Visually this LED filament bulb looks nice, and to be fair it does light up... But with no smoothing at all?
I fully understand the reliability of minimalist circuitry, but I wonder what effect the potential inrush current will have on the LEDs over time.
When you have a simple capacitive dropper and power is applied, then the discharged capacitor will represent a near short circuit initially, And that results in a current spike limited only by the wiring resistance and the LEDs. It's a very short pulse of current, but if you turn the power on at the peak mains voltage it will be a high current spike. That's even worse with a traditional dimmer, as half intensity will be turning the light on at peak mains voltage twice per cycle. That's 100-120 times a second.
My guess is that this is the cheapest and smallest universal LED drive module they stock. Primarily intended to fit into very small lamp caps, but also slapped into all the others too.
Mid-play video-adverts are annoying in technical videos, so I don't enable them. If you appreciate that and enjoy my content then you can help support the channel with a contribution of a dollar or two a month on Patreon. That also lets you critique the (advert free) videos before they are released, gives a more direct means of communication with me and also gives access to the regular relaxed Patreon live streams.
https://www.patreon.com/bigclive
#ElectronicsCreators

It's time to do a recreational teardown of a lamp because this thing that I'm currently reverse engineering is proving to be quite complex and irritating to reverse engineer due to its custom chips which doesn't make things easy no sign of them. are you type numbers in the internet and it literally comes up with two random results. It's not very good, However, this is what we should take a look at. Instead This is a quick doodle I did in the front to show what's in it and they're going to run in advance.

This is flickery. That's partly why I want to take it to bits Now let me grab a lamp holder and I'll show you this. So here's a lamp holder and we'll test its power. I'm going to guess its power is going to be in the region of 2 Watts because it is that using the one watt filament.

So let's bring this the happy so plug the flickery hop in and then it will get competition of something else that will. Flicker and I'll just put this back in this bag if something is tried to sneak into shot there. So I'm going to run you in advance, this is going to flicker. I Know it's going to flicker because I've seen it and it flickers.

Oh, it's not showing up in the camera as being flickering. That's interesting. Uh, to the naked eye. Yes, it's strobes, but that's not bad.

It's not flickering on camera at one point. No more, so close enough. 16 milliamps point Four Power Factor: Typical set of things looking up the end of the land I Was hoping to see the usual soccer chair, a little round circuit board, but there's not. There's a bit of heat shrink sleevie in there, which partly explains the flickering because there's a tiny tiny circuit board in it.

Get this out the way and that probably means it's going to be a tiny little ceramic plaster and a bridge rectifier. And if we're very lucky, uh, in Rush Limiting resistor and no smoothing. So the only way to open this because it's the lamp cement. It's not likely to come off easily without breaking the glass.

This is so dirt. So I'm just going to nibble my way in. Kiss goodbye to this lamp. It's it's about to meet its maker in the name of science.

So let's see if I can do this without cutting into myself. So the values I Get into these is I will peel the metal away and this is where the Anderson screw winds over the bayonet cap holders because the beanet cat ones are so much harder to open for some reason. Well, they're smooth. For a start, there's nothing to get a grip on.

Here's the circuit board. There is a circuit board with what looks like a ceramic plaster on it. I Will draw out the schematic. it's not going to take long I wonder if it's got any in Rush Limited at all because that would be annoying if it doesn't.

All right, tell you what? Where's my sneaky little side? Cutters And I'll get in there and cut that wire. I've missed the wire I think I've missed the wire? Yes. I have it is cut. Behold.

Uh, there is the circuit board. Bend that metal out the way here. There's the other wire going down to where it solders on and that means that I can now theoretically just bend these wires in and get this cover off. Oh yeah, where's the redirect fire? The redirect fire is on the back, the Uh capacitors in the front.
There's a discharge resistor cross capacitor, right? Say what? Where's the notepad? This isn't going to take too long. On a plus note, it's so simple that if that uh, resistor value doesn't mess things up too much, I should be able to measure the value of that capacitor. I'm going to guess it's 470 nanofard. Let's put this to the two microfarad setting and then go across this capacitor 260.

But is that an accurate value? Because uh, it does have the discharge resistor across it? It's actually, it probably is an accurate value 260. that's an odd value. Let's say a typical wire might be 270 Nano fired for that. Hmm, Okay, on one hand it's Lauren I was expecting, but then again, it's quite a low current going through this, right? So what? Where is magnifying glass to see that resistor which has a value of 304? That's three zero and for zero something is just Avalanche.

That's okay. Um, so that's 300K and on the back is just a bridge rectifier right here. Oh, anything else there? Oh, there is a little damper resistor on the other side of it 204 which is a 200 key and that's just kind of pointless. really.

That's uh, the one that's just designed to damp it slightly so it doesn't glow I Suppose it is helpful to stop it. gluing. This is that horrible, uh, dark glass that when you try sweeping Off the Bench it will cut to me now. I'm just going to sweep it with the notepad to the side and then I shall vacuum that up safely later.

Let's draw the schematic. so I'll zoom out just to tell you a bit for this, but not too much. We've got live. We've got neutral.

We've got that capacitor which was about 270 nanofarad. Um, not sure the voltage rate will be always a bit concerned about those tiny tiny little ceramic pastures because they're so small, but this one is a fairly decent size and I do know they make them in Fairly high voltage forms sometimes to get the voltage they actually have. Uh, if these are the plates at one side, they'll actually Bridge them with a central plate and then more plates with those connected to either side and that puts effect. They took plasters on series but increases the voltage rating.

but anyway, we have that and we've got a discharge resistor. was that 400k I said see I've distracted myself 300K which is an odd value 300K and that's a discharge resistor so you don't get a tingle when you touch the back of the pins. the other connection will be going to the bridge rectifier. At least it's rectified.

