Hi I saw this on Twitter. Some leads are lighting up when they go through the reflow oven. Uh, what the what's going on here? A lead across a capacitor, Heat it up, it lights up. I gotta check this out.

So where does this come from? Well I heard about this from Uh. Greg Darville on Twitter. Highly recommend you follow. Greg does great.

uh, like close-up macro photos of like Smd and soldering among other things. Really great. Anyway, he says weird phenomenon during reflow, some green lead power indicators are emitting slightly. Seems like maybe uh, during reflows.

Solid polymer capsule is an effect on creating small electrical current. Does anyone know what this effect is called? Maybe Some further reading and he did a test with a 220 mic. uh, 50 volt electrolytic surface mount cap like this heat gun onto the cap with a lead and sure enough it lights up. So very interesting.

And Ian here, uh pointed out this Uh. Stack Exchange article about somebody saw these leads light up um in the reflow oven. But I suspect if you notice there's a thermocouple there and the thermocouple looks like it might be touching a pad that's coincidentally near all those leads. So I suspect that is not due to what we're seeing here because there's no capacitors.

Um, this is due to uh, capacitive electrical coupling. Whatever it is through the thermocouple, uh, through the Um shielded bottom and everything else. so I don't think, uh, any things. there's anything to see there, but there.

There you go. I'll link in the Uh thread down below, but fascinating. You can light a lead by heating up a capacitor. Hmm, let's do some experiments.

I've got an ammeter, I will hook it up in a second. I'll show you that um, it's on microamps range and I've got a, you know, a selection of Smd electrolytics here. I've just chosen a 470 microfarad 16 volt. Um, Jobby.

I don't know the brand of these because this is one of these, uh, just generic cheap ass uh kits. But it does have the split in the top there, which indicates um, that that is not a solid, uh, polymer capacitor. That indicates that it's an a, uh, wet electrolytic type capacitor, which it looks like from the photo is similar to the one Greg's using. Somebody else on Twitter um mentioned that it could be some sort of pre-charge slash dielectric absorption of the capacitor linking dielectric absorption.

Never done a video on that. I'm sure I've mentioned it at least many a couple of times. Uh, in videos. Anyway, if I hook this up.

um, I've just had this sitting here before. It was actually hooked up and shorted out with the load of the milliamp, uh, the micro amp input here, which is what? 1k or something like that. But anyway, I'm going to hook this up and watch watch the reading. There will actually be a charge.

It jumped up to a couple of microamps there, so there is some sort of um, dielectric absorption charge building up, but It can't be that, um, on its own because the lead would instantly just drain any of that away. Anyway, here we go. I've got my, uh, heat gun set to a hundred degrees. um, Celsius.

so I don't wanna. You know, take it to like reflow temperatures. yet I wanna do it. Just a low temperature.

So let's see what a hundred degrees celsius does. Here we go. Yeah, Yep. Sure enough, it's going up.

but not much. I mean, you know, 1.2 microamps? That's a sniff of an oily rag stuff that's half a bee's dick. There's not a huge heat sink effect in this because this is like a dull and, uh, plastic or whatever. But you know, look, heat is actually doing something to it.

So there you go. We're getting up to a micro amp. but uh, Greg said that he was seeing like tens of microamps or something. Okay, I'll ramp the temperature up.

Okay, let's take it up to 250 degrees celsius and hopefully, um, I don't melt any of my stick vice here. I don't actually know what the temperature rating of this is, but anyway, and point three. Well, yeah, yep, that something's no something's going on there. Yeah, I won't do that anymore.

Okay, let's try that again on a metal surface this time so we don't damage anything. Oh yeah, it's faster. Yep, yep, it's going higher here. We go here, We go.

Now we're talking. Now we're talking. Sure enough, once you get into the tens of microamps range, you should definitely be able to let light a really high efficiency. Ah, there we go.

It's dropping back down. It's dropping back there you go. So it peaked at about 30. that is clearly going back down.

