Driving Nixie-tubes with arduino

[RocketScienceSmurf]

There are many circuits out there using arduino and some extra components to drive nixie tubes but I am trying to figure out a cheap way that don't take up too much space so I would like to use ULN2003 if possible.

When looking at the IN12 Nixie tube drive voltage seem to be around 145V DC and recommended current is 2,5-3,5mA. Do these thing behave like a resistor? Can I assume each digit is like a ~48k resistor? If so I should be able to use a bussed resistor array with 7 pcs of 22k resistor to make sure the ULN2003 is never exposed to more than ~45V (50v is ABSOLUTE MAX according to the data sheet).

Assuming that the internal DC resistance of each digit is never LOWER than ~48k I should be safe right?

The other potential problem is the current when the digit is NOT lit. There will still be close to 2mA running through each digit. Will they be dark or will the probably be dimly lit?

[klugesmith]

Nixie tube digits are not even close to constant-resistance devices. 
The normal glowing state is much like a LED: small, and sometimes negative, voltage change with large current change.  There are 10s of volts of hysteresis at the off/on transition.

A digit with 2 mA will be seen as plenty On.



A Google search for ULN2003 Nixie revealed some circuits that work. 
Your mission is to learn how they work.

[Mads Barnkob]

When I built a Nixie tube clock I used common resistors for the Cathodes. I used high voltage NPN transistors MPSA42 to interface the 5V logic to driving the individual numbers.

http://kaizerpowerelectronics.dk/general-electronics/nixie-tube-clock/

[RocketScienceSmurf]

Perhaps SN75468 with its 100V rating would work instead? A transistor array instead of using discrete transistors would save space and make it much easier to connect.

[johnf]

the best way is to find some 74141 ics that were designed to drive nixies direct from TTL logic