Could have been so much worse. but uh, having said that, in this case, it would kind of only really work with that if they did this and then the output is going to those filaments. I'll just draw them as just a couple of big LEDs but in reality each one of those filaments will have quite a voltage drop across it. Potentially they could have thought your chips and that would be say 90 volts across each one.
Just a guess or it might be lower, but that would suit the UK Supply There is no real inrush limiting when you, particularly if you put this in a dimmer, which is probably not great, although they probably describe it as dimmable. Knowing them, it means that as the when the dimmer switches on, you've got a pulse of current that goes through the LEDs. But there is also that other resistor which I think was 200. Okay, I'm gonna have to remind myself of that one again.

Cross the bridge rectifier. Yes, 200k. and the purpose of this one is just to put a slight load on so that uh, if you've got leakage through the The switch wiring capacity leakage, it doesn't make the LEDs just glow a little bit and that's it. Um, it's just capacitive dropper rectifier.

Absolutely no smoothing. The smooth smoothing would have been nice here, and there's loads of room in here for it. but maybe they did it for reliability and just say, well, nobody really means some flickering. I Mean them flickering? It's annoying.

It continually stimulates the peripheral vision out the side of your your eyes. and, uh, it can be quite irritating. It keeps your brain alert so to speak, which isn't necessarily a great thing, but that is it. It's a very simple lamp, much easier to analyze than this thing, which I'll work on.

It's weird, and I suppose Ultimately, it's quite stylish. It's they've actually made it using traditional. Lamp Manufacturing means that they do do met or dummy metal the sort of like the metal that's designed to match this pinch seal. and also they've got nice solid uh stems in here that they've spot welded the filament on time.

They've spot melded a extension onto this so they can use this as a standard base. Quite interesting, so actually, quite a neat lamp. It's just a shame that they didn't go that one step further and they had a bit of smoothing just to get rid of the flicker, but quite nice nonetheless.

12 thoughts on “Very tacky ebay led filament bulb”
  1. Avataaar/Circle Created with python_avatars Hola! Cucumber Man says:

    Nice video Clive thanks, theyโ€™re getting less and less in them lamps nowadays!

  2. Avataaar/Circle Created with python_avatars Acamperfull says:

    I wonder if the phosphorus layer around blue leds could dampen the flickering from the LED, in a similar way as a tungsten filament does not flicker either? That could also make them compatible with some traditional dimmers, dimming with a capacitor in series already works very nice and makes them much more efficient and long-lived.
    I think that build-in load resistor to make them not glow when off (with capacitive leaking switches and or wiring) is a pity, otherwise it could be a dual purpose feature; a night orientation glow against hypnagogic hallucinations with virtually no (extra) electricity consumption when turned off and a normal light bulb when turned on!
    A nightlight faintly glowing effect could still be achieved with these bulbs by putting a tiny capacitor over the switch just enough to get a current that offsets the load resistor inside the bulb and make the LEDs glow with an electricity consumption that still is close to none.

  3. Avataaar/Circle Created with python_avatars Blitz Roehre says:

    …looks like it came straight off the production line for low-watt high pressure sodium lamps. I wonder whether they filled the globe with nitrogen to aid cooling the LEDs…

  4. Avataaar/Circle Created with python_avatars Fluxkompressor says:

    That lamp reminds me of an Sodium lamp, with that thick wire going to the other end. They probably just reused some old assembly line for this, just swapped the sodium burner with LED fillaments

  5. Avataaar/Circle Created with python_avatars Erik Swiger says:

    You mentioned flicker, and this raises a question. Typical movies are shown at 24 frames per second, yet as far as I know, nobody has ever complained about the flicker rate of movies. So, if our LEDs are flickering at 25 to 30 times per second, why is it that we're aware of the flickering? Seems to me that we shouldn't notice it.

  6. Avataaar/Circle Created with python_avatars Byron Watkins says:

    Ceramic dielectric constant typically has a very high temperature coefficient.

  7. Avataaar/Circle Created with python_avatars Michael Skelton says:

    Do you think you could do a project where you buy a some cheap filament bulbs and then make a DIY Dubai lamp from the parts inside them and combine the filaments to underrun them?

  8. Avataaar/Circle Created with python_avatars VarionJimmy says:

    Nice video, with a simple schematic, to watch before sleep. (Itโ€™s 7:09 in the morning here in Sweden..)

  9. Avataaar/Circle Created with python_avatars Der Karhu says:

    at my previous position, we did some research, many years ago, into sensitivity to flicker/pulses, in LEDs, and there is quite a wide variation in sensitivity, something also noted in monitor studies …

  10. Avataaar/Circle Created with python_avatars SeanBZA says:

    Bayonet easier to open as you have the pins you can pull out, giving that small edge to grab and peel the base off. Lamp itself is probably still using the old lamp making machinery, just they have disabled (or it just broke) the vacuum side of the massive carousel assembly line, with probably a half dozen workers sitting there with the formed filament bases spot welding the LEd bars on, and then placing them into the machine manually, where it runs and seals the base, before it melts off the pinch. Of course they use the old dumet feedthrough, as replacing it with something likely is going to be expensive, and that wire is not expensive at all, especially if you do not care too much about the seal being good all round, just has to hold and accept the spot weld.

  11. Avataaar/Circle Created with python_avatars Mark Markofkane says:

    Opening the metal tip reminds me of those corned beef cans that had a key to wind around the can to open it.
    I have a 4 watt LED bulb. It's shaped more like a bulb than long. It doesn't flicker that I can sense. Interesting video!

  12. Avataaar/Circle Created with python_avatars Mike says:

    Love to see what this would look like at 120 V.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.