Hi. If I don't sound enthusiastic, it's because I have to shoot this entire video again because I just shot oh how many? like 35 minutes one take. um, a continuous thing and my audio wasn't you can see I'm supposed to have waveform there. Go right? So in a previous video linked up here and down below and at the end.

if you haven't seen it, highly recommend you do. It's an aircraft transponder tear down and I had lots of comments and also emails about this one from people. Um, saying oh, we spotted something in the video and I thank you. I love getting comments and things and emails and messages when people spot something in the video.

but in this particular case, I did actually see this and there's a reason why I didn't mention it in the video. Anyway, let's take a look here. Let's go full screen here. Oh, look who we have down here.

I shouldn't use that. He's haunting me. Louis Rossman, He's haunting me. Look at him.

Ah, shouldn't have had the green screen then I would have covered him up. Anyway, go and watch Lewis. Let's go right? So um, here is the particular thing I had. Lots of people comment, uh, saying dave, did you see these this short? This short between two pins up here? Oh, what's the deal with that? Um, there's obviously you know this is dodgy design.

Look at the short there and um, yeah, there's a reason that short is okay. it's actually designed to be there. And I thought I've mentioned this many times over the years in many videos. I couldn't even point you to, but I'm sure I have and it's actually a common design technique.

But I thought, you know there's always people who are who just aren't aware of this. They think that it's just a, you know, a manufacturing fault. but it's not. So let's actually call up a higher res photo of this.

This is the, uh, a 4k screenshot and sure enough, look in there you can see that there is a definite solder short between those two pins. So I did actually see this when I shoot in the video, but I didn't really mention it because the video was already going to be long enough and I've mentioned it many times over the years. And uh, so why is this not a problem and why is it a deliberate design short? Well, it has to do. When I immediately saw this and I went, oh, it just that's not a problem because I the solder mask expansion.

Now, I've done videos on solder mask expansions before, but well, in this video, we'll be going over that again. Okay, you can see that there's no solder mask between pins here and when you get a short, which just so happens to be practically exactly, I can guarantee it's exactly in the center of these pads as we'll go into like that with no solder mask. you know. Aha, you get 99.9 confident that short is supposed to be there.

So let's find out why. By the way, check out that bodge there. It looks like if they like lifted what have they done there right? So you can see that this chip is a classic example of having solder mask removed. You can actually see it like that it's been removed as one big square which will jeez my my five-year-old could follow lines better than that.

There we go, that's better. You can see the square solar mass expansion so you get that's the green stuff. You get no solder mask between pins in there and that can actually potentially cause a problem because when you get your when you solder this chip either hand solder it or drag solder it or reflow solder it however you want to solder this thing. If you get too much solder on there it can just bridge between pins and the smaller the pin pitch gets.

And these modern newfangled devices these days you know less than like point five millimeters pin pitch. This particular one. this picks uh C 756 that's a 0.5 millimeter pin pitch so that starts getting down there. But once you go below that as we'll see shortly then you know you can really, uh, really come a gutzo with shorts between pins.

So it's important to try and have solder mask between pads. But you can't always do this as we'll discuss shortly about various manufacturer tolerances and pin pictures and solder mask expansion rules and minimum solder mask Uh, rules and and manufacturing rules and stuff like that so you can't always do it. But anyway, this board dates from about uh, 2000 or something like that. So let's go to a Pcb, shall we? I just picked this is just a random example.

I'm using that circuit studio here for no particular reason and we can actually see that. this is because people are asked uh, it's a look, looks like a four channel, uh Pc, um Rs 485 serial interface. There you go, but it has a nice big quad flat back here which we can experiment with and if we go right in here, we're in the 3d view in the 2d view here. we can actually go into single mode here so we can only see the blue top layer.

You'll notice there is no solder mask expansion around these pins and if sure enough, if you go into 3d view, you can see that there's no solder mask expansion between pins. So ah, look, Oh, look at this. Oh, there's a you can see the render thing. You can see how the solder mask expansion this is the white one here goes big over here and little over here.

That's the perspective. That's the perspective change. Isn't that nice? I really like that. Anyway, you can see that there is no solder mask expansion around that pad.

It is basically the solder mask does not. Solder mask expansion means it expands past the width of the pad. The pad is the copper one there, that gold color and the blue solder mask. none of that green rubbish.

This time. the blue solder mask just does not extend beyond that And we can. Actually, I'll show you what it does look like. We can go here to the individual pin.

We can specify the solder mask expansion value. In this case, you know you can come from the global rules. You can do it on a global basis or a chip, basic component basis or whatever. Those depends on your package you're using Or you can do it on a Pm by pin basis.

You can see it's zero right? But if we set that to 0.1 millimeters, it doesn't sound like much, but it will be. Watch. Whoa. That's heavy.

