My question is long, rambling and weirdly specific. I hope I'm not violating like 50 stackexchange rules, but I'm completely lost. Long-time lurker, first time asker and pretty new to circuits in general.

Here's the short version: How are two batteries that should output 3V pumping out upwards of 15V, and why do just a few LEDs drawing 3 ma completely drain 5,000 mah worth of battery in just a few minutes of intermittent use?

Long version: I've been working on a project to replace the dead microcontroller on my Razer Blackwidow Chroma TE with the microcontroller of a Logitech K800 wireless backlit keyboard, thus bringing my mechanical keyboard back from the dead AND making it wireless. The mod is based on this guide: http://probably.ninja/keyboard/

Everything went fine until I started wiring the LEDs. The K800 circuit board has two pins for the + and - of the caps lock LED, plus a pair of wires that originally powered a string of small white LEDs to serve as the backlight. My mechanical keyboard has a separate RGB LED for almost every key (85 total LEDs). After researching how LEDs work like crazy, my plan was to wire them all up in parallel, with a resistor for each LED. I tested my idea on a breadboard, learning how to get various colors and brightness levels using different resistors on the various LED anodes. I could make different keys red, green, blue, purple, teal, orange and yellow. This was gonna be great.

(I can't get the caps lock LED pins to light anything up no matter what I do, but that's another problem).

I moved on to actually soldering the wires and resistors to the LEDs. I kept it simple by starting with just one LED, lit red with the cathode connected to the keyboard's LED + line and the red anode connected to the keyboard's LED - line through a 100 ohm resistor. It lit, everything seemed perfect. I added a few more LEDs in parallel. Then everything died.

The batteries were dead. Weird coincidence, as I had been running on the same pair of Eneloops for two weeks. I replaced the batteries with a freshly charged pair of Eneloop Blacks (2,400 mah). A few minutes of wiring and testing later, the keyboard died again. This pair was also dead.

These batteries are relatively new. I charge them with a Lacrosse smart charger and I believe the batteries are in fine condition and full. The same batteries were charged and put in other gadgets, where they ran just fine. Something was up with the keyboard. I had no idea how, but the keyboard must be discharging my batteries really fast.

I put in some full Eneloops straight from the charger. I plugged the LED wires into the power rail of a breadboard and started testing everything I could with a multimeter. Here's what I found.

The total voltage changed as I added more to the circuit. With a simple 100 ohm resistor circuit, I got a current of 9 ma and a voltage of 0.3V across the resistor. Next I added a red LED. Across the LED, I got 1V and 3 ma. Across the resistor I got the same as before - 3 ma and 0.3V. The voltage between the two LED wires was 1.3V. I added a blue LED in place of the red, and got 1.9V across the LED, 0.3V across the resistor and 2.2V total.

(It took me a second to realize that I was actually shorting the power leads by measuring the current across the resistor with no LEDs, which must be why that measurement was different from all the others)

I tried putting more LEDs in series. One red and one green was 2.4V total, 3 ma. Adding a yellow increased it by another 1.1 to 3.5, still close to 3 ma. I got up to around 15V before the LEDs totally dimmed, and the current stayed around 2.6-2.9 ma the whole time.

Also, does it make sense to wire every LED in parallel, or should I wire as many LEDs in series as the power supply allows, and then wire each of those series in parallel with one resistor per series?

I haven't tried non-rechargeable batteries yet. I don't want to risk killing every double A in the house.

Another thing I didn't mention - the LED brightness is controlled by Fn+F5 and Fn+F6. I have the LEDs set to maximum brightness. I haven't tried it, but lowering the brightness would probably reduce the current to something below 3ma.

edit: I tested with a standard non-rechargeable, fresh pair of AA batteries. The power light went red after less than 5 minutes of on time and less than 30 seconds of total lit led time. All I can find when I search for this topic is people TRYING to quickly drain their batteries. I think they just need to come borrow my keyboard.

Here's a pic of the whole setup on the breadboard that drains the battery in minutes: Here's a pic of the whole setup on the breadboard that drains the battery in minutes

