In my previous blog, I described how I came to decide to construct my own 8x8x8 3D tri-colored LED cube.
As we discussed, I purchased 1,000 tri-color LEDs for just under £35 along with a 500g reel of 20SWG tinned copper wire. For this type of project, I personally prefer the silvery appearance of tinned copper wire to its plain copper counterpart. Some other constructers I found on the Internet used 22SWG wire, but I decided that 20SWG wire would be sturdier ad would result in a more robust structure.
The problem with all this is that when you unroll the wire from the spool, it curls back on itself and it's really difficult (if not impossible) to straighten it by hand. Consider the two pieces of wire shown below; which would you rather work with if you were creating a 3D LED cube of your own?
Now consider the panel shown below. As you can see, there are 8 horizontal wires and 24 vertical wires, and this is just one of eight such panels.
Obviously we're going to need a lot of pieces of straightened wire, but how is this to be achieved? Well, we'll look at that in a moment, but before we go there, let me summarize first how the testing of the LEDs turned out.
In the end, it took around two weeks to perform all of the initial testing. Admittedly, I typically tested only a few LEDs each day (this usually occurred while watching TV in the evenings after I'd returned home from work). Out of the 1,000 LEDs I started with, two failed the electrical test with the blue LED element being non-functional. Also, around 80 to 100 LEDs were rejected for defects in their diffusers, like air bubbles or scuff marks on the surface.
In hindsight, I was possibly being a little over-fussy here as -- with the exception of the two failed blue LEDs -- all of the remaining components were perfectly serviceable. On the other hand, I have LEDs to spare (I started with 1,000 and I only need 8x8x8 = 512) and I want my cube to be as perfect as possible. Furthermore, the rejected components can be used to verify my construction methods.
While testing the LEDs, I canít honestly say that I was able to detect any differences in the quality or brightness of the red, green, blue, and white comparisons -- not with my eyesight, anyway. If I were to build another cube, I might choose to not test to the same extent and to confine my testing to visual checks only. Each LED will be tested before it is soldered anyway, and tested again at various points during the construction, so any failures would be detected in plenty of time before the final assembly.
Now let's return to the task of straightening the wire. In order to achieve this, we're going to need some tools -- namely an electric drill, a pin chuck that grips down to 0.5mm diameter, a small vice, and a pair of side-cutters.
We will consider the actual process in a moment, but first consider the image below. This shows a simple jig comprising three lengths of wood with end-stops to the left. The strip of wood in the foreground is 8.75" long; the one in the middle is 10.5" long; and the one to the rear is 8" long.
The way this works is that I can push a piece of straightened wire up to the appropriate end-stop and then use the side-cutters pressed to the end of the corresponding length of wood to snip the wire. As we can see in the image below, the end results are reasonably consistent.
I donít know about you, but I always think that it's a lot easier to understand how something is done if you can actually see someone doing it, so I've created this video to show the process in action.
OK, so now we have the LEDs tested and the wire straightened. In my next column we will dive into the main construction; in particular, I will present the jigs I created to speed the construction and improve the accuracy of the assembly process. In the meantime, as always, I welcome your comments, questions, and suggestions.