Look at. Look at that. All right. You can see how huge the pin there it is down there.

that that's the white stuff. That's the fiberglass right? So solder mask expansion like that? Okay, so you can see even 0.1 millimeters was a huge solder mask expansion right? So we can go back in there and go like point. Oh well, let's go 0.025 which is you know a reason it might be a reasonable solder mask expansion value. Okay, 3d view? There you go right? So we can do this Actually on.

We will do this on a global basis, right? So let's go into our rules up here. We can Go into our solder mask expansion. it's currently Zero. so global we'll just do 0.025 millimeters.

Okay, so there we go. Let's change that. So we're going down here 3d view mode and you'll see that we have, you know, a reasonable amount of solder mask expansion around the pads. you can see the tracks, individual tracks coming off the pads there.

Okay, so you know we now have solder mask between the pads and we can actually measure, uh, the difference in there too. Between there, there you go, that is actually at naught point one millimeters or fourth out. In fact, this is a good example of where you can come at us. guts are and I've done a video on this.

but I'll go over here again because it's appropriate. Um, if you I use Metric uh, for all of your board stuff, which you can do, there's nothing wrong with that. I'm still. you know, old school.

I've been doing this almost practically my whole life, right, as well as, uh, professionally. So uh, you know I. I still use uh, Imperial thousand Mills for track and space. but I use Metric for board sizes, hole sizes, and for uh, you know, Smds.

They're all in Metric these days and stuff like that. But it's still like doing track and space in Thal just for you know, old school, but some manufacturers Pcb manufacturers will have. they'll specify Imperial although and specify Metric. or they'll try and specify both, right? So let's say the manufacturer specified a minimum solder width of, uh, fourth out They say, no, we're not going to manufacture below this value right then.

Uh, if you did all in Metric, you know you're going to be using like 0.1 millimeters. That's fine, that's near enough. 3.937 Geez, that's near enough. But no, um, your manufacturer.

Especially if you're getting a cheap prototype service. They will just reject your board automatically because it's less slightly less half a B Dick, less than four thou. It doesn't matter because you're sharing the panel with, you know, a hundred other customers. They're not going to dick around if they've got automated software which checks all of their parameters, and if it's under that fourth hour, you can come a gutser.

So there you go. Um, and just be careful with that. I've done a whole video on that, so that's a good example. But anyway, this one should be easily manufacturable, right? And you've got some solder mask expansion.

Now You want some solder mask expansion around the pad like this Because the when they manufacture the board, the copper layer, and the solder mask layer, there can be like little alignment issues like that, right? They can be tiny little alignment issues. and if they happen to be misaligned, which is a normal thing, especially on your low-cost boards where they're not, You know, taking a huge amount of care there, you know there's a reason you get the board for two bucks or five of them for two bucks, right is because it's a really cheap, uh, service, right? That you can pay for a more high quality service and they'll take more care in the alignment and the tolerances are better and all that stuff sort of stuff. but you'll really pay for it, right? So on the cheap, uh, services? Um, then if you've got no solder mask expansion, no, then your solder mask will cover some of your pad and when it covers some of your pad, then the solder mask and then the paste eye can be applied on some of the solder mask. and then you're going to get little solder balls everywhere and it can really ruin your day.

That's why you should be looking to get, um, you know, some solder mask expansion. Always okay. In fact, some manufacturers say, in fact, we can go over here and we can have a look. In fact, let's go to Jlc here.

Let's go to their solder mask stuff and have a look, right? And they're just a cheap prototype manufacturer and this is their I Believe this is their prototype capability. Is it or is it just the general capability? I don't know. Anyway, they might have different capabilities for their higher cost production panels and for their, uh, cheap prototype servers, so just be aware of that. Okay, but in this particular case, they don't actually specify here: a solder mask minimum width right? Or a solar mass slither it's sometimes called, but it can be inferred here.

But look, they actually tell you, you must have a solder mask opening a minimum of 0.05 millimeters right around pads. So right there, what we thought was a manufacturable board. You know you might look at this. Oh, that looks reasonable, right? 0.025 millimeter solder mask expansion.

Bingo. This would should don't be surprised if it gets automatically rejected by Jlc. in this case that they go. no, you don't have adequate solder mask expansion on there.

So oh bugger that, you'll have to go back into. uh, your rules in here. Your global rules So you know you've just been rejected. You thought you finished your board, and you haven't.

You've come a gutser because you're uh, you forgot to match their, uh, actual requirements. So 0.05 millimeters, right? So there's our global rules set to match Jlc's requirements. Let's go in here and have a look. Oh geez, look at that.

That's thin. That's thin as right. We've now got a tiny little solar mass slither in there, which is what we can actually measure that near enough we don't have to be exact. Uh, we're talking 2 Mil 2 or 0.05 millimeters, right? So 0.05 millimeters? Does that meet Jlcs requirements? Let's have a look.

