Let's take a look at a low-cost electronic ballast for fluorescent lighting from AliExpress This cost three pounds including of shipping and another 60p was added on by the English government for their massively efficient running of the country. That was sarcasm. So this is by far one of the flimsiest and latest ballasts I've come across. I Guess these days there's not as much demand for these now because we're moving away from fluorescent lights to LED Although having said that, in many ways, fluorescent lights lasted longer than many of the LED alternatives.

so this one is marked as being for one 40 watt tube now. I'd like to demonstrate this, but I do not have a 40 watt tube and the close I've got well is 13 watt and that would not be very happy being connected to this. Maybe it blew the electrodes at the end. However, the first thing I noticed about this is that it's super lightweight and flimsy.

The cover on clips and is excruciatingly thin inside. We have the circuit board with a plasticized paper in the shoe liter at the back. It's very common construction for these things. Let's Zoom down this a bit more.

and the first things I'm seeing here, the maintenance coming in. then I see a discrete project far. I'm seeing two capacitors to 250 volt capacitors so they must be wired. They are wired in the series.

here. there's the two transistors that do the push-pull thing. Uh, this is probably the capacitor that's looped. It is Loop between the heaters at the end of the electrodes just to pass some current through them.

Where is the rest of the current? limiting though, Is it purely done by the inductor here There She? It usually charges and discharges a capacitor, but you know this will be revealed because I shall reverse engineer it and we'll see what the circuit is like I Was hoping this might be one of the more complex ones with a diac or that might be a Diac there. Um, okay, tell you what. I'll take a picture of this. both sides are more.

Reverse engineer it and see what the circuitry looks like. One moment, please, reverse engineering is complete. Let's explore I shall zoom in in this just a little bit just to make it easier to see. That should do it.

Things worthy of note on this very standard electronic fluorescent ballast: Very basic one. The incoming Supply goes to these diodes. here. the the four diodes form a bridge react for fool rectfar foe Redfar and you've got two smoothing capacitors.

They're right, about 250 volts each at 10 mega far, but they're in the series and it's interesting to know a center mid point tap is taken off these and goes to one end of the tube, which is quite unusual. Uh, here is the main current limiting doctor that limits the current through the tube and there is the feedback Transformer that arranges the switching. It senses the current flowing through the yellow winding and it's got a red and a blue winding smaller one and uh, that drives the base of either of these transistors, giving a push-pull effect. Anything else worth mentioning here? Yes, this component here is indeed a Diac.

It's a classic system used to start these fluorescent ballasts in a controlled manner. Um I Think we should go straight to the schematic for this. Interestingly I've not got a trip here to test this because it is uh for a 40 watt tube I don't have that four foot tube. Uh, it does say power factor of 0.92 I'm not overly convinced about that I think they might have just printed that one just for sure, but without being able to test it I cannot tell I've color coded some sanctions this just to make it easier to follow.

So here's the power supply. the AC comes in goes through those diodes positive to that. real negative to that real. but that has a midpoint there tap going over to one end of the tube.

so it's going to be about in the case the UK It's going to be about 340 volts on this rail. zero volt circuit reference in this year will not actually zero volts and then a midpoint voltage of roughly 170 volts going to the tube. The tube has the classic little capacitor 4.7 nanofard between the two heaters. and then it's got the current limiting inductor here.

that's the big huge choke inductor and then it's got that little Transformer which I've color-coded orange here. Thanks for watching when you start this up initially, the tuber initially has a high voltage. it requires the electrodes to be hot to be emitting electrons and therefore lower the voltage reduce the cathode drop. So initially if the troop hasn't struck, then current will flow through the heaters through this capacitor here, and the heaters will heat up when the tube strikes, and it happens virtually instantly and the voltage across the tube will drop to about 100 volts typically, and that will result in current continuing to flow through the heater electrodes, but just at a reduced level and it just maintains them at that hot state.

For Uh running the tube, it's treats the tubes quite well actually electronic ballasts. So let's take a look at the startup circuit first. that is based I've colored it purple here. that is based on this resistor charging this capacitor up.

so initially when you turn it on, it's unlikely that their circuit will just start oscillating itself. So in this case, this resistor charges up this capacitor until it reaches the threshold voltage of this diac, which then triggers at Round probably about 32 volts and it suddenly dumps that small value capacitor into the base of this transistor, turning the transistor on as soon as it does. So, current starts flowing um through the tube or the heaters through the inductors, through the sense coil down through this transistor to the Revolt rail. but it also pills the capacitor circuit.

