My Nixie Tube Mini-Tester
by Peter H. Wendt

Rel. 0.9.2, Last Update: Wednesday, 2001-12-19


Click Image to Enlarge

Originally I just had the idea to make testing of these Nixies a bit easier. So I built a small circuit from parts and bits I had laying around. It had no particular "Design" - basically just an oscillator and a 10-stage divider plus some driver transistors. I had the circuit connected to a big tube power supply that had 2x 6.3 VAC and regulated 180 - 280 VDC for the anode voltage. Worked fine so far. But was -of course- limited to one particular family of Nixies, the ZM1020 or equivalent types.
After a while I added a second socket when I got some ZM1040's, which have a different pinout. When I added the third socket to give me the ability to also test ZM1180 and compatibles it was a real rats-nest of wires. So I took a piece of experimental printboard and mounted the sockets into it.

Then things started an own life.

First and important: you should be able to track the anode current Ia. For most -but not all Nixies- the anode current Ia equals the cathode current Ik. Since you do test only one tube with only one active cathode at one time it would be sufficient to have one point in the circuit where a Multimeter in Milli-Amperes range could be attached to.

Second function: a decimal counter that activates all 10 cathodes sequentially to allow looking after the numbers (or symbols) if they are displayed properly. You should have a switch that allowes switching between "automatic" (perpetual 0 - 9 display) as well as "manual" steps to allow measurements for individual settings, when the Nixie has one particular cathode activated.

Third function: a testing point where a multimeter in Volts range attaches to and which allows reading the values for

• Supply voltage Ub
• Anode Voltage Ua
• Voltage at the cathode pins Uk0 to Uk9

The latter one is important when you wanted to find out to what "idle voltage" the inactive cathode pins go up when particular cathodes are activated. I use it to find out the value for a pull-down resistor, that must be used when the driver transistors cannot handle a voltage up to Ub and to find the value where the inactive cathodes are just not activated by the pull-down resistor. Ukx can go up to near Ua when no pull-down resistors are used.

The first part of the NTMT ("Nixie Tube Mini-Tester") is a power supply. I use the same circuit that is used on my Nixie Clock Project. It consists out of two 230 / 12 V 300mA transformers installed "Back To Back". This gives a relatively safe Ub of 200 VDC for the supply of the Nixie anode voltage.

Here's the schematic:

For the prototype shown in the pictures above I replaced the left transformer with a 12 VAC 500 mA wall-transformer, which originally supplied an old 2.400 baud modem. That's better and safer, because I don't have mains-AC in the unit.

Next in the NTMT is a circuit that generates a clock pulse - preferably with variable timing. I use the very common NE-555 8-pin chip for it. It is cheap, it is simple and it will suffice for the low needs. It is followed by a CD4017 10-stage counter / decoder, which advances one step with any positive pulse. Between the two is the switch for automatic / manual stepping.

This is the schematic:

Each of the outputs Q0 to Q9 on the 4017 is connected with a driver circuit - the same as I use on my Nixie Clock Project, consisting out of a 33 K resistor and a MPS-A42 transistor. For the NTMT you need 10 of the stage on the left side of the following schematic:



Finally there is the wiring for the Nixie-Sockets and the instruments switch. The instruments switch is a rotary type with 1 common pin and 12 positions (like the one made by LORLIN).

1. Testing the numbers
   
   

   Very simple done. Just insert a Nixie into the appropriate socket, set the mode switch to "Auto" and switch on the anode voltage. Now watch the Nixie cycling from one number to the next.
   Are all numbers displayed ? Does the Nixie "fire internally" (means: show unwanted glow along the internal wires) or are some numbers only displayed partially ? When done switch off the anode voltage and swap the Nixie if required.
   
   

2. Measuring the Anode Current
   
   

   Remove the jumper wire and install your Multimeter in mA-range. Set the mode switch to "Manual" and switch on the anode voltage. The meter should read something between 0.8 and 1.1 mA. Press the "Step" pushbutton to change the Nixie display to the next number. When done switch off the anode voltage and swap the Nixie if required.
   
   

3. Measuring the idle voltage on inactive cathodes
   
   

   Install a multimeter in Volts-range to the voltmeter connectors. The voltmeter must be capable to display voltages up to 200 VDC and should be a high-impedance type (1 MOhm or higher).
   Set the mode switch to "Manual" and switch on the anode voltage.
   With the instruments switch 1 you can select which of the voltages Ub (Supply voltage), Ua (Anode Voltage) or any of the Uk0 to Uk9 (Cathode Voltages) you want to test. The voltage for the activated number will read close to zero volts, because the driver circuit pulls it to GND for activation.
   
