That's interesting, I've never heard of this issue. Do you know the problem it causes and how big if an issue it is?
Up and down lights, is there a problem??
- Thread starter mg123
- Start date
Help Support UKworkshop.co.uk:
Some led bulbs may not as reliable as others and the on board electronics possibly more prone to either total failure; or at least reduced driver output. I have an outdoor led lamp in a bulk head installation with integral PIR. The bulb failed in that housing, whilst in other non PIR areas (indoors in a garage) the dame type and wattage have lasted, are lasting, far longer. The PIR unit has frequent on/offs due to assorted critters critters triggering it (think moggies, foxes erc.). Current bulb of another brand thus far is holding up ok. Flashing led systems have different electronics and so on and are designed for that flashing operation; other bulbs not so. I’m still inclined suspect the bulb and its built in driver in this case, producing correct volts hence the problem as described. The problem with led system with non replaceable bulbs is one never knows if it’s the drivers, or the led effectively failing as any diode system can have leakage across the diode(s) or go high resistance (the latter being unlikely though). I have seen led bulbs start ti show reduced output before they actually fail (before the claimed life of sed bulb).
LED's have zero issues with being 'flashed'- they are a solid state component!!!
The failure of 'cheap' LED bulbs is usually manufacturers overdriving the LED's inside badly, in order to use less LEDs to get the same output (this is the cause of most early LED bulb failures) and there are some that are guaranteed to fail in only a few hundred hours or less of use- with them being overdriven by 200% or more of their maximum allowable current!!!
(there is a youtuber named Big Clive from the UK that his entire channel is 'teardowns' of electronic devices- and LED bulbs feature prominently on it)
It isn't the 'only' failure mode (excess heat can cause failure of the drivers true- usually capacitive dropper style circuits although these circuits are rare these days, and really only were found in the earlier LED bulbs) but overdriven LED's are by far the biggest issue- the cheaper the bulb the more likely it is that they are being overdriven to cut down on manufacturing costs...
Apparently they get disoriented and it causes issues with their flight paths. They are the genetically close to humans and can communally hold back giving birth due to weather and other circumstances and have communal nurseries.
They are protected and are becoming more scarce.
Wonderful little creatures.
Have you had the circuit tested properly? Could be a loose neutral causing issues or maybe a volt drop issue if the circuit is too long (unlikely in a normal size properly thou) ! I remember a job from years back where customer was having issues with her living room lightbulbs flickering and or not switching on sometimes; that turned out to be a loose neutral. Did the symptoms start after a noticeable change in ambient temperature ?
That wouldn't cause just the top bulbs in a twin light fitting to fail...
If the two lamps are being driven by the same voltage o/p from the driver /power module, i.e. they don't each have their own driver module... and they're connected to exactly the same voltage point, then I would be inclined suspect the dimmer lamp is simply suffering from old age or excessive abuse... I would however expect the two lamps to be isolated from on another? If they are indeed isolated from each other then it may be that the dimmer lamp has a failing volts supply.
The driver module - 230 volts in and whatever volts out - will have an SCR or its current replacement chip. SCR have a history of slowing failing - as in leaking and thus a drop in the DC volts out.(in days of yore... When an SCR was failing there was often smell like rotten fish (actually it was silicone dioxide...) and It was warning to locate and replace sed SCR; and also to ensure the area was well ventilated as the gas being emitted by the failing SCR was toxic... If as is often the case today these units use switched mode psu or similar (all on a single chip) ... the same can apply, and again one of the o/p might equally be failing whilst the the other is ok (presuming again the two lamps are isolated from another). Is it not possible to replace the dim bulb with a new oneto see if it makes any difference? It it can be replaced and condition remains the nit points to the driver module. If the replacement lamp is normal o/p then probem solved?
incidentally, slightly off topic... if you're wearing hearing aids and switch them to 'phone mode... approach any led light source and you'll hear just how much junk leds radiate - as you'll hear them "sing"... Buses, trains etc. are saturated with their noise... as are most stations, shopping centres. Whereas in homes (mine at least) there is almost none. Presumably this is because the typical domestic led bulb replacements for incandescents are designed somewhat differently to those used in commercial and public areas, and transport and so on...?
