Yeah I was a dummy and forgot to relabel the pinout before I uploaded the schematics to the manufacturer (still a noobie with creating custom PCB's). I had the pins inverted because the autoroute option in my program was creating about 7 or 8 via's and I wanted as little complications with the design as possible so mounting it the other way created only one via under one of the IN-17's.
Reflashing the ATMEGA would explain the wifi issues. I put an order in for an ESP8266 to USB board and that will arrive Jan 5th roughly.
Since it's the weekend when I get home I have doubles of all my parts minus the nixie driver and wifi boards. I'll do another full sweep of the routes for my wiring. Once the adapter ships in I'll reflash the wifi module and give it another shot.
I think your routing is alright on your board. If all the digits work, they are all the same number, and they all change synchronously in test mode, that pretty much confirms it for me.
I'm really pretty sure you'll find something in the interconnection of the 2 boards. Individual wires on the pins to connect them would allow you to test them 1 at a time if necessary. Even double checking that cable with an ohmeter (with power off of course) to verify the connections are actually to the right pins would be a help.
The test sequence starts at "0" with the LED's lit "red". The LED's are once again "red" when it reaches "6". At "4" and "8" the LED's should be "blue". I've just verified this on my clock. That should be enough to tell you if the clock is displaying the same thing that it "thinks" it's displaying.
I don't have any LED's hooked up at the moment. I was going to use far too many than the driver would handle so I have a completely separate power supply and LED setup. I do, however, have both common anode and common cathode RGB LED's laying around. I'll get some of those set up and see if the color coordination synchronizes with your description. PS, thank you for all the words of advice and help.
Just posting to get myself subbed to the thread (since my global notifications seem to have quit working since the big upgrade). I have nothing to add at this point - Jim is IMHO on the right track with suspecting the connections. If it was the K155ID1 all tubes would exhibit the same problem since the same K155ID1 serves all of them.
Look into it later when the dust is clearing off the crater.
If you have a switch over of some digits, this is the place to change it, rather than pulling up tracks. You go into the table with the index of the digit you want to display, and the result of the lookup is the digit on the K155 that gets set. So if you go in with "0", you get "2" out (index 0 of the array results in the value 2), if you go in with "1" you get "3" out, and so on.