Hi, we're going to talk about Apple MacBook Pros recent model ones and how there's a potential design issue in them that's causing these failed connectors here and it's a real problem compared with the older design. MacBook Pros Now Luis Rossman over at Russman group subscribe I'll link it in down below a link in both the videos and Louis's channel Louis is great. He does Apple repair videos and business advice and no commentary. Love it! Anyway, he did this video about these Boches in the design of the recent I believe from 2016 MacBook Pros onwards the previous ones didn't have the same issue or we'll get into the technical details cuz the devil's always in the detail here so we're going to have a look.

Is this a design oversight is a bad design? Who at Apple which engineer at Apple is responsible if any, because it's a real interesting question, we might dive down the rabbit hole a bit here so take it away. Louis Another Apple design flaw. Many of the design flaws that I've discussed an Apple products affect products from 2008 9, 10, 11, 12, 13, 15. However, today we're gonna get to a flaw that's affecting machines from 2016 17 and 18 I Want it to be clear when I say this that it's not just pure hatred for Apple products that I do this.

It's because I hope that the consumers of Apple products when they realize that their design in this way ask the company. Can you do better? As a trillion dollar company that sells three thousand dollar machines, can you do just a little bit better than something we would expect from a first-year intern? So today we're going to be going over a very common problem on the 13 and 15 inch. MacBook Pros with USB C ports which is no image I Want to go over the design flaw that causes no image, how it manifests itself, what it looks like, and why this is a really bad design and contrast older machines with many of the design flaws that I discuss here. I'm talking about things like a scamp failing, or this thing being placed in the wrong place and stuff where you could.

Okay, it's kind of understandable that this mistake could occur and it sucks that you didn't fix it after a few years, but understandable to some extent here. This is just objectively terrible in a way that I would like to show. So so is it objectively terrible. Well, we're going to get into it.

Let's check it out. Oh Lincoln The Lewises video down. You can watch it for yourself. There's another video here which I'll link it as well, which is one of our Louis's cohorts in action.

What we have here is a 1707 that sustained liquid damage and as most of the time they do when they get liquid damage just because of the way, the liquid gets into the Machine And here's the connect. It destroys this connector right here. This is the LCD connector that goes from the motherboard I've so this connector is for ouch. loosening.

That's nasty. See the carbonization between the pins ena and what you're looking at here. that's terrible is this big pin is what carries the backlight voltage to the screen. That's about so the biggest.

So this is the biggest voltage that you're gonna have and this pin right here goes to on this machine. Luckily, it goes to a muxing circuit that changes between the integrated GPU and the discrete GPU. Okay, so the problem here is that the backlight voltage pin on the LCD connector is right next to a data line, which is going to either depends on the model, either goes to a MUX chip or goes directly to the CPU. And if there's any sort of arcing or any sort of water gets in there and then bridges out between those two pins and I'll do a demonstration shortly.

Stick around for that that can cause a low impedance. Really sort of carbonization. For what of a better term between the pins in the connector causes that low impedance. Generally, the backlight power supplies are reasonably capable.

you know they capable of at least you know, a handful of watts. And then if you've got a low impedance, over to a data pin, a lower voltage data pin. What? That's going to ruin your day? Current can flow into the data pin and not only in the data pin, but through the protection diodes as well. I've done a separate video on that and that can just take out all sorts of stuff.

So I can either take out your MUX chip or the CPU off it. Taste it up if you've got the model where it takes out the CPU It's a ridiculously expensive of course. Apple don't sell any of the parts so you've gotta scrap him from existing boards he apparently can buy. And but I think Lewis says they're like, you know, a hundred and eighty bucks or something like that and then you're not even guaranteed that they gonna work.

Did they scrap him properly? Also, all that sort of jazz. So yeah. really nasty problem. and the older ones didn't have this sort of.

Well, they didn't have the same issue that the later Mac balls. Now I've got the the board open board view software written by Paul Daniels I'll link it in down below. It's brilliant software so can have a look. This is the MacBook Pro board if we have a look at the connector up here and these are the two pins we're talking about why they've labeled in forty three and one like this: I don't know it will go into that shortly.

but anyway, this is where the fault occurs. Pin 43, as you can see is the LCD backlight voltage Lewis says that's about 52 volts but that can vary between the models and then pin number one. here is this auxiliary data pin just goes off to either a max or a CPU or whatever so there's there's no ground in between them. There's there's a little bit of space in because it's physically a much bigger connector and we can actually have a look.

