I go to signal ground or hear the signal integrity. Now I am. just.

I haven't probed all the signals yet. but there is one. Let's check. We're 500 millivolts, uh, per division.

So let's look at that. We have a sync pulse here. You can see that that's a some sort of synchronization pulse that doesn't change at all with my finger on the touchscreen. but this other one.

There are test pads for these and here it is. It looks very much like an analog video signal. If I move my finger, look at that, look at that. it changes.

So I'm moving my finger, uh, vertically along the screen. so that axis that there you go. it depends on the location. It's changing this analog signal like this.

Why it's like messy like that. Why it's got a big dip in there? I don't know. Um, but it is certainly an analog system. Why uh, x direction? Then it does that too.

but only because like, it doesn't go all the way. It's limited range so it's in view of the sensor. So that is some sort of analog image sensor. Like you know, as I said before, like a probably like a single line thing.

It's not like your proper uh, you know, it's not like a 640 by 480 little Cmos camera or something like that. I think it's just a single line jobby and if I probe the other one here, we've got. there's three signals here. We've got our uh, trigger signal which doesn't change at all.

We've got our video signal now. once again, I'll go vertical and that's changing. It looks like the same signal really, doesn't it? Well, I can store that all right. I've set that as a reference waveform.

I've done a complete video on this which actually got bugger all views Well for my channel. I know, like 24 000 views or something. Come on. Um, I'll link it in anyway.

I've saved that as a reference waveform. so that's that second channel. Let's go back to the original channel we looked at. There you go.

No, they are. Oh look, it's it's almost. It's kind of opposite isn't it? It's kind of got like an opposite peak there, but uh, yeah. so different orientations.

There you go. so superimpose it over the reference waveform. and if I put a fatter finger in there, let's say the tip of my pinky like that. it's really narrow and if I don't touch it, I'm not touching the screen.

You notice it just dips down a little bit there. But if I touch it like that and then if I roll my finger like that to give a bigger footprint. and if I put my whole hand on there, wow, look at that. I can block out the whole sensor.

So if I actually block out the sensor, there it is. I mean, you know it's it's basically Gonski. I'm covering the entire sensor there. Do that again.

There you go. I'm covering that whole sensor. You might be able to see the Pc moving up there as I do the touch. But yeah, it's definitely an analog image sensor.

and if we zoom out like there is, there is nothing else there. It's just continually repeating that uh image pattern. Cool huh? But wait. there's one more signal here, which I didn't see on the other one.

that one that looks like a clockity doodah. There we go. That's a 100 kilohertz clock. Ah, that's interesting.

I expected it to be significantly higher than that. There you go. So there seems to be once again, that won't change with my finger. There's a hundred.

You can see the uh screen in the background there scrolling. So obviously those things signals are going back over. uh, the main connection here, and I presumably back into the main processor there because there doesn't seem to be anything else on here. There's a couple of other chippies, but they go off to the Lcd here.

All right. So I'll attempt to explain what I think's happening here. I'll probe this sensor down in this corner here, so we'll get that. We'll stop that.

Okay, so what we've got here. Obviously, this is, uh, like one continuous line scan of the sensor and the reason that we're getting this waveform. Just ignore the beginning. I might be able to explain the beginning, perhaps? Anyway, right? Our senses in this corner here.

So let's say that this start point is, uh, the first line on this sensor. What it's doing is the reason that this slopes down like this is because, uh, the attenuation of the signal coming back. So there's a there's the lead emitter here. I still don't know if they multiplexed the lead emitters.

I don't think so. I think they just flood fill it all and they're just looking for the uh detection shadows. Basically, I believe that's all they're doing. So if you've got your sensor over here like this.

Okay, so let's say your infrared Led is shooting out. it's bouncing back off that retro reflector like this. Okay, you get, it's a relatively short path here, isn't it right? So this is going to be this bit here. and then at some point, it's going to start attenuating because it's getting further and further away until it's bouncing right off the opposite corner over there.

Which would have to be that. And then it would start. Then the leads are like pointing. Let's just say right, that they don't actually leads, don't actually scan like this.

But just like for the sake of argument, like the beam angle right where as it gets towards here, then it ramps back up because it's a shorter path between here and that corner and here and here. so it's bouncing back off that far wall over there. right back to this sensor. and that's why we saw the two waveforms are swapped over in terms of their uh, bottom peak here.

When we switch from this sensor measuring this one to this one over here because they've got because they're physically different orientations and but they're going to have the same effective pattern but sort of like a mirror image based on the geometry. Cool. Actually, this is interesting. I'm pairing a test point called Power Lead Power so I think what's happening here is a negative.

On here is actually the lead turning on. so the lead is on for all of that period and then when it goes high, it actually is the lead switching off. So that makes sense. I don't know why you wouldn't leave it on all the time, but maybe they're I don't know.

They're doing some little magic in there. And for those who want to see the circuitry, the three signals I was probe in here uh, that one was the video signal, that one was the sink and that one was the 100 kilohertz clock. They're just three series uh series jumpers there. So yeah, I was able to measure those and it's the same.

Over here they've got uh test points. That was the video test point and this was the sink test point. Okay, so what I'm going to do now is, I'm probing this sensor down in this bottom corner. So if my theory is correct, the bottom of that or the you know, the bottom peak here should be this corner over here.

So I'm going to stick my finger in there. and yep, wiggle wiggle. See the little wiggle wiggle? Yeah, and even if I get closer, Oh well. If I go like that, there you go.

It changes around that point. So my theory was correct that this sensor, of course, if I put my finger right across there like that, goes all the way up. and yeah, slightly change. Just look at that little tiny point there.

So there you go. Theory: Correct. So that goes to show that the odd wave shape we got here is just in non-linearity in the uh. attenuation of the reflections across the screen for the particular sensor in this case, this one down here.

but it's the same. If you've got this one up here or this one up here, you're going to see a similar waveform. So if we were probing this one up here, then the space would have. You know it might be over here.

and then that one would dip if I put my finger right in that uh corner. So there you go. That proves the theory. It's not too hard to figure out when you just start probing around.

So there you go. That's absolutely fascinating. I never looked at these optical screens before and it looks like they use an optical uh image sensor uh, line based one because you wouldn't need anything otherwise and it gets a like an analog signal out and they use three of them. Now I I do believe you could probably do this with just two Uh sensors on the top side, but they've obviously decided that they need three for better resolution or whatever.

I don't know if you, you know, I haven't really given it much thought, but uh, if you're in the know about how they actually process this, or you've got like a link to you know, a data sheet or a paper or whatever it is, Um, how they actually do this, Please leave it in the comments down below so I hope you found that as interesting as I did. Please did, please give it a big thumbs up. As always, discuss down below and check out all my alternative platforms. Catch you next time you.