Nixie Tube - Current across anode

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
Amateur nixie-er here,
I have a handful of z5900m tubes that I just tested (their specs below).
The issue is, when I use a 20Kohm resistor on the anode (calculated by supply voltage - maintaining voltage / current), the anode current is much too high. Here are the results of my testing (resistor value - current):

20K - 9mA
30K - 6mA
47K - 3.7mA
56K - 3mA
68K - 2.5mA
82K - 2.1mA
91K - 1.9mA

When I finally got the current down to a "safe" level, the digits were much dimmer, but still lit just fine. The maintaining voltage was at 130v.
Any idea why the resistor has to be so much larger to drop the current to a recommended level?
Now that I think about it... this type of tube has two anode leads. I connected both to one 20K resistor. Do each need their own resistor?

--- Tube Specs ---
Min. Supply: 170v Min. Current: 1mA
Ignition: 150v Max Current: 2mA
Maintaining: 140v Max Pulse: 10mA
Extinguishing: 120v
 

ebp

Joined Feb 8, 2018
2,332
What supply voltage are you using?

The voltage across the resistor for all of you tests is around 170-180 V, which suggests you must be using a power supply of around 320 V (rectified & lightly filtered 240 VAC?)
 

ebp

Joined Feb 8, 2018
2,332
It simply does not add up.

Take your last line:
91k & 1.9 mA = 172.9 V
If the voltage across the tube is 130 V, then the supply must be 173 + 130 = 303 V
If the supply voltage is 180 V and the tube voltage is 130, then the current measurement must be wrong. A supply of 180 V and a tube voltage of 130 V with 91k should yield 0.55 mA
 

ebp

Joined Feb 8, 2018
2,332
What type of power supply are you using?

[EDIT]: Is this with steady drive or some sort of pulse or multiplexing scheme?
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
This is a ~170v direct drive power supply. Is the anode voltage actually the difference between supply voltage and maintaining voltage? If so that does yield 93.33kOhms from the calculation, which is what was seemingly necessary. Seems like a much larger resistance than other similar nixies though, and the digits are quite dim compared to when I was running it with the 20K.

If it is relevant, the digits took a couple seconds to fully dissipate. This seems out of the ordinary compared to other nixies as well.
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
Dug a little into nixies with two anodes, and found that they are "bi-quinary". This appears to be exclusive to the z series of tubes, and require a resistor on each anode. Source:
http://www.dos4ever.com/nixie1/nixie1.html#zm1030
The diagram in the link above shows a different resistor value for either anode, however; I don't understand why this would be necessary. I suppose I will need to do some more testing.
 
Last edited:

ebp

Joined Feb 8, 2018
2,332
The sustaining voltage is the anode to cathode voltage once the digit has turned on. Since the cathode is normally either open circuit or connected to the supply negative ("ground") by a transistor switch, the voltage measured at the anode relative to ground would be within a volt or two of the actual sustaining voltage. It takes a somewhat higher voltage to get the tube to "strike" initially, which is why a supply of around 180 V is typical. Because the current flow is very small at the start of striking, the voltage drop across the anode resistor is small so you get essentially the full supply voltage to cause striking. Once struck, the current rises, the anode-cathode voltage drops, and voltage is dropped across the resistor due to the current. The resistor serves as what would be called a "ballast" for other lamp types such as mercury vapor or fluorescent lamps (which are also mercury vapor lamps). The sustaining voltage won't change very much with current that is within the allowable range.

Can you post a drawing of how you are measuring the voltage and current? A photo of a hand-done drawing on paper is fine as long as it is legible and right way up. At this point I'm baffled and can't make sense of the values.

Try measuring the current in the anode circuit then in the cathode circuit (for the same lit digit use for the anode current measurement). They may be slightly different due to leakage in the digits that are supposed to be off.

I don't know why the anodes would require different resistors. It may have something to do with the screen not being connected, but I'm really just guessing.
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
This is how I ran the circuit, and where I measured with a multimeter. I modified the original image to account for both anodes. One thing that I did incorrectly was run both anodes through a single resistor; I was not sure how to handle both anodes when testing this earlier. Perhaps that is the issue.
nix_th01Edited.jpg
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
UPDATE 9/17:
Tested the tube with only one