In general, if you follow the manual quite closely step by step, there is a rational order described in there, based on experience.
In particular, one of the little tricks is to make the first tube holder on the right (1x seconds) and then use that as a jig for the other holders. Place the male parts into the female headers on the board, and then place the board over the unsoldered assembly and solder it. This means that the headers go in straight and are aligned.
Another few little tips: The NeoPixels are very heat sensitive: use lead/tin solder (it has a lower melting point) and keep the iron down to 280°C. Make sure you test all of the holders before mounting the tubes. It's possible to change the NeoPixel after soldering the tube (that's why they are mounted from the back), but it's a sweat. There is a spare or two in the bag, just in case.
The NeoPixels are in a daisy chain and light from right (1 x seconds) to left (10 x hours).
But other than that, like Ty said, do what feels right. The only thing I would ask is that you do try to mark off the test steps as you go through. Many people just check it on the board and then turn it on at the end, expecting everything to work first time. Often this is not the case.
Got my tube socket PCBs all done and installed it on the main board the two on the left are not lighting, I have swapped them all around and it’s definitely both of these two. reflowed the LEDs to the socket board and checked conductivity between pin number 2 and Pin number four on the next board , they have a good conductivity. before I replace the LED can anybody think of any further troubleshooting to do at this point . Is it possible to apply any external external power to the socket PCB itself to check and see if the LED lights up or not. Thanks for help
There's no practical way of testing those LEDs by applying external power. The won't light because they're "smart" devices - they need a serial data stream to tell them what color and brightness they should display. If you want to get further into how they work the data sheet is here:
It sounds like you've narrowed it down to the problem being with particular LEDs, not positions, or IOW it doesn't matter where you plug a bad one into the clock board, it's always bad. Right? The only thing I'm wondering about is did you try putting the "bad" ones in the rightmost (seconds) position to see if they might work there? As Ian described above, the data flows from right (seconds) to left (10xhours), and each LED has to pass the data on to the next LED in line. If a LED is bad in the seconds position that proves the problem is with that LED and not a case of the LED to the right of it failing to pass data.
Look into it later when the dust is clearing off the crater.
I figured it was something like that, that you couldn’t test them the way you do normal LEDs, I did have one spare with my kit so I removed the LED from one board and re-soldered in the spare and it works fine , so evidently I got it too hot when soldering it. So I will be needing to order a replacement LED. I didn’t see them for sale on the clock biz website so it looks like I will be eBaying them. Could someone look at this eBay link and see that I’m looking at the proper LED to order.