[0:00] If you watched the previous video, which should be appearing somewhere around here hopefully,
[0:17] you’ll know we’ve been playing around with some high-voltage circuits to drive these
[0:20] LED filaments, which I got out of some broken light bulbs.
[0:23] It’s a very simple circuit, there’s a MOSFET, an inductor, a diode, and a smoothing capacitor.
[0:29] And we’re simply switching the inductor off and on at a higher frequency.
[0:33] So currently, I’ve got it hooked up to my signal generator, and we’re running a duty cycle of 10%.
[0:38] So if I just ramp up that duty cycle, then we can get up to a pretty high voltage.
[0:44] So these diodes start to light around about 90 volts, and we can go up to around 95 or
[0:51] 96 and get sort of fairly bright light.
[0:54] This meter only starts to register current after about 20 milliamps.
[1:00] So this circuit works pretty well, but as I say, it is quite Heath Robinson.
[1:05] So you can see all of this stuff on the breadboard.
[1:09] This is the remains of a capacitor that got blown up.
[1:13] So the problem is, I’ve only got some fairly low voltage capacitors.
[1:18] So maximum of 100 volts, and I went over 100 volts.
[1:23] It worked for some time, and then the capacitor exploded.
[1:27] Now I don’t know if that was a failure of the diode or theOSFET, because theseOSFETs
[1:34] have also been blowing up.
[1:35] So I’m just testing one of the failed ones here, and it’s turned into a resistor divider,
[1:41] which is not ideal.
[1:44] So I thought what I’d do is invest in a proper DC/DC converter.
[1:47] So I bought one of these.
[1:49] Now this will go from 45 to 350 volts on the output.
[1:55] I have both the negative and positive, so in theory we can get 700 volts out of this.
[2:01] Now I have seen some videos on YouTube.
[2:03] These have been around for a couple of years, so there’s plenty of videos to find.
[2:08] You can get the negative to go almost minus 400.
[2:12] So if you go from negative to positive, you can get around 800 volts out of these circuits.
[2:17] So that’s pretty impressive.
[2:18] Well obviously this one’s dead.
[2:21] I started using it, it was working, but it was making a very strange sound.
[2:25] It’s sort of clicking sound, and then it just failed.
[2:29] And the input is now shorted together.
[2:32] There is a vaguely suspicious burn mark here, but I can’t remember if that was there before.
[2:37] Now this does get pretty hot, but not excessively so, and most of the video reviews I watched
[2:43] talk about how hot this board actually gets, and it does get too hot to actually touch
[2:47] when it’s running.
[2:48] There was a connector for a fan, which may be useful if I actually get another one that works.
[2:54] But let’s pop this under the microscope, and see if we can work out why it blew up.
[2:59] But before we get into the PCBs, let’s talk about PCB when they do multi-colored PCBs,
[3:05] they do flexible PCBs, they do 3D printing, and they do CMC machining, they pretty much do everything.
[3:12] So check them out, there’s a link in the description.
[3:16] So I think the first thing to notice is this board’s pretty disgusting really, it could be improved.
[3:23] I mean my soldering’s not great, but this does look pretty bad.
[3:27] I mean, could that be a short there?
[3:29] I mean, I don’t think it is, but it definitely looks a bit crusty.
[3:33] So I think what we should do is clean this board up, and have a look at it once it’s nicely cleaned.
[3:39] So a bit of isopropyl alcohol, and a toothbrush, and our board is now nice and clean, nice and shiny.
[3:47] So let’s look at what we’ve got on the board.
[3:49] So let’s see if I can read the numbers.
[3:51] So that’s a 7-8-L09, so that’s a 9-volt regulator.
[3:57] Here’s the switching supply, so UCS3843B, I’ll try and dig out a datasheet for this.
[4:07] LM393, so an op-amp, let’s see if we can find a schematic online, there’s a few diodes dotted
[4:13] around, and that’s kind of it, really.
[4:17] So just checking that our short circuit is still there, and it is, so I’m getting a
[4:22] beep from the multimeter.
[4:24] So yeah, definitely still have a short circuit.
[4:27] Let’s have a look at the output side of things.
[4:31] So we have these two diodes, so let’s check these.
[4:35] So I think this diode is the positive supply output wheel.
[4:42] That seems to be fine.
[4:44] Yep, that’s definitely working okay.
[4:49] And then this one, well, that’s a bit weird, this one seems to be a short circuit.
[4:57] So that’s definitely not what we want.
[5:00] So something wrong with this diode, now I don’t know if that’s a different problem from the
[5:04] short circuit on our input, but it’s certainly not right.
[5:08] So what I might do is I’ll remove this, and I’ll probably also remove the MOSFET and we’ll
[5:15] see if we can see if they’re working okay.
[5:18] So I also want to take a look at this, what could be a burn mark?
[5:22] I’m not sure if it is looking under the microscope, but let’s see if we can peel up this insulation
[5:27] and see what’s on the side.
[5:29] It’s quite difficult to actually get this off.
[5:33] This transformer seems quite well constructed.
[5:35] Well, that looks completely fine, yeah, I can’t see any problems there, so not the issue.
[5:43] So let’s get this diode and the MOSFET desolded, and then we’ll do some testing on them.
[5:48] So I’ve got the diode here, and I’ve also got the MOSFET, right of solder on some little
[5:56] wires to the MOSFET because they were very short.
[6:00] Let’s test out the diode first, let’s go into, sorry about that, clicking, let’s go into
[6:06] the diode mode on the multimeter, and I think you can see that, and let’s shine on you slightly.
[6:13] That’s still pretty bad, there we are, that’s better, you can see that okay.
[6:19] So let’s see what our diode is sharing, so going this way, yeah, well that’s definitely
[6:29] not right, so we have 0.05 volts drop across the diode, and that’s pretty much the short
[6:39] circuit isn’t it, and then going the other way, what do we have, yes, short circuit as
[6:49] well, pretty much, pretty much the resistor isn’t it, let’s see what the resistance is,
[6:53] if you can actually measure such a low resistance, 30 ohms, so our diode has turned into a resistor,
[7:04] not very good, and then let’s see what our MOSFET is, well yeah like the MOSFETs that
[7:13] I’ve destroyed, this one’s turned into a voltage divider with a very low voltage between pins
[7:19] two and three, so that would be the gate, so drain and source are basically shorted together,
[7:27] so also not very good, well yeah this didn’t last very long did it, yeah not ideal, so yeah
[7:37] we’ve destroyed it, I’m obviously more destroyed, now I’ve removed these components, desoldering
[7:42] the MOSFET, I lost one of the pads, desoldering the diode was pretty okay, I should have
[7:49] probably used my hot air workstation for the diode soldering, but I tried to do it with
[7:54] the soldering iron, so I’ve ordered another one, a replacement, hopefully that will come
[7:59] soon, and we can actually do the experiment, we initially set out to do, which is lighting the stupid filaments.