May or may not have been the first one to think of this, but it’s bare WS2812B and tin coated copper wire (carefully) soldered together to make a very dense rail of LEDs, 10 LEDs in 2.75 inches which is about 144 LEDs per meter. Going to eventually upgrade all the LEDs I put in my PC with these rails.
Thank you! No, for now i just used duct tape to line the LEDs up face down, then held the wire with pliers and soldered everything together. After the LEDs were attached I used a block of aluminum I have to flatten them out. It actually worked really well!
Honestly, I had thoughts of doing exactly that with the firm intention of putting together a 16X16X16 cube !!!
I was considering using slightly smaller solid copper wiring to minimize the visibility of the cube’s skeleton and of course, putting some distance between the LEDs.
However, experience with the WS2812b on miniboards (see my post from Nov 25th) have not been very successful. I figured that I was simply overheating the little buggers with my soldering. Considering that my miniboards had some built-in heatsinks I did not dare to go and order over 4,000 of them and that idea stay dormant for now…
I would suggest that you run a long test with the LEDs on full white for some time. You may find the same weird behaviour that I encountered.
Please keep me posted about the results… thanks !!
What temp are you soldering at @JP_Roy ? Sometimes hotter is (counterintuitively) better, as you spend less time heating up the component - in and out in a flash. A jig is also a worthwhile investment in time.
I use a very old but nice 60 watts Weller model W60P. Specs says it has a narrow temperature controlled tip of 700°F.
I have used it for over 30 years and have only had to replace the tip on occasions. I even successfully did, with some difficulty, some surface mount soldering with it.
I am quite good at electronics soldering as I have a significant amount of experience over the years though it does not mean that I did not screw up the soldering in that case…
I did use a very simple jig ( a small low pressure clamp actually) that held the devices in place for soldering.
From what I read, nobody experienced similar problems with WS2812b devices set at full brightness white for some time so my
conclusion is that I managed to overheat the insides of some of these ws2812b devices to the point that they did not like to be driven 100% white anymore.
I still have some doubts leftover as to the manufacturing quality of these devices so I decided to not undertake any more projects where I would have to solder them.
.
Hi @Jon_Burroughs , I would certainly recommend adding that capacitor, although you could probably ignore that requirement (without noticeable consequences) on very short lengths.
You could also get away with using only one capacitor for every 2, 3 or even more ws2812 devices !
@Jon_Burroughs There’s an example schematic on the datasheet that shows one being used per LED, I’m not exactly sure what it would do though. This is more of a test than anything else. I’ll run it on a high brightness sketch overnight and post my results tomorrow morning. Anyone know what the function of the capacitor is?
Hi @Kyle_Halvorson , the purpose of that cap is for power decoupling.
Because of relatively high current, fast switching within the device, a local capacitor will supply that immediate current and will prevent excessive switching noise on the +5Vdc wires.
This past week I tried to do something similar and seemed to ruin the leds, I was soldering at 400 degrees and thought excessive heat was the problem. Maybe I should have gone much higher?
That’s possible, I melted a few small parts of some of the cases of my LEDs, but they’re all still functional. I use a xytronic 168-3cd soldering station that has a +/- 2 degree temperature controller. I had it set to 700 and had no problems with having to keep the tip on the LEDs for extended periods of time.