Understanding commercial LED backlighting circuits (I don't!)

[suicidal_orange]

I'm sick of not understanding commercial LED circuits.  Each time a question comes up I look, get confused, then run away.  This is no longer acceptable!

Here is a nice pic of a Pok3r PCB I'm borrowing from someone going by the name of yabbadabado on imgur, the circuit looks the same as other boards.



There is a trace connecting the "+" legs for the LEDs going down the column of keys 3, E, D, C - makes sense.

But then the trace continues to the resistor for each LED, and from the other end of the resistor to the other leg of the LED - by my understanding this puts the resistor and LED in parallel?

Every circuit I can find from the newbie LED on a breadboard tutorial to a DIY RGB matrix clearly shows a resistor in series (one per row or column in the case of the matrix)


Please can anyone explain what commercial keyboard designers are doing and why, or correct my understanding of what I'm seeing if that's the problem?  I need to understand

[suicidal_orange]

*bump*

200 views and no replies - even a "yeah, that's parallel and they're being dumb" would be something

[rowdy]

I can only sympathise and say that this exceeds my humble understanding too.

[vextanys]

I could be wrong, but the columns are connected to a single resistor (R40 at the top), and each LED is connected in parallel to an inductor (marked LR12, LR20, etc). I would hazard a guess that the inductors are there to "smooth" the PWM brightness and/or LED matrix between the output cycles so that each LED can be individually controlled.

[xondat]

yeah, that's parallel and they're being dumb

who am I kidding I have no idea how this **** works

[Oobly]

IF the traces truly go to either side of the resistor, then that's a PCB design flaw and it simply won't work. This could be the reason they say you can't add LED's to a Pok3r: https://geekhack.org/index.php?topic=74170.0

All the LED's should actually be wired parallel, but in series with a current limiting resistor.

From what I can tell, the one pad of each resistor should go to ground instead of the positive terminal of the LED.

[obones]

Is this a single sided PCB? If not, it would be interesting to see what's on the other side, connected to the LED legs.
If that's single sided, then the resistor value would give an indication as to what is happening. Can you tell us what the markings on them are?

[suicidal_orange]

Thanks Rowdy and xondat - glad to know I'm not alone in this )

@vextanys interesting suggestion - I don't know what an inductor does and it seems crazy to label it LRx rather than Ix but I'll look into it.

@Oobly - I agree with your "shoulds" but you can add LEDs to this very PCB.  Most work at least...  That was the thread that sparked this one.

@obones - the pic is huge, click it then click the arrow-square thing in the bottom right corner to make it even bigger and you can read all the labels   It's definitely double sided...

[tufty]

I could be wrong, but the columns are connected to a single resistor (R40 at the top), and each LED is connected in parallel to an inductor (marked LR12, LR20, etc). I would hazard a guess that the inductors are there to "smooth" the PWM brightness and/or LED matrix between the output cycles so that each LED can be individually controlled.

Generally you'd use inductors in parallel with LEDs as a "flyback" circuit, enabling you to, for example, drive it with very low voltages.  Connect a LED / inductor pair to (for example) a 1.5V battery *the "wrong" way round*, then disconnect the battery.  Hey presto, the LED flashes.

CBA to work out if that's what's going on here, though.

Edit - here you go - http://web.engr.oregonstate.edu/~traylor/ece112/beamer_lectures/led_flasher.pdf