  • 1
    \$\begingroup\$ Are you sure you are acting on the batteries directly, not on some boost converter which used to power the leds? \$\endgroup\$
    – Posipiet
    Commented Apr 21, 2016 at 6:03
  • 5
    \$\begingroup\$ Post a circuit diagram of what you have wired up. a) We cannot see what you have done wrong without it and b) Just the process of getting the diagram together and drawing it will probably have you go 'doh! facepalm!' and solve your own problem - see this \$\endgroup\$
    – Neil_UK
    Commented Apr 21, 2016 at 8:38
  • \$\begingroup\$ If your project looks as messy as the guide you used, it's most probable that you have an intermittent short somewhere which kills your batteries. \$\endgroup\$ Commented Apr 21, 2016 at 8:45
  • \$\begingroup\$ I mentioned in the question that I'm measuring from the LED pins on the microcontroller. I'm quite confident that I'm not measuring directly from the battery, and that's part of my question. I presume the keyboard some circuitry to boost the voltage but as a noob I've never heard of such a thing. \$\endgroup\$ Commented Apr 21, 2016 at 16:17
  • \$\begingroup\$ I could post a diagram but I'm not sure it would be helpful. I disconnected everything from the keyboard and used only the breadboard, and the circuit I made was STUPID simple. + -> LED -> 100 OHM resistor -> -. I also added some more LEDs in series with the other but the circuit remains just as simple. The mass of wires connecting the keys all works fine, with no weird behavior or battery draining. I had no problems until I started connecting the LEDs, and I continued to have the same problem even when I moved everything out to a much simpler circuit on the breadboard. \$\endgroup\$ Commented Apr 21, 2016 at 16:22

2 Answers 2



Op is using a keyboard controller that apparently has a built-in step-up switching regulator IC or circuit for a constant current LED setup, with no idea of the specs, target current, voltage, or wiring for the LEDs. Considering the Logitech K800 is stated to have a 10 day life on fully charged NiMH with the backlight on, there's PLAINLY a short or something on the board or wiring. Check the original board for wiring details, and double check your modifications.

Old Answer: There is no way for 2x AA NiMH Rechargeables to be 15V. The nominal voltage will be 2.4V. From your description, you're obviously guessing the led voltage based on their typical recommended max, which is not how it works at a lower voltage. For example, a typical blue led will only be 3.3V Forward Voltage at 20mA. At 1mA, it may be down to 2V.

As for measuring it, the voltage across a circuit will always be the voltage across it's voltage source. And the current through a circuit is equal throughout the circuit.


simulate this circuit – Schematic created using CircuitLab

Notice, these 3 leds, with a total Forward Voltage of 6.3V at 20mA, will only have 0.85V or less across them, and 9 microamps. That's 0.000009 amps or 0.009 milliamps. The total voltages add up to the 2.4V across the batteries.

  • \$\begingroup\$ Great, there is no way. So why am I measuring 15V? As I said in the question, the circuit is powered from 2 LED power pins on the keyboard, but there's no reason to believe those are connected direct to power. There is probably some circuitry in there manipulating the power before it reaches the LEDs. Does there exist some kind of circuit that could produce 15V when only powered by 2 AA NIMH batteries? Because that is 100% what is happening here. Instead of dividing the source voltage among the LEDs, which even I know is supposed to happen, the source voltage seems to increase to match load. \$\endgroup\$ Commented Apr 21, 2016 at 16:27
  • \$\begingroup\$ I'm not guessing anything - every number was taken directly from many, many multimeter readings. The red LED had a FV of 1, the yellow and green were 1.1 and the blue was about 1.6. The problem was that the total voltage between the power pins increased by the LED forward voltage each time I added one in series. I tested the same circuit on a standard 5V power supply and it worked as expected, and was only able to light a couple of LEDs in series before they got really dim. The keyboard was able to light more than 7 LEDs in series on 2.4V, which is a mystery to me. \$\endgroup\$ Commented Apr 21, 2016 at 16:32
  • \$\begingroup\$ @JupiterJesus you said you connected the batteries to a breadboard, and then a resistor. The added an led. then another. and another. \$\endgroup\$
    – Passerby
    Commented Apr 21, 2016 at 16:41
  • 1
    \$\begingroup\$ @JupiterJesus in that case you have an undocumented boost/switching constant current led driver, and the project you copied doesn't give any details on the led wiring either. Look at the original board wiring for a clue. \$\endgroup\$
    – Passerby
    Commented Apr 21, 2016 at 16:57
  • 1
    \$\begingroup\$ @TimSpriggs The logitech k800 supposedly has a 10 day life with the backlight on. But it also has a built in NiMH recharging circuit when plugged into usb, and it has a proximity sensor (large capacitive sense board) to automatically turn the lights on and off adding to that time. \$\endgroup\$
    – Passerby
    Commented Apr 21, 2016 at 18:38

I can only assume there's a fault in my board. I don't know how or why or when it happened, but I cracked open another keyboard (I had two dead K800s in my junk box) and resoldered the wires to the new board and it seems fine. Haven't tried the LEDs yet.

I'd prefer to wire the LEDs to the caps lock pins since they give a constant, predictable voltage of 3V, but then my lights would only work WITH CAPS ON. Unless I can find a workaround for that (the modder mentions using the caps lock led power but does not give any workaround), I'll have to hope the LED backlight power works and doesn't fry the board again.

Note to self: the original backlight is a simple LED strip with 8 small white LEDs and a resistor, sandwiched in some kind of diffusing plate. I wonder how useful it would be to just jam that thing as-is in the case and ignore my per-switch LEDs. That would be no fun though.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.