And here's the actual wire connector I think it's a Nova Stack one. It might be an equivalent. There might be various because once you can see, it's physically like a larger pin in there and that has the backlight voltage. Normally you might just use that as a big round pin or something like that, but Apple decided to use in the backlight.

There's nothing wrong with that because it does. It is a larger physical contact pin, so they might have needed it for the higher current. So physically, it's one of these just flat flex board interface connectors and there's nothing wrong with the choice of that connector. In fact, if we go over here to the spec sheet, the voltage rating of the connectors actually 60 volts, AC or DC per contact.

It's basically a 60 volt rated connector. so the design engineers of Apple they've chosen the right connector. It's rated properly, no worries. So the first thing we're going to take a look at is conductor spacing because the higher voltage you go, the more distance or creepage distance and clearance you've got to have between your pads now.

I've mentioned clearance and repeat before, but say clearance. In PCB terms it means an air gap. So if you've got to physical pins like or to wires or whatever with physical air gap between them, that's the electrical clearance. That's different to what's called creepage which is, let's say you had this is your PCB Here you got one trace and you got another trace on top of the PCB next to it.

Just imagine it there, then it. the. Then the voltage can creep along the board and from one trace to the other can follow the contours. Or if you've got a connector like this for example, this would be a matter of creepage, not clearance.

So the distance between the pins in here actually goes along the connector and if they've got little ridges in there, it follows it, have got little bumps. It'll follow it. That's what we're talking about. the creepage distance between those pins.

but I've already seen. This connector is rated for a nominal 60 volts DC AC So it's fine. We don't even have to get out the Saturn PCB calculator here and actually calculate it. but let's just look at some typical numbers.

from 51 volts to a hundred to 100 volts I Only specify arrangers. You know there might be some other calculators on the market that you can type in the actual voltage value, but it's a bit how you're doing. So if we look at a bare PCB with just bare copper traces on here on the surface like this: 51 to 100 volts to have a safe creepage distance between them and or conductor spacing, you need to have not 0.6 millimeters so that's actually not a huge amount. That's why this connector can have that 60 volt rating with those small pins and a lot of people don't realize that that'll change with elevation as well.

So if we go and use your MacBook Pro on the top of a mountain over here, you've got to use it on the top of a 3,000 meter mountain. but if you did that then you would is how high as Everest Anyway, if you did use it above that, technically you have to go to one point five millimeter conductor spacing and once again there might be some how you doing formulas out there for the height of things like that. Anyway, you see we're already starting to go down the rabbit hole of PCB design here and clearance conductor spacing and creepage. but it'll be different if you're If you have sold a mask on the PCB, it's only not point one three millimeters once you add solder mask um, soda mask is an excellent insulator.

It allows you to get closer conductor spacings. but of course when you've got a connector, then you've got the exposed pads of course. But if you can get the solder mask between the pins, which you should do to prevent shorting between the pins during reflow or hand soldering. but then you get much better conductors, that much tighter conductor spacing.

But all bets are off in the case of the MacBook Pro where you get liquid damage and I'll do a demo in a minute. Stick around where water can get between your pins and then doesn't. The solder mask isn't going to save you in that particular case. And by the way, if you actually have it embarrass ID an inner layer of a PCB then that's the best.

Really, it's better than solder mask only naught point 1 millimeters compared to the 0.13 there. Anyway, conductor spacing is fine. so Apple have done their job in terms of choosing the right connector and conductor spacing on the PCB. That'll all be fine, no worries, but let's go do a little demo to see what happens if you actually get liquid damage on your PCB and see what can happen.

It allows you 50 volts. Hmm, this could be interesting. all right. So what I'm gonna do here is a little demonstration: I'm going to get a 52 volt power supply I've got 32 volts in series with 20 volts here both set to 2 amps and I'm gonna connect that down to a breadboard down here just to adjacent traces like this.

So this will like simulate the 52 volt rail and ground or some other data line that it can go to and you're gonna pour various things on it and see what happens. Okay, I've got some water here, It's just been in a bottle that subbed out and I've been sitting on the lab shelf for a while so let's put that on there. See what's gonna happen? anything? I Think we see some bubblies see some bubbling action. Of course the gap between the Mace's I think is probably slightly larger than what's on the connector.