Okay, so I've heated that up. So what Uh Greg said is that, um, it didn't work after he'd already heated at once. So I'll wait for that to cool down a bit, blow on it, and then we'll see. Oh look, it's gone.

Negative. Isn't that interesting? So there's some interesting physics going on here. Like my first thought was that, yeah, it was like a thermoelectric effect with the dissimilar metals Um, like in the junctions of the leads and the capacitors and all sorts of stuff, Right, There's bond wires in the leads, there's uh, you know, there's actual uh, metals used inside the leads, in the, and there's the foil in the capacitors and everything else, right? But if it goes up to like, uh, 30 micrograms or something or whatever we saw there, and then it comes back down, speaks up and this starts coming back down, then we've got something very interesting happening with. um, almost certainly the dielectric material in here, as if it's like it's boiling off or something, and then it's not going to happen again.

So if we reheat this and then we find it doesn't happen again, then that indicates that there's something in the capacitor. Most likely the dielectric that like is boiling off or something like that doesn't really affect the capacitance, because everyone knows that you can reflow these things. They're designed to be reflowed, not continuously, and certainly not repeatedly reflowed. but you can, certainly, uh, reflow them once onto the board and everything's hunky-dory They still remain their capacitance, Their remains their Esr remains everything else.

right. Here we go. But will it get back to 30 odd microamps and you saw like it's slow at first, but then it's sort of like the heat got internal and then it says going up? Here it goes. Here it goes.

It's accelerating, but will it get high enough again? It's not. It was going up faster than that before, wasn't it? It's but yeah, it's not getting. it's not getting this high. That's it.

And it's gonna drop right. It reached eight this time. Aha gotcha got you got you gotcha? And that's just oh no. It's yeah.

it's it's gonna drop. Oh, it actually started dropping drastically there. All right, it's cooled down again. Let's try it one more time.

At 30 odd microamps, the first time, I should be recording this. The only difference between science and mucking around is writing it down. Hey Jenna, Yeah, I don't think we're gonna get there. I think there's diminishing returns now.

Oh no no no. hang on, you can do it. And then it goes down. down.

down, up, up up. Is that just me moving? it? Is that just me? Like doing the airflow? Anyway, yeah. like we. we take it off and immediately it drops back down.

Okay, let's try that again. But I've got the 121 Gw here which allows us to, uh, just log the current here and this will, uh, potentially present a, uh, different load than the Bm786. Multimeters can have different um, shunt resistor values on the uh, microamp and current amps. so let's do exactly the same thing again.

So I'll start the data logging. Sure enough, ramping up 32 and we're going back down. Back down, Back down. Okay, it's cooled down enough.

Let's log that again. This is cycle number two and we don't expect it to go back to where we 30 odd microamps we got before. It's going to have a much flatter top on it. I can tell you that for nothing now now it's dropping, it's going to be a much smoother, more stretched out graph the like.

It's not going to have a really high peak on it, and if you're wondering what we did to that poor little sucker, even though that was, you know, like 250 was like a reflow, uh, temperature, uh cycle. Pretty fairly typical. So like 407 microfarads for the 470. um.

dissipation factor. What else have we got? We can give you the Esr on that bad boy 0.464 And for reference, here's a brand Spanker. I haven't cycled it at all. In fact, it's series resistance is actually, um, higher.

There you go. So yeah, I just haven't damaged this at all. As far as you know, your regular parameters go all right. I'm going to work with a known quantity now.

Uh, so I went to the bunker. All the best stuff comes from the bunker and here's some Panasonic Jobbies 220 mic, 16 volt. And there you go. For those playing along at home, you can look them up and you can get the data sheets.

I don't think we're gonna have a shortage of them for testing. And the Panasonic that's been sitting there for like 10 years? Here we go. What happens if we hook it up? Whoa. look at that.

That jumped up a lot. Wow. Okay, so here we go. heating up the panasonic.

