Adventures in Nixie Tube Socket Manufacturing


Not too long ago I got two old cash register displays full of perfectly functional nixie tubes from a friend. My goal was (and is) to use these tubes to build a pair of clocks. I am by no means new to electronics or programming so I undertook this project completely unaware of the challenges that lay before me.

Look at these beauties:

IMG_0244

My initial thought was that I would just use the board as it already existed. It seemed as though it would be a simple matter of pulling the datasheets for the tubes, and tracing the connections back through the big mess of on-board transistors to figure out how to to drive it. This actually didn’t work out as well as expected for a few reasons, the boards were corroded pretty heavily, there seemed to be additional boards in the original cash register required to power the tubes, and lastly, many of the electrolytic caps on the board were dead.

Knowing all of this, I undertook the process of removing the tubes and sockets and designing my own board.

Removing the tubes was actually quite easy, they just pull right out of the sockets (use a paper towel or gloves to avoid transferring skin oils to the tubes).

IMG_0245But removing the sockets wasn’t going to be that easy. My attempts to desolder the sockets and extract them failed hilariously, The solder on the boards seemed to be some sort of ancient irremovable metal and even my +15 wand of soldering was incapable of phasing it. In a final effort, I got out my heat gun, turned that sucker up and attempted to get the whole board hot and yank the sockets free.

IMG_0246I then learned that since sockets are made of ceramic they explode when you yank on them with pliers and sling shards everywhere that stab you in the feet for days to come. Equally ineffective was attempting to use the dremel tool to carve the board up and remove sockets one pin at a time.

It was at this time that I realized that I should test to make sure the tubes themselves even lit up before spending anymore time on the sockets.

IMG_0214VICTORY!

So now I had to come up with a clever way to either build my own sockets, or set these free, I chose to build my own.

I fired up SketchUp, dug out the data sheet and started modeling them out.  One site I used heavily was tube-tester, an online compendium of nixie tube knowledge. From there I was able to figure out that the socket my tubes used was called an RTS-11 and at what radii and angles the pins needed to be placed. This is what i came up with:
sketchup_socketI kicked around the idea of making these on my CNC machine but decided that all the heat-proof materials I could find came in huge sheets that would cost more than just ordering 3d printed sockets from shapeways. So I had them made of one of the materials that can handle temperatures higher than the max temperature the tube datasheet gave. After a week these showed up at my door:

IMG_0247

But, there was a problem:

IMG_0253I’m still not sure how this happened (probably something incorrect in the model I uploaded)  but there was a thin layer of plastic where the cylinder met the rectangular base.
Tactical Solution:

IMG_0249

IMG_0255Problem solved! And now the tubes fit beautifully into their new homes. Bonus, they face forward, something I wasn’t one hundred percent certain I did correctly.

IMG_0252

Next time I will be designing the control board and getting that fabricated, so wish me luck!

Leave a Reply

Your email address will not be published. Required fields are marked *