Christmas Lights

Black Hole

May contain traces of nut
I have a string of LED Christmas lights (procured post-Christmas in the sell-off), 10m long with about 100 LEDs on it. There is a wall-wart transformer (just a transformer it seems) outputting AC (the connector is unpolarised, and doesn't seem to mind which way around it is plugged) which runs to a small control unit with one button. Press the button to change the light effect - constant, chase, flashing (and variations thereof). Okay so far.

The string of LEDs itself has four conductors in some places and three in others, linked between the individual LEDs at 4" intervals. The LEDs appear to be connected in series. Okay....

Here's the killer: there are only two conductors between the control box and the first LED in the chain!!!

This clearly requires further analysis.
 
I'm surprised that they would use addressable LEDs rather than a bit more wire, it sounds like a more expensive option
 
I'm surprised that they would use addressable LEDs rather than a bit more wire, it sounds like a more expensive option
With the cost of electronics vs labour these days I'm sort of surprised, but then not really.

But if they are using those why would they need 3 or 4 conductors in places? Wouldn't just 2 suffice all the way along?
 
Information to ponder:

[Xfmr] -2- [Ctrl] -2- [1] -3- [2] -4- [3] -4- [4] -4- ... -4- [49] -3- [50] -2- [51] -3- [52] -3- ... -3- [99] -2- [100]

Where [n] is the number of the LED in the string and -n- is the number of conductors between them.

In fact, all the odd-numbered LEDs operate together, and so do the even-numbered LEDs (there's no proper running light action, which would need at least three groups). And, given that this string has a broken wire resulting in half the LEDs in half of the string not working (25 LEDs), it appears there are two independent sections of 50 LEDs each.

So the operating modes can be explained if the odd set is on with wire A more positive than wire B, the even set is on with wire B more positive than wire A, all off with A and B equal, and all on by rapidly alternating A and B (so they are not actually all on at the same time, they just appear to be).

Hypothesis: there are 25 LEDs wired in series in each half on the 100-LED string, so the voltage to them has to be >50V.

Further examination should confirm. The control box looks difficult to get into without destroying it.
 
The transformer is rated 30V AC 2VA. That's enough to generate a DC rail of about 80V.

Here's what I'm thinking:

EB4E9EBF-00F0-471A-A265-DBE5BB56AACD.png
Note the control circuit could be alternatively implemented without the bridge rectifier, using synchronous AC switching (just bring the AC directly to the bridge driver supply).

Update: general wiring scheme confirmed.
 
Last edited:
I just had a walk around the same shop (where the string came from last year), and a string of 20 indoor LEDs (battery powered) can be had for £1.29! Compare that with buying the parts separately!!!
 
My wife has a string of LEDs similar to yours. I thought they used a LED multiplexing system known as charlieplexing (google knows). I have never gone as far as investigating the string, but your schematic could be more correct.
 
I don't think I've heard the term "Charlieplexing" before, but looking at the Wikipedia entry there's nothing new about it and we were using similar techniques well before the quoted 1995 coining of the term. LCD driving schemes also rely on difference voltages being insufficient to excite undesired segments.
 
Back
Top