The driver module - 230 volts in and whatever volts out - will have an SCR or its current replacement chip. SCR have a history of slowing failing - as in leaking and thus a drop in the DC volts out.(in days of yore... When an SCR was failing there was often smell like rotten fish (actually it was silicone dioxide...) and It was warning to locate and replace sed SCR; and also to ensure the area was well ventilated as the gas being emitted by the failing SCR was toxic... If as is often the case today these units use switched mode psu or similar (all on a single chip) ... the same can apply, and again one of the o/p might equally be failing whilst the the other is ok (presuming again the two lamps are isolated from another). Is it not possible to replace the dim bulb with a new oneto see if it makes any difference? It it can be replaced and condition remains the nit points to the driver module. If the replacement lamp is normal o/p then probem solved?
incidentally, slightly off topic... if you're wearing hearing aids and switch them to 'phone mode... approach any led light source and you'll hear just how much junk leds radiate - as you'll hear them "sing"... Buses, trains etc. are saturated with their noise... as are most stations, shopping centres. Whereas in homes (mine at least) there is almost none. Presumably this is because the typical domestic led bulb replacements for incandescents are designed somewhat differently to those used in commercial and public areas, and transport and so on...?
In the case of LED 'bulbs', there is usually a constant current source driver, driving a high voltage string of series LEDs
This is from Big Clive, two typical bulbs, one from Poundland, the other from the local supermarket (despite being from different sources, and having a different number of LEDs, they actually use the same PCB inside!!!) this is the circuitry found under that opaque 'domed cover'
Interestingly, the 8.5W one is the far better choice for long life in this case, as each LED pack (these LEDs have multiple chips inside each LED body in series) is being driven with far less wattage dissipated in each LED... 0.5watts per LED package in the 8.5W bulb, as opposed to 1.26W per package in the 6.3W bulb!!!!
As you can see, the circuitry is rather minimal- a bridge rectifier, the current regulator chip (SM2082) 4 resistors and a capacitor besides the LEDs themselves...
The LEDs have multiple actual LED substrates in each LED housing- the number of which can vary depending on the LED chosen by the manufacturer of the bulb- the rectifier and smoothing cap give you a approximately 340vDC supply in the UK which is applied directly to the series LED string with its series current control IC (the 2082)
This design has two features which are at the same time an advantage and a disadvantage- they work from 100-240v, so the one design works worldwide, but they aren't dimmable...
This is the circuit commonly found in many dimmable LEDS- and its even simpler!!!
Same bridge rectifier changing the mains AC to approximately 340vDC, same smoothing capacitor 1uF but this design simply gets rid of the constant current driver IC, and uses the dropping resistors only- as a result as the incoming mains voltage changes, the DC voltage changes as well, and the LEDs change their brightness
(this particular one has really cheaped out, as they have also save the extra resistor across the 1uF cap- which means this is one of the bulbs that has the possibility of 'ghosting'- that dim glow seen on some cheap LEDs even when the switch is turned off- however it is also the best choice of the three for 'long life' as each LED package is only dissipating 0.33W, against 0.5W and 1.26w on the first two bulbs up above... plus has basically no active components apart from the bridge rectifier (which really shouldn't be counted as an active component imho lol) the rest being passive components like resistors and the capacitor)
Which shows how truly efficient LEDs really are- the stray voltage picked up from the parallel conductors to the switch (one live ie supply- to the the switch, the other coming back from the switch to be 'on' or 'off' depending on the switch position- and unless they include that 1.5megaohm resistor across the cap, just random stray voltages inductively/capacitively picked up will light them up dimly!!!
There are some other designs used, but this is the most common one used- simply because its cheap and has minimal componentry to go wrong- the PCB can be made small enough to fit any bulb or housing (it is also found inside the small 'downlight' style bulbs and even tiny little SES bulbs sometimes found in some table lamps and the like)
The full video I borrowed those screengrabs from is from Big Clive and is 24 minutes long and can be found at with hundreds if not thousands more of his 'teardowns' on his channel...
(it's his 'thing' lol, tearing apart stuff to see how it works, or more often why it no longer works lol...)
As I said before, the most common failure in modern LED bulbs is the LEDs themselves- this is not because of an inherent flaw in them, but because some manufacturers 'cheap out' and save the price of a LED package or two from each bulb, driving the others harder to get the same effective 'bulb brightness' from a smaller number of LEDs- some cheap bulb manufacturers drive them by 200% or more of their maximum actual led chip ratings and these are guaranteed to fail early on...
(it is a testimony to just how robust LEDs really are that they can survive such massive overloads and still function for as long as they do...)