So they don't actually specify as I said, a minimum width, but 0.05 so they imply point. You must have a minimum of 0.05 here between copper and your solder mask. And here, they specify minimum copper to copper, right? So 0.2 millimeters minus 0.2 times 0.05 is 0.1 millimeters. So from this, we can infer that our 0.05 millimeters is going to be acceptable because they didn't actually specify a minimum.

and this is quite common, right? They may not. Actually, some manufacturers will, some won't. they will. Some will have a minimum.

Slither in there. They just don't like it. If it's too thin, they'll reject it. But in this case it sounds like Jlc will just have a go.

They don't care, right? If the solder mask doesn't work, it doesn't work. That's your problem. That's not their problem, right? And when you've got soldermask that thin, it may not like it. Adhere to the fiberglass or whatever right? It could just easily peel off or simply.

But if you've got your solder paste on this pad here, and this pad and this tiny little thin bit of solder mask there, then uh, yeah. it could easily bridge across. It's almost as if you're having nothing. Some solder mask is better than none.

But if you have it too thin and your pins are too close together, then you know the thin solder mask isn't going to help. But anyway. Um, actually, don't pause this video. Now it's going to say pause this video.

Don't because that'll ruin the watch, time, metrics or whatever you know. audience engagement metrics. So after this video, go get a Pcb with solder mask. Doesn't have to be blue, could be you know that green rubbish or it can be black.

Anyway, go and try and put some solder onto a solder mask. The whole idea of solar mask. It's well, its purpose is to prevent uh, solder mask over bare copper smobc. It prevents oxidization of the copper, but it's one of its primary purposes is to, uh, stop shorts between pins.

Like this solder just does not adhere to solder masks. Try and put solder onto a solder mask. It'll just ball up right and then it just fall off. You know, Or do it pick off because the flux will kind of hold it there.

but you know you can just like flick it off. No problems once. it gets everywhere lightly. Little solder mass balls, pain in the ass.

Anyway, so yeah, um, that's the purpose of having solder mask And that's the purpose of having that between pins is to prevent uh, shorts between this. But anyway, back to the original question Dave, Why is there a deliberate short in here? I guarantee it's a deliberate short. I'll tell you. let's go in here.

Here we go. Here's an unused pad, right pad 22, so let's actually just change that so it's actually ground. It doesn't chuck a wobbly when we try and try and connect a track to it. Okay, so when we're laying out a border like this, Pcb designers, we love to use snap grids.

If you're not using a snap grid, you're not a proper Pcb designer. Okay, your designs are going to be crap, right? Snap Grids: I've done whole videos on that of what you know, the best. snap. You're always changing snap grids when you're doing a layout because there's not just one snap grid to rule them all.

So we can go like 0.05 millimeters snap grid so you see so you can see they're the little dots in there. Okay, so everything's going to snap. So when you're actually routing traces, you want them to snap and you particularly want them to snap to the center of the pad. You'll notice that when I move my cursor like this down, Well, look, look, look, it's it's trying to drag me.

Oh look, it's dragging it right. It's automatically snapping to the center of that pad. It wants to be there. And trust me, you want your track to snap to the center of pads.

That's how you want to do it. And uh, so a natural thing. If you had two ground points like this, well, you should actually just connect that up to the ground plane up there. The copper paw, right? But let's say right, this layout was tight as a nun's nasty.

It was just really, you know, so tight that you could not route uh, traces up here. Okay, then there's absolutely nothing wrong with going. Well, in theory, there's nothing wrong with going between pads like that. and that's an age-old tradition of doing that right.

And um, that's and that's what the P I guarantee you. That's what this Pcb designer has done here. So if we go to our 3d view, you can see the trace in there. So let's just change our solder mask expansion back to uh, what was it? You know, 0.1 millimeters or something like that, right? Something crazy so that there's no solder mask between pins on there.

That bingo. We now have a copper short, a deliberate short between those two pins. and when you see lack of a solder mask between pins and a short that looks like it's directly in the center like that, then you know you can almost 99.9 Sure that that is going to be a deliberate, uh, trace across there, so no worries whatsoever. This is why I didn't mention it in the video.

It's just a common Pcb designer thing to do that if you're running out of room. In this particular case, I will actually criticize the designer because this truck, look, there's a whole area of room up here. Okay, this particular trace could have come up here like this and gone like that and up. Or it could have, you know, simply got up there like that and across like that, leaving room for this one to come across and drop down here as well as down there.

And I can show you that. So let's assume that we were going to do this on our board here. and let's say we had absolutely no room. Okay, but we had no room to route it under the chip here.

But we want to route it outside the chip. So let's get rid of this. uh, polygon. it's gone ski.

Okay, then we can actually route our trace on here. We want to drop. that's too thick. Okay, we want to.