Yep, she's rising similar to the other one we had so two entirely different brands and it's going back down. So what? It reached 16 points? Something there and this is our second time. No, there you go. it's dropping back down.

so check this out. This is absolutely fascinating. We have the data here. Uh, the orange one here is the no namer.

that was the first test and then the blue was the uh second test. So you can see how that the uh first test. It peaked right up to like 33 microamps or something like that. and then it just dropped fairly quickly and then here.

this must be where I removed the heat gun and then it dropped off and it went negative down here. But as you can see the second test I've had to sort of like shift the data here to because I didn't like line up the exact um you know thermal profile so they've just been shifted. You can see, I think it does take a similar amount of time to get to the peak though. um but then it just stays there.

It just like it a stays there and then this must be where I remove the heat. So then you've got the Panasonic one. The gray one is the first test here and you can see it's kind of got like a little humpy, sort of not a not harm plateau of like a front porch so to speak. and then it ramps up and goes off and it has a it doesn't have the same sort of nice profile that the no name.

It does it sort of like goes down a linear slope and then it's got a larger slope like this and then it goes negative and then it starts oscillating. Is there more data to that? Actually yeah I thought there was there. It is. Um yeah, there's extra data.

It went like this. now I can't remember if I actually took the heat gun off at this point we're at negative or if I kept it on there like that. but there's some interesting little negative action happening there. But anyway, the second test is the yellow one here.

uh, for the panasonic jobby and it actually ramped up quicker. If you, you know, I don't know exactly where the heat started, but if you shift it over, um, yeah, it's sort of ramped up to its peak quicker. and then it sort of went down. and then it did a sort of like a plateau kind of thing.

So it's a different profile to what we got for the no namer. So that's interesting. And these are the same heat, same heat flow site, right? You know, roughly the same distance. All the conditions are basically the same into the same load.

uh, meter. And yeah, they're different profiles. So there's something happening there interestingly, with the internal chemistry and or the physics. I don't know.

Um, it could be some obscure physics thing. It could be a combination of obscure physics and chemistry stuff. It could just be pure chemistry. It could be metallurgy type.

Um, you know, there could be some thermoelectric effect. As I said, um, happening there. It could be a combination of any of those, or all of those, I don't know. but it's It's absolutely fascinating, is it not? You can generate tens of microamps by heating up your electrolytic capacitors, and unfortunately, I haven't been able to find any leads.

I found leads that work at 30 microns. when I put them on my current generator, they work fine. When I put them on the cap either cap, they don't work. I found a red one, couldn't find a green one yet.

But anyway, now the reason why my leads didn't light is likely because they're non-linear devices. Yeah, sure, you can push that 30 microamps through. but if there's not the compliance voltage required um, then like the actual voltage generated on the cap in this case, then well, they're not going to light. So I definitely have to let that lit in room light with 30 microamps and it does nothing with this cap.

So yeah, you got to get a specific, um, like really high efficiency, uh, lead and stuff like that. So it's you know and don't be disappointed if you try this and it doesn't work. But you saw Greg's lead actually light up so I'm sure it works. I've just got to find one.

So yeah, an interesting follow-up experiment might be to actually uh, plot the voltage as well as the uh current as well. So then you can actually see the compliance voltage and the actual power. Uh, delivered the power and the capable power actually delivered from this thing. It's not much, but it is enough to light up certain types of leads.

but that is interesting. I might have to ask one of the capacitor manufacturers. I'll reach out because I have a contact and we'll see if they have any theory that explains this. but if you do, or if you've got, if this is like known thing and it's published somewhere or it's in some app node and buried away somewhere, then please leave it in the comments down below.

So thanks to Greg for finding that. It's absolutely fascinating. There's some weird stuff happening here, something very interesting, so if you want to experiment yourself, it's it's. pretty easy to do so It's good fun and that might do follow-up videos on this if there's enough interest and things come to light.

I'm here all week. Catch you next time you.