UnintegratedCircuit
Part 3 - The IN-19V Tube and PSU
​
The IN-19V Specifications
Since catering to the specifications of the IN-19V Nixie tube is likely to be the most difficult part of this project, it makes sense to start here and work around it, rather than trying to build it into existing circuitry. The first place to start is the datasheet (as always), an excerpt of which is shown below in Figure 1. This snippet was originally taken from Dieter's archive of Nixie and other display tube datasheets, but a working link can be found for this here.
​
Now, while this excerpt is exclusively in Russian, common sense (or the Google Translate app) is all that is really needed to decipher which figures relate to which specifications. The 'B' character is actually the Russian equivalent for a 'V' (i.e. volts) in English and 'mA' should be fairly self explanatory. Since it is already known that the Nixie tube requires a high potential to 'strike' and illuminate a cathode, the top few lines probably relate to voltages that determine the illumination of cathodes.
​
In fact, the top two lines here relate directly to the striking voltage stating that it will be no more than 170V and no less than 120V. Both of these figures are actually important since we know that whatever power supply solution we use must be capable of providing at least 170V in order to guarantee that ANY cathode on ANY tube will be able to illuminate. The minimum voltage however, is needed in order to determine the voltage rating of the open-collector transistor drivers that will be used to interface the logic circuitry with the cathodes themselves. If the total voltage across the tube (i.e. the supply voltage minus the transistor breakdown voltage) is allowed to exceed 120V, there is a potential that some cathodes in some tubes will remain permanently illuminated since they are always able to strike.
​
The third line gives what is known as the 'sustain voltage' and is needed in order to set the current that flows through the tube once a cathode is illuminated. This sustain voltage is a property of all neon lamps, Nixie tube or not, and occurs due to the ionisation of the neon gas inside the tube. When the gas is in its normal state (when no cathodes are illuminated), its conductance is very low and hence its resistance is high and, ideally, no current flows through the tube. In order to ionise the gas and make the cathode illuminate, a high voltage is needed; however, once the gas is ionised, its conductance increases and hence its resistance drops. Due to this drop in resistance, a lower voltage is required to maintain the ionisation of the gas and consequent current flow through the tube.
​
Shown below in Figure 2, is a voltage curve of a neon lamp, emphasising the strike and sustain voltages.
​
The fourth and fifth lines in Figure 1 give current specifications for the tube, likely a maximum continuous current for DC operation and a peak current for rectified mains operation respectively (since it will not conduct for a fair proportion of the mains cycle so the higher peak current is acceptable).
​
This small snippet of datasheet provides us with all the information we need in regards to designing any Nixie tube related circuitry: power supply voltage, driving transistor voltage rating, and a suitable value for the anode resistor.
​
Choosing The Power Supply
The power supply for this project was of my own design and its design process is described in depth here. Essentially it is a flyback converter based around an IC that was originally intended for charging large photoflash capacitors to a high-voltage. I then modified this application in order for it to provide a 180V output to a low-value capacitor (the stereotypical output capacitor found in most SMPS's) from merely a 5V input.
​
The only modification from the original schematic was a slight tweak in the feedback resistor (R1) in order to nudge the voltage down to about 170V (I believe it ended up at closer to 166V) in order to reduce stress on the cathode driving transistors. This lower voltage was verified to be acceptable with THE EXACT tube I would be using in the final build, other tubes may not have all cathodes strike (although since it is so close to the upper end of the specification, it should not matter too much).
​
Of course, a module from ebay could easily be susbstituted for my module; however, these modules rarely cater for a USB input - most are only specified for operation at 9V or 12V - and if they do cater for a 5V input, they are often considerably more expensive - probably about £15 for one module in 2020 as opposed to a similar cost* for 5 of my homemade modules. The final reason for picking my module is that it is small, very small, having a footprint of less than 1 square inch (22mm x 25mm or 0.85" x 0.97"). Having said all this, if money, space, and input voltage permit, buying a module off of eBay would be a quick and easy solution.
​
Now that the power supply has been decided upon, the next installment of this project will decide upon choosing a suitable cathode driving arrangement and sizing the anode resistor appropriately.
​
​
​
*At the time of writing this, I was a university student and was, therefore, able to obtain samples from numerous companies. In making this module, the two most expensive components, namely the flyback transformer and the controller IC, were obtained in small quantities for free by Coilcraft and Analog Devices respectvely. In fairness this does save about £5 of cost per module.
​
​
​