I haven't actually measured it? Yeah, you can see it's starting to react. This is just plain old water that I've had lying around. Doesn't include any gunk or anything like that in it. It's not coffee or some other liquid.

And of course this is what would happen if you happen to have a relatively high power source inside a product like this and you put liquid in there then and if there's exposed contacts, it's It's not going to suddenly just arc over and and carbonize everything. but you can see that it's certainly do. It's changing color I Think we're getting some sort of I don't know. Carbonization is that the correct word Between the traces you can see it's getting really dark.

You can see it will slowly deliver power into there and then break it down. and once it like carbonized for want of a better word between the traces gets lower and lower impedances can dump more power into that trace that's getting really quite black. Now it's not looking good is it? Remember this is only 52 volts? Sure, it's coming from a fairly high heavy-duty power supply. In fact, you can see the parrots delivering.

They're not particularly high power at all, and a backlight driver can easily deliver that sort of current, right? So we're not talking high powers here. and of course is this was a solder mask board. Then of course it would have to get through the solder mask first. So the solder mask is an excellent insulator.

and really, in theory the water should. just, you know, plain water shouldn't do anything to the solder mask. You should just sit there and pool and not and not do anything at all, but eventually it might break down. The solder mask is, you know there's always contaminants in the water.

It's never going to be pure water, it's picking up all sorts of gunk as it's going inside and but of course you would get this between two pins of a connector and they will of course be exposed. The solder mask only goes up to the pads, but then between the pads and then with inside the connector itself. which is what has happened. It seems to be happening inside the Apple The water gets inside the connector and then just forms a creepage path across there and it just breaks down and it starts to become more and more conductive.

and it pumps more and more power into there and it can accelerate and get worse. And yeah, it's pretty ugly. Alright, I've given that like a couple of minutes. Of course you know you might leave it in there for weeks, a month or whatever inside your products.

So what I'm gonna do is just blow that away I won't brush it I'll just yeah, you can really see the carbonization or something between the tracers. It's still drawing eight milliamps Wow Check that you can see the see the smoke coming off there. now. remember 52 volts.

it's just still drawing 13 milliamps like we're we're talking like you know it's 600 milliwatts, half a watt. It's not much. does not like that at all. it's just completely broken down Oh Starting to get a bit smelly.

Don't set the fire alarm off I'll call it quits there. Damn. I didn't start recording. um what I've done here is another test because I thought it was oxidization on the existing traces on the board and I just I cleaned it with a steel wool, cleaned it all up so there was nothing on the board so it was just two pure copper traces on there.

I Just poured the water on like seconds ago and it instantly started to bubble up like this. perhaps more than what it did last time. So that's fascinating. So we've got some sort of electrolysis happen in here.

And between you know, two brightly polished, you know, pure copper traces on there with water that just came from the bottle? Sure, it's been sitting on my shelf for a while, but it is like, you know in their regular spring water. Once again. I'm getting that. you know, 1819 milliamps or whatever.

So it's basically in the order of a watt. Pretty much so. let's blow that away. You can see one of the traces is really getting green.

now. Hang on, let me show you. Yep. look at that.

Wow We've got electrolysis happening there. You can see it, you can still see it bubbling and smoking. That is really quite something. Just tarnish in a way.

completely. love it. Okay, one more test. Curious to know what happens if I increase the gap in there.

So I've pulled off the inner copper trace there. So exactly the same water as I had before. Let's go let's see if that does anything. Oops.

I'm an idiot. didn't turn the power on. Don't there we go. Yep.

18 Milliamps? Yep. Yep. So that's interesting. Even if you increase the spacing like that, it's not really going to help you.

Maybe the Apple engineers weren't at fault. It's just the nature of having a high voltage. You know, Reasonable current, reasonable power capability, power supply, and you add liquid, it's going to ruin your day. I Mean there's going to be a point like it's some physical separation distance, but we've got quite a large separation in there.

Like if I was the layout person doing that board and I wanted to add separation in there, it certainly wouldn't be any more than that. But we are drawing significantly less current than before. We're drawing about 7 milliamps this time, so there is a function of the distance in there. But still I mean that's like within seconds that just started reacting and it's all corroded you can see.

