Dimmable LEDs

If the power supply is intermittent and you want DC from it there's going to be a drop in voltage = ripple.
In very round numbers the bulbs I'm talking about are 10V (ok 12V) and 10W means 1A.
10% ripple might be OK, or not.
--


The data sheet is very poor about specifying the output. What it quotes could well be some sort of average.
In my exp if there's no tolerance given, it turns out to be not good enough
Most applications wouldn't be fussy about variations during the cycle so they wouldn't need to make it good, so there would be no need to spend money on that.

In the circuit shown there's too many components to fit into a bulb I think. Big cap or two in there. 75V = a separate supply thing.


I'l play around with a few dimmable bulbs & when I break one I'll cut it open

Ripple on the input cap voltage won't affect the LED brightness as the switch mode control loop will be fast enough to adapt to it.

This is from a non-dimmable mains-input LED lamp; the module is approx 24x16x15mm.
The only large components are the transformer and the input cap, which is 4.7 uF 400 V.
Everything else is tiny. In fact they've not even used the smallest surface mount resistors; they could have fitted twice as many components in the same space.
The additions to make it dimmable would increase the number of components, but the volume would not have to increase much.

Here is the datasheet for the chip in that module:

http://www.bpsemi.com/pdf/BP3102/BP3102_DS_Rev 1.0 EN.pdf

And here is a similar one from the same manufacturer that supports triac dimming. Note the additional components.

http://www.bpsemi.com/pdf/BP3108/BP3108_DS_V0.2 EN.pdf

endecotp said:

In fact they've not even used the smallest surface mount resistors; they could have fitted twice as many components in the same space.

Click to expand...

If it becomes too small and fiddly they can't use children sat at kitchen tables for 16 hours a day to put them together.