   

4. Testing for cathode pull-down resistors
   
   

   Basically you rig up your voltmeter as for measuring the idle voltage on the inactive cathodes. But you also connect a 2 MOhms potentiometer parallel to the voltmeter leads. Start with highest resistance. When you connect the potentiometer to the circuit you will see an immediate drop of the idle voltage - which goes down to zero when you reduce the resistance. This however will cause the corresponding cathode you'd selected with the instrument switch to light up.
   You will have to find a value where a) the idle voltage has dropped down to a safe value of e.g. 60 VDC for all positions of the instruments switch and for all displayed numbers on the Nixie and b) where the selected inactive number just doesn't light up.
   When using pull-down resistors you will however get a slight reduction in the Nixie display contrast, while all cathodes are pulled closer to GND and very, very slightly come up. They produce some sort of "ghost image" in the Nixie.
   You should avoid using a circuitry which uses pull-down resistors. This measuring however gives a good indication at which voltage (in respect to GND) a cathode starts to light up noticeably.
   

The design of the NTMT is limited to test Nixies that have one common Anode and have unmultiplexed cathode connections. Those "Bi-Quinary" Nixies cannot be tested without modifying the circuitry. Bi-quinary Nixies have two anode pins and only half the number of cathode pins. Two cathodes share one common pin. To activate a number the cathode pin and one of the two anode-pins must be selected. Unlike to "normal" Nixies which have a pinout of 1 anode + 10 cathodes these have a pinout of 2 anodes + 5 dual-cathodes. The reason was to save pins.

Also the Burroughs B-7971 multisegment Nixie cannot be tested. On normal Nixies the single anode is tied to a sufficiently high voltage across a current limiting anode resistor and there is only one active cathode at one time (plus probably a decimal point). On the B-7971 the principle of the alpha-numeric capabilities requires to have multiple cathodes activated at one time to form the shapes of numbers and characters. Here the anode is connected directly to the supply voltage Ub and for each cathode individual cathode resistors are used. On the B-7971 it is important to calculate the individual cathode resistor properly for the allowed cathode current. The anode current then is the total sum of all cathode currents.

Of course you should pick a better case than that what I made. I started with some experimental board and made the sides also from this stuff - simply using thicker silver wire to fix the board "pieces" inside. I didn't even bother with using nuts and bolts for most components.
But that's easy and you will surely be able to make it much better than I did.

In addition you might alter the driver circuit a bit (what I did in the meantime). Just cut the common GND-wire from the driver circuit and install two round jacks: you will be able to measure the cathode current then. If you use your normal desktop DMM for that purpose and don't have a separate instrument for that you need to either install a bypass switch or a jumper wire again like I did it on the anode-current testing point.
If you get instruments for 0 - 250 VDC and one or two instruments for 0 - 5 mA you could build your "professional version". No need to fiddle around with a desktop DMM and its meter leads anymore. No need for more connectors and more switches or jumper wires.

A nice bit is also to have control LED's for a) the low-voltage supply ("Power On"), b) the anode voltage supply Ua ("Anode Voltage On") and c) the clocking signal ("Stepping Pulse"). Still not enough stuff for the "Total Control Freaks" but gives some infos on the NTMT status.

You could install a potentiometer to make the anode current variable. I would recommend that you keep a fixed 22 KOhms / 1 Watt resistor installed to avoid problems when an accidentially short-circuit occurs. With a 22 K resistor the maximum short-circuit current at 200 VDC is 9 Milli-Amperes .... theoretically. In practise the unstabilized transformer output voltage will collapse and both, the effective DC and the current drawn will be much lower. The potentiometer should then be a 47 KOhm linear type in series with the 22 K resistor. That should give a setting range from 22 - 69 KOhms, which equals about 0.9 - 2.7 mA given an anode voltage of 140 VDC. This is within the range from 1 - 2.5 mA for most Nixies and will do for most testing purposes.

And -finally- building adapter(s) for the "wire-end" Nixies like the very common ZM1080 or equivalent. Probably a dead "Socket-type" Nixie like a ZM1040 or ZM1020 can be opened, the glass envelope cut down to the solid glass bottom plate with the wires. Then solder cables with little clips to these wire ends. That should do for testing most stuff. Don't forget to use different wire colors / clip colours to mark the corresponding pins.

The following table shows the influence on a ZM1180 Nixie when cathode resistors are used to pull down the inactive cathodes closer to GND to reduce the idle voltages.
The supply voltage Ub is 195 VDC, the anode voltage Ua is set at 135 VDC with a 54 KOhm anode resistor. The anode current Ia is 1.1 mA in this case. I used 470 KOhms resistors to pull down the inactive cathodes.

The following table shows vertically the tested pin and horizontally the number shown in the Nixie display. The upper voltage is the one without, the lower value with the pull-down resistor installed.

As you can see the values for the circuit without the pull-down resistors reach values up to 125 VDC - while the measures for the circuit with the resistors installed stay below 60 VDC.
After reading that you might conclude that it were recommended to have the pull-down resistors installed, but -No- this is not always true. As explained a bit further above: the pull-down resistors cause the cathodes that should stay inactive to lightly light up. This louses up the contrast on the active cathode within the Nixie. You should really save this sort of idle voltage reduction if you don't have any other choice, like - for instance - in a driver circuit layout that requires to have transistors or driver-chips with Ucb0 below 60 VDC.

This is a first, rough and incomplete list of Nixies that I'd tested with the NTMT so far and which seem - within limits - to be interchangeable. You should however consult a datasheet if you manage to get one. I did not use "primer" pins or such like, just the bare cathodes that display numbers or symbols.