Yep, you can see that positive terminal over there. It's not a happy camper. So yeah, that's interesting. So as you can see, there may not be much you can actually do in terms of conductor spacing and creepage distance when you've got liquid damage inside your product like this.

So does that mean that there's nothing the Apple designers could have done well? No, that's not really the case. They could have taken various measures to not necessarily prevent this, but it helped mitigate the problem when you get water ingress. Is that their folder that I don't know that's still up in the air? By the way, this board view our software is just fantastic. This is the open version.

There's also a pro version as well. No wonder Lewis can like repair these Apple products quickly. When you've got this, you can cross-correlate pins with the schematic and everything. It's just gold really is fantastic.

Hats off to Paul Daniel So did they goof up by having the high voltage pin next to the data pin like this? Well, yes and no, right And your average designer might have come along and said well, okay, let's just leave out a couple of pins spacing to get extra creepage, distance, extra clearance and that that would be like standard practice, but as we've seen, that may not have been enough to actually do the business here. You still might have got arcing across here. So really, what you're needed to do is to ensure that none of your other chips blow up. So that's fault mitigation in another aspect of design that you may or may not consider.

And I don't necessarily blame Apple for missing that. Do you design your products to be tolerant to water ingress like this: Do you have them failsafe? For example, how would you make this failsafe? Well, this high voltage penny. You could still have it right next to this pin. or maybe one over this data pin.

but then yet, maybe could have made pin one a ground pin and then maybe you would have ground guard traces around here. You might even have exposed tracers. Yeah, I've seen this. This is quite common in very low current designs.

were seen it in, you know, Keithley electrometers and stuff like that where you'll leave an exposed guard trace around there that prevents that. The leakage from a particular pin in this case due to either moisture or water ingress on the surface of the PCB is then shunted to ground instead of to a sensitive MUX chip or a CPU that can just have the magic smoke released when it gets low enough impedance to go across. So you know, really and you might have had room in there to actually put some guard traces in there. But really, I would have left a pin here and made that one ground that just would have been.

You know, good design practice if you would. Thought of this problem, water ingress problem and you are trying to mitigate any potential failures of other devices in your product. And that's not necessarily that kind of standard practice inside products have. It's always a trade-off how far do you go to try and prevent these sorts of issues? So if you had a big guard trace right around going through this pin and right around there and like what as I said, once it gets under the solder mask, it's pretty fine.

But it's on the connector that you've got to worry about and it's on the connector where we've seen this horrible damage like this. It's just like it's just carbonized that completely and just and destroyed any chip that this was connected to either directly or other things via the protection diodes builder than a chip. And it's just yeah. it's nasty.

So that pin was ground and then the others were all data. Maybe you know this wouldn't be a problem. Sure, it might short out your backlight power supply, but then you could have like a Poly switching your backlight power supply or some other fuse that's much easier to replace. But Apple don't necessarily design their products to be like component level, trace and replace they do.

You know major board swaps and things like that. and that's one of the criticisms of Apple and other companies there at Apple aren't the only ones that do this And and you've got to ask. Well if Apple do know about this problem and sure enough, if they missed it, that would be fair misted in the design process. Fair enough.

But what happens if they then knew about it, they do a new revision of the board. You would upgrade that to fix the problem. Surely they would. Any you know, any competent engineer would look at that and go, oh yeah, that's a problem.

Whoops. Sorry, we'll fix it in the next revision, but apples seem to have gone backwards in this respect because the older designs actually did shunt at the ground and I think it blew a fuse or whatever and you know it was much easier to fix than replacing a big expensive BGA chip. So yeah, oops. and this leads me to a comment on Lewis's video.

This is from an older video where they talked about this and this person rainy days I think where they can see where the error arose in this case. Here we go. No single design engineer is responsible for the design from start to finish in a large corporation. That's very true, especially in terms of something so complex.

Mike and Mac. There's no way only just one person did the schematic and the layout and everything. Y'all see, you've got at least one person doing the schematic. The schematic might have even been divided up.

You've likely only got one person doing the PCB layout because cooperative PCB layout it's a great in theory doesn't really work that well in practice, but it's possible. but it's most likely that I mean that MacBook Board that was seen not particularly a complex layout at one. design engineer can easily do that. and it's better if you have the One Design Engineer Do it.