The Uh charge capacitor here if it's not dumped completely through this uh, the Diac, it pulls it down via this diode and this basically means that once this circuit is oscillating um continually, then because this diode is being pulled down to the zero volt reel continually every time this transistor turns on, it means after it started oscillating then although this resistor is still trickling current to this capacitor, the capacitor current can't charge up because it keeps getting discharged via this diode that is the starting circuit when it has started and current is Uh flowing. The using conventional current flow is going from the midpoint of the capacitors through the tube and it's going to the zero volt rail through this Transformer and in doing so, it induces a current in this Transformer which then drives the base of that transistor. These transistors are both Npn. Normally, you'd expect an Npn to be switching to the zero volt reel like this one is But.

Ultimately, because they're not relying on a emitter reference to zero volts, they've effectively got a closed circuit here with this little coil. So the coil itself, the little Transformer winding is actually powering the transistor directly so they they could be treated as almost like isolated sections of circuitry. And now here's the odd thing. Normally there'd be a capacitor in a series of the tube, and when the capacitor had fully charged or discharged, cut it would cease to flow on that half cycle of the oscillation and the current through this Transformer would drop to zero and then it would slam into reverse.

so to speak. Once it dropped zero, it's turned this transistor off and it'll turn this transistor on. but in this case because it's got a center tab. I Guess they're possibly doing this for the lower voltage to minimize the amount that gets dropped across the rest of the circuitry here.

Um, but because of the way they're doing that I Reckon that the main limiting factor is the saturation of this little Transformer the Little M feedback to ride with the yellow winding in the tube windings, the red and the blue because when it saturates to the point that it can't couple current across to the other coils, then it will have the same effect. and uh, it will sort of slam into reverse. when the magnetic field in this collapses, it will start driving this transistor, which then pushes current through in the opposite direction and uh, then induces current and its own base drive, but also make sure that this one is shut off. It's notable, there is a little diode here and it's positioned with these resistors such that, um, when the voltage in the opposite direction in the coil, it's going to just create a shunted loop here, and it's not going to actually try and reverse bias the transistor in Anyway, These resistors then will actually help them at the current.

It's a very efficiently designed circuit, so the net result of that is that, uh, it basically alternates this line to the positive and negative rails. Uh, that's more or less it except for this bit in pink, which someone speculated was for providing a sort of like a ref sort of voltage reference to the transistor to be able to turn on. but in reality because that's provided by the little coil that does the transistor directly. Uh, I I Think this is actually filtering I Think it's a just a very simple filter to provide stability.

Maybe it helps set the frequency, or maybe it just purely is to avoid harmonics, orbs of high frequency that could end up resulting in false triggering of transistors or even electrical noise. but I Think it's more likely to be for stability of the circuitry. Uh, that is it. The circuitry is not complex.

I mean I Say it's not complex when you do a component count on it. But what's happening is complex. And having watched these lamps evolve from the early days, they used to be stupidly complex, but now there are a whole lot simpler. The one thing that's uh, notably different from this versus a really expensive ballast is that normally, before these smoothing capacitors, you'd normally just have, say, a filter capacitor and then you'd have what's called a perfect correction circuit that would, uh, try and draw the current over the full sine wave.

This one doesn't have that. So I wonder if I wonder what the power factor is for this deep down? I Wish I had a four or a 40 watt tube here to actually try that? It would have been interesting checking that, but kind of irrelevant because, well, who uses fluorescent tubes these days? But that is it. It's nice that it's done. the single-sided board.

Um, it's a nice, simple logical layer. They've included extra component positions, notably a resistor, bouncing resistors across each of these capacitors and also discharge resistor I suppose. but that's something that is to prevent a voltage imbalance if the capacitance fires are slightly different, or an asymmetrical waveform. But um, they've not done that here.

I Guess they thought, well, we just don't need them so we won't put them in. Um, but that is it. It's neat. It's a a generic fluorescent driver on how many of these are in use.

it looks so mass produced at them. Pretty sure that many cheap fittings will have them in them. and uh, it's just nicely simple and, uh, refined. It's exactly the same circuit, just in a larger size that you'd find in a, uh, compact fluorescent lamp.

Now what are these transistors? I Didn't know this down, but I completely forgot what they were. They are mje13003 13003 Um, Mje, it doesn't actually cmge, but that's a one Three double O three which is a very standard uh transistor uh used in these. It's commonly used in many of these ballasts, but that's it. It's quite interesting.

It's super flimsy and lightweight. It's cheap and nasty, but ultimately when it comes to Crunch, it will do the job. And if you consider the ones in the compact fluorescent lamps that just got baked and just kept going, it's very refined. It's A.

It's just a super simple, standardized, reliable design. Actually, very neat.