This stinky little thing is a dusk sent so let me Zoom down it and this was sent by Dave and he got it in a bulk headlight fitting from a UK supplier called tool station that supplies trade materials and it was a very basic dusk Center and he said that as the light transitioned to dusk, this started flickering horribly and he went. Looked at the reviews and they said the only solution is to get the three wire dusk sensor and replace it. but he sent this one so we could explore it. and the first thing I'm noticing here is it's held together by sticky tape.

It's literally got a label that has been folded around it to hold the case together. That's quite odd, but having said that, this is inside an enclosure, so tell you what we'll do. I Shall Zoom back out again I Shall bring in a light bulb a low power light bulb I've got the little Chinese test unit here I've got a socket here and we can use this suitably trashy. Chinese uh, wgo clone or Vago clone if we want to pronounce it correctly.

So so I shall hook this in line in series with that socket. I Shall put this into this connector noting that this is off at the moment, the display stays lit all the time. I should really add an indicator. like to tell me when this is on just as a reminder.

It would make sense to avoid nasty tingles, but I'm going to that stuck in I Definitely put that into the connector. Well, didn't I What did I do there did I not push it in Far Enough you get over complacent with these things I Guess it's the fact it's soldered and it's just not not gripping onto that well. H The other one seems en tily makes mental note: don't put soldered wires cuz this came soldered into the Vago type connector. Maybe the real Vago would actually work better.

So let's screw this in. I'm leaving it in the Box because I think can leave it in the Box because I want to Shield the sensor from the light itself. There we go. it is.

it is now attached so I'll point that in that direction. Here's the light sensor. I'm going to have to turn the light off for this and I will warn you in advance that this may get flickery. So I'm just going to set the lighting up right now one moment, please.

Okay, it is now detecting that it is dusk and I've got a LED flashlight here I'm going to Shield it and I'm just gently going to pan it around to the point that it just barely. And yeah, you can see it actually visibly shimmering and flickering at the sort of low light level that it's just neither fully on or off. That is Very. That's very indecisive.

It's not a snap action, it's literally trying to dim it up that is not working very well. Is it okay? That's excellent. That's exactly what we want. Okay, H Tell I'll pop this open I'll take a picture and reverse engineer and we can explore the circuitry one moment.

Please, Reverse engineering is complete. Let's explore. It is not sophisticated. The black sticker that I thought was holding the case together I think it.

although it will have had some effect in holding it together I think it was mainly to Shield it from light because they have used a clear plastic case. So the PIP for the light sensor can be the same plastic but as a result of that it can get light in from the back and that will affect its stability at night. Uh, on the circuit board I'll Zoom down this. on the circuit board we have a discrete Bridge W far made from four diodes across.

That is a little fist R here. I'm going to guess it's an MCR 100 which is a very sensitive Thyis and this blob of silicon. Here they've made a little aluminium heat sink that folds around that I think it's aluminium and they've sort of glued on, but they've also glued glued onto the capacitor. This is a bit that's going to get hot with traditional tungsten lamps and there it is glued onto the electrolytic.

That's not going to extend its life much is it? Uh, there is a current limiting resistor here. 47k there's a voltage clamping xener out here. I Measured just over 8 volts so 8.2 volts is standard value. 100K in series with the Uh Ldr light depend resistor and that is it.

I shall let you Feast Your Eyes Upon the schematic here. Well, the C board and I shall bring in the schematic and we can explore its Simplicity and why it's so unstable. Here's the incoming Supply it goes via the lamp or bulb or Globe whatever you want to say. I've drawn the old retro tungsten one since that's prob with that circuit was really intended for goes to brid R fire and the way the lamp is switched on is with this thyris here shunting out this bridge Rec far.

so when this gets triggered in each halfwave, it turns on. current flows through the diodes and in here through the ther, keeping it latched on until it get reaches the next Zer crossing point. It turns off and then gets triggered again. I say the zero Crossing Point Well, in a sense it is a zero crossing point, but keep in mind this is a humpe DC it's fuel wave rectified um AC into just basically UNS smooth DC There is the current limiting resistor up there 470k.

quite high value and there is the xener diode that gives a fairly stable-ish voltage here. and there is a potential divider 100K resistor and the Ldr the light depend resistor probably cadmium sulfide and it then disc goes over to the gate of that thyer with a little 47 microfi capacitor which is probably just to add at least some element of stability. Uh, and that Uh turns the Thyris on when the light level is at the correct level. And what that means is that during daylight, the current will be flowing through here and the voltage in here will be low because the resistance of this is is low.

But when it gets dark and the resistance of that is high enough, the current then chooses to. well, the voltage then reaches the Threshold at which can trigger the thyer. That is it. It's very simple.

There's no hysteresis, which is the main problem here. The circuitry reminds me a lot of the very basic Uh children's nightlights that just had even less. They literally had one diode uh, and then the Thyis turned. It just drove the lamp halfwave.

But there it is. That is why it was flickery. Not ideal for electronic loads, and even while it's turning on and getting unstable at night, that's going to result in a bit of electrical noise from switching. Tungsten lamps, Not tungsten lamps are use much these days, but they will go through a stage that it is probably triggering.

Sort of kind of halfwave is actually I. Suppose the Zen is there to actually stop that. it's going to cap the voltage from uh, going up to the sort of M sort of midpoint of the sine wave. Um, but uh, certainly it's going to be creating a little bit of noise when it turns on and it's going to be doing that flickery thing.

Fortunately, this xener probably actually does ensure that uh, it's more likely to stay. The switch on point is going to stay fairly close to the Zer crossing point, which is pretty good H But there we go. A very, very basic circuit I didn't draw my usual diode in here I've drawn it. Now that's for the brid direct far h a very standard approach You you'd wonder why they didn't just use a track.

uh, rers are more sensitive. this one probably has a gate current of like 40 microamps or something like that and it's just a very simple way of controlling uh loads with a bridge W far and shunting it with the thyer. So an interesting circuit not ideal for modern LED lamps, but will actually kind of work if you don't mind that slight. Shimmer at the dawn.

well, at the start of dusk.