So they might have had a PCB layout person, but that piece of be layout person might have had absolutely no visibility into the schematic side of things. Anyway, let's continue the era started with the CAD librarian responsible for drawing schematic symbols for new parts and maintaining the library who incorrectly labelled pins 143 to be far apart numerically when physically they are close together. Now, this is interesting because if we have a look at the connector here, this is actually from the Apple schematic and it's actually drawn this. There's two ways to draw a schematic symbol like this one is A as a physic as they've done here, a physical representation of the connector and you can see the backlight power is up the top here and they physically show that as a separate pin.

They like that there's a box around here showing that these are all signal pins and this is a physically separate power pin. And of course, that's what you see on the connector. That's one big, larger tab pin. So they've actually I don't see any fault whatsoever.

From the librarian point of view, who would have drawn this schematic symbol? Now, you could argue that why did they number it pin 1 and 43 up here? Well, it might be understandable because you might start labeling the signal pins first and well I You know, and is that a thing? I Don't know. They don't show the other one down the bottom, which is interesting. So maybe there's a bit of inconsistency that other large pin down the bottom. So maybe there's a bit inconsistency there in terms of drawing the schematic.

but that's not where this problem originated. So I Don't put any fault on the librarian. In this case. if it is a PIN number like this, the person laying this out the pins actually, even though the pin numbers are there, they've physically put it next to that.

And any competent design engineer, especially ones familiar with the connectors in these types of products, you'd have to be when you're designing these types of products, they would have gone. Oh yeah, that's probably the larger tab at the top of the connector. And you've worked on other designs and you've got physical examples in your hand and like, yeah, it's like I can't see any error happening there. The person who drew the schematic knew that that backlight pin was close to this auxiliary power pin, but they might have had, or they may have had the decision of buying the connector they might have.

They certainly would have had visibility. This part number up here would have been mapped. They could have gone and checked out the datasheet in their Apples internal library systems, almost certainly would have linked to a datasheet for that particular product. There, The link to the 3d model and a photo and everything should be linked together in their system.

If it's not, I'd be very surprised, but they should have. The person laying out this schematic should have known that that pin was physically close to there, so that if you could blame anyone, it's the person who did the schematic knowing they would have known that pin was physically close. So that's the first goof You probably should have said. Well, let's add, let's skip the pin down there.

X Look, we've got lots of ground pins down here. Tons of them. You don't need that many ground pins to in this particular case. they're twisted pairs, so it's not like they're getting shield in between the individual ones or individual pins or anything like that.

So really, they could have shifted all that down and it wouldn't have. you know it wouldn't have mattered. You could have got at least pin 1 and pin 2 there as ground so that you could have a guard shield around her. But obviously they weren't in that mindset when they were designing this.

So I don't think that's a valid criticism. A completely different mindset. What the librarian had: I mean the library is probably following the company's standard. They probably would have had something like Oh connectors must be physically similar to what they look like rather than just numbering and 1 through 50 or whatever and then not have physical representation.

just one big role like this and not even having two sides separated. So I don't think the librarian was at fault then when this is passed on to the layout engineer. Personal Responsive. actually, rounding the design on the physical board here outs the design blindly and does not catch the area because you does not know I really care about how the circuit functions.

That can be very true. There's nothing that says a PCB layout person has to be. Well, they they have to have electrical knowledge. Okay, but they're not necessarily a design engineer.

I've worked with excellent PCB layout people way in the past and in fact this was a carry on. They were originally CAD engineers, they were autocad and then in the early days of CAD they'd come over and they'd be laying out PCBs without really much electrical knowledge. It's more important these days you've got to know about you know, impedance control and and twisted pairs and and termination loop area and star ground in and there's tons of stuff you have to know as a piece of your designer. So so I'm not necessarily sure the PCB layer person would route it that blindly, but I wouldn't blame them for not capturing that problem.

And I think if you're going to blame anyone here, it would probably be the schematic designer which is more of the main product designer because you could say PCB design is just more perform a professional PCB designer myself and I hate this term. but grunt work right laying out the board is kind of right. It's it's the bigger picture actually designing the schematic, designing the product, the overall look and feel and the physical form factors and how it all goes together. and the thermals and the whole you know things like that.