Ah, let's none that metric rubbish. Okay, let's let's just say we had a five foul trace like this, right? You'd go like that out and you'd go like that. Haven't properly set the snap grids and things like that. But there you go, right? and you can see.

we've automatically got a design rule. Online design rule. Uh, minimum. Um, between pad and track here.

So yeah, that's doesn't matter, but you can see. Yeah, so you know a proper Pcb design. If you can route the traces out like that, it just avoids any confusion whatsoever. especially.

um, an optical camera inspection system could actually see that. could actually deliberately flag that up. But of course it would be the same on every single board. so you just mask out that error.

Um, if you're doing an optical camera inspection, What an optical inspection system. As soon as the ball comes out of the other one, so it would go along the conveyor, it doesn't sound like that, but you go along, the conveyor belt come out of the oven, it's still smoking hot and then it might go into an opt-in a manual inspection. Then it might go into an optical inspection machine. big camera on top, and specialized software.

It takes a high-res photo of the board and then it just analyzes and looks for things like shorts and it'll compare. but in this case it'll probably compare it to a reference so it may not pick that up. But anyway, might be intelligent enough to go. Oh yeah, there's a short in there and you know something like that.

Or you know, somebody else inspecting the reporter? you know, somebody in youtube comments could go oh, there's a short in there that products crap when. no, it was actually designed to be that way. So anyway, the moral of the story is, you know, don't try and do that. but I'm still guilty, guilty as charged Of, you know, just put in between pads because if you've got soldermask between your pads, which you should be doing anyway if you can, then it's not a problem.

It's only a problem when you remove the solder mask like that and you'll notice that if you do it on a pad per pad basis, there's like it. It move like it's got these little half moons like that, right? It follows the shape of the pad, but this one over here. Although these are square pads, so it's hard to say whether or not this or they've done this manually because I've you know, I've seen that done over the years. I've even done it myself on occasion.

Depending on what package I'm using. for whatever reason, you might actually deliberately put a big square of solder mask like that so you could actually manually do that. But generally you know that's you don't want to do manual stuff like that on a board, so on the top solder mask layer there, you could actually, uh, go in there and just put manually if you wanted to like a big square. There's occasionally reasons that you go in and manually tweak stuff like that, but generally you don't want to.

You want to do it on a global basis or an individual component basis. especially if you've got a critical part like a Bga or something like that. You don't want to be using global rules. Otherwise, you'll come a gutser so you want to, You know, specifically, have that chip have its own local uh rules just for that chip for solder mask expansion.

But anyway. yep, that's all solder mask expansion. so that's all it is. I know, I've waffled on for quite a lot of time and all this is covered in many, many videos I've done in the past, but that is the reason why.

uh, you get shorts like that on a board. It's deliberate, in fact, I guarantee it's deliberate. In fact, I guarantee so deliberate. I'm now willing to go to the bench.

Well, I'll solder with the solder up. I bet you there's a copper trace there. Let's go. All right here we go.

Tigano Microscope. Let's zoom in. Where's our short? There it is. It's all a bit hairy.

scary, isn't it? It's been in an aircraft for what 20 years or something. So just put some flux on there because there's never enough flux in the bloody, um, solder wick. because this stuff just, uh, dries out ages and not like a fine wine. Not that I drink wine anyway.

let's wake that up. Ta-da There you go. Needs a good clean. Oh, this actually was.

I can. I forgot. this actually was conformal coat too, by the way. Probably why this looks a bit janky.

Yeah, there's absolutely no doubt that that down in there see that is copper. Genuine copper. That is not a short. Well, it's a deliberate short.

It's supposed to be there. But as I said yeah, poor work on the Pcb layout. That trace should have gone up there like that. and they should have just went around there knowing that they were going to have no solder mask expansion in there.

So there you go. Um, I hope you found that video interesting. Waffled on a bit. It's been covered before, but a lot of people didn't seem to reload.

So please, no more comments in future videos when you spot shorts like that. Um, and hopefully you can now recognize them and go. Oh yeah, I'm pretty sure that's deliberate, although you know, not always. You know Murphy can get you and you think that it's all you know just happens to be right in the middle of the pad and there's no solder mask, but that short shouldn't have been there.

But yeah, this particular case. Yep, that's a deliberate design thing, snapped between two pads. Absolute classic. So now I've got a video to link to people when they comment on this in the future, because I'm sure we'll see that because I'm still guilty of going between pads like that.

But anyway, there you go. Hope you enjoyed the video. If you did, please give it a big thumbs up. As always, discuss it down below over on the Evblog forum and check out my alternative platforms here if you don't particularly like the Youtubes Catch you next time.

I have no idea if this was actually better the second time around. I think the first time the first time I shot this, I nailed it. Shoot it again. Damn it.

You.