Note from the author (== Plons): It has been a while since I used DHE-editor to create webpages. I did some family webpages with the free version of this editor. You can find this free version at
But since this webtool generates much nicer webpages than the archaic ones I used to publish, I purchased a licence to allow me to use it for personal and professional use.
This page shows my first steps with this new tool ..... uh .... toy  :-)

Large fontsize huh ? I use it as it's more comfortable for me AND for collegues from the same age-group.
24 button IR controller
On this page you will find all the information you need to build your own Led-light controller.

I reverse-engineered the protocol because I like this remote, but the implementation of the actual controller (the power-module) did not give the result I needed for my application.
Opening up this remote would destroy it, so I had to come up with a non-invasive way to measure the carrier frequency.
I took a 680uH inductor, connected it to the oscilloscope, pushed a button and found:
Yes, but it uses the current transients in the IR-diode, so put the inductor like shown above, and press a button.
An inductor to measure the IR-carrier ?
The oscillations you see in this picture are so-called parasitic oscillations: the inductor has some parasitic capacitance and so does the probe. The transients in the current through the IR-led are picked up and make the circuit oscillate.
But that is not what we are looking for.
When the current is turned on, there is a sudden change in magnetic field. But also when the current goes off: so the time of two oscillations reveals the IR-carrier frequency. The measured time is a tad more than 25 us: appr. 26us. That comes down to a 39 kHz carrier. A 40 kHz IR receiver (like the TSOP1740) will work fine, although it's slightly off.
Okay, we know now what carrier we're dealing with, so time to hook up the logic analyzer.

The w i d e  picture is a screen-cut from what the Intronix Logic Analyzer picked up from the TSOP1740

The application software is very powerfull, and it was a piece of cake to measure the intervals. Counting 32 bit in a row is a different story.
With notes on a sketchbook I used my PCB-design program SprintLayout to draw the complete pattern, and exported it to a pdf.
A .png file for this webpage and the .
pdf as a downloadable link
AHA ! 40 kHz ? Nope, 39 kHz is the correct number. The 40 was my first rough estimate. But I am too lazy to correct it in the files and exported docs.