So the higher level design engineers they were really responsible for ensuring that those pins didn't have a guard ground between them. For example, all that the backlight wasn't fused I haven't looked at the rest of the schematic was is the backlight fused or whatever if it's not and it didn't have a ground path going to ground, then there when the ball goes through testing. Nothing is ever caught since nobody has the foresight to purposely try to short a design during testing and that can be quite right. But when you're designing and you've been designed an Apple, have been designing consumer products for a long time, they should know about water ingress.

especially Macbook computers, right? People sport poor stuff on the keyboard. It gets in there and it's just right. Ruin your day. It's a common failure mode, but is it the mindset of the company? This particular case? Apple That look, we don't care about if they users dumb enough to spill water in them.

Hey, we'll sell them a repair service where we'll replace the whole motherboard, replace the whole lot, or just sell them a new one. and that's you know. So Apple have some blame there from that point of view. But do you blame the design engineers for not mitigating against water ingress like this? Because it's not just the 52 volt supply.

You could actually get electrolysis happening on other rails as well, and that could ruin your day. So really, how far do you go to ensure that that backlight wasn't going to go into another chip? With hindsight, it seems obvious. Yeah, don't put it right next to a dunno fifty-two volt rail right next to a data pin like that's just yeah. Asking for trouble without having some sort of ground guard trace mitigation in there to shunt that power into ground instead of it being going into your big BGA processor or whatever.

So with hindsight, it's obvious. But do you blame them? Yeah, it's hard, right? But once they know about this problem, yeah, if they don't fix it, then I know that's that's a problem. Ultimately, this is a flaw. Jutsu, a misunderstanding mismanagement between multiple engineers with different functions.

This would never happen a smaller company where the zone is owned by one person from start to firm. So Apple being rich and large actually works against them because all the engineers that work on the board need to have the exact same design philosophy or at least adhere to the same standards. This is basically true. It's yeah, when you got, we subdivide your tasks up everywhere.

Kind of. You know it's common just to wipe your hands, throw it over the wall, so to speak. I Finished the schematic job done, throw it over the wall, throw it over the cubical wall, to the layout engineer, and then they do the layout. Then they throw it over the wall to the production engineers who then have to get it assembled and built and tested and all that sort of jazz.

So there is. There is that silo aspect to a large product design. But and like anyone in the chain here could have seen this potentially. and then were they did, they speak up.

Maybe they didn't say hey, I think this is a problem or hey I'm hearing on the forums that these notebooks are failing due to water ingress and it's always this line here and it's causing a problem. Maybe you should fix that on the next revision. or maybe we should be careful on the next revision. But then maybe they're rotating through a design engineer.

So the 2012 MacBook Pro design team may not have been the same as the 2016 MacBook Pro design team or one person left and or it's not documented or there's no company culture - at the engineering level to actually look at this sort of problem. Go. Hey. Ok, what happens if water ingress? Water ingress is one of the big fight.

You know you can do failure mode analysis on these things and you can argue that they certainly should honor. You know, a high volume, expensive consumer, big profile consumer product like this. Ok, a common fold. everyone knows it's spilling water and coffee and whatnot on your keyboard.

if that gets in, what do you know what happens? But hey, they could have tested this and looked around and went, hey, it's not a problem Like I said. Don't necessarily blame them for missing this, but once they know, should certainly fix it. This floor is still, of course unacceptable and can be remedied with better communication, a set of internal design philosophy standards, and that's certainly true. You can implement a fire analysis on these things where you can look at a design, and in fact, you could have somebody devoted to this or a team devoted to this once you've designed your first prototype.

If your product, right, how do we make this thing fail? Do they care about that? Like, or do they simply blame the customer for getting water ingress? And then even if you do identify this problem, well, you can. ok route this. Okay you? Okay, We're going to change the pin out of the connector. We're going to make this a ground pin.

We're gonna have ground guard traces around here. We might have some of the solder mask on the traces to actually do that sort of thing, but then it might happen somewhere else and there's but you seen the board here. Like you know, it's a reasonably complex board. There's just tons of connectors in here.

There's just tons of things going everywhere. and well, it's not just about the high voltage backlight, other issues, other electrical failure modes, and other night men. this is tons of stuff that can go wrong in here. but this one, with hindsight, seems blindingly obvious and you could argue.

And I wouldn't argue against it that. yeah, it's a pretty dumb designed to put the 50 volt supply right next to this like this. If it was a, you know, a 3.3 volt backlight or something like that, you know I had a large bunch of parallel LEDs and was just really high current. You wouldn't have had this sort of issue, although you might still get some electrolysis happening between the pins and stuff like that.

And yeah, that could ruin your connector or ruin your day. and you might have to clean it up or do whatever. But it's not going to go in and take out your CPU or what does it do to take out the CPU down here or the MUX chip? Apparently if you've got one of the designs, it only takes out the MUX chip which is a MUX I. Don't know which will chip it is, but it could be like that one or something and it's a much smaller and simpler chip to actually replace.

Then say the CPU over here the massive big BGA gob over here. Yeah, and then do you go even further and go? Oh well. we're gonna get look this connectors here and it's near this joint on the top of the keyboard and if water goes in here, it's always going to go around. I'm just guessing here from the fee.

I Don't know what a MacBook Pro looks like and assembled but like water always gets in here so maybe we should partially conformally coat this board. for example. that's an extra production process I Don't think I've ever seen and conformally coated consumer notebook computer. Maybe they did in those you know, Panasonic toughbook SPAC in the day or something like I like I Don't know, but you know that'd be a drastic over-engineering measure.

Audio add, physical rounded out slots in your PCB between the pins to prevent like you can actually do that. You could leave this pad off here and the pads on the connector. You could actually put a bigger routing path right through here. so you had no pin, one pin, but you had a big slot.

You could have it right down here like this. As a matter of fact, if you just left pins 1 & 2 unconnected, you could have a big massive slot down there and the pin just flaps around in the breeze. It's just happy. You know I need to be fine, just happily sitting above that slot.

No worries whatsoever. And then you just get this big so there's no creepy - across your board like that. I'd have to creep over to here. You can see that on this side of the connector, there's nothing over here really that you'd have to worry about.

You know, the creepage paths are very large, so you know maybe they could have done that. For example, that that would have been an obvious option. but then they could have done their water ingress testing for example and found that world water didn't get under these connectors and it could have tested fine and and ship it. And maybe it's only one in every 10 cases of water ingress where water actually migrates its way into this connector because people like you spilled something in there for example and it's got the top cover on the connector and then they went and you know, store their notebook upright.

they folded it up and you know I don't know, shoved into their bookcase or whatever. or did whatever stand-up vertical and then the water flowed off the board and then down into the connectors in it. Does it happen in every single case? You can't. So you don't know if they could have actually tested this and went.

Yeah, no worries, that's fine, and in that case, no. You certainly can't blame the engineers at all. So we really don't know unless whoever was on the design team here. But if previous ones, if they actually thought about it or is just luck, where it actually shunted that power from the backlight to the ground, whether or not they deliberately thought about that, or is just luck at the design and these didn't think about it in this one at all.

But no, there's certainly many cases where I've done designs and we've tested it thoroughly and multiple engineers look at it. We do failure analysis and we figure out all the different ways that can fail and it still finds some unique way to fail. Or we did actually test that particular thing, but it only fails one out of every ten times. And of course, you're not gonna attend test it ten times.

you're doing it a few times or you know, Murphy's Law get you every single time. So so anyway, it's an interesting discussion about the infinite variability in layout and things going wrong and whether or not you do fire analysis, Whether or not it's good enough, whether or not you're mitigation issues are enough. but yeah, this one seems a bit dumb with hindsight. but are they Are they to blame? Are the engineers to blame? Let us know your thoughts down in the comments.

So thanks Luis Um, who subscribed to I'll link it in at the end of the video and sharp they're there somewhere, losses videos and let us know what you think about this whole sure muzzle that is the Apple product design. and of course, but yeah, Apple really need to lift their game. They're not even off the floor. they need to at least lift them off the floor out of the gutter to make products more repairable and the right to repair bill which Luis is involved in and all sorts of stuff.

very important. And anyway, there's lots of interesting stuff and we could go deeper down the rabbit hole if you really wanted to. but I think we'll call it quits there anyway. Hope you enjoyed that video and you found it interesting.

If you did, please give it a big thumbs up. As always, discuss down below or I run the Eevee look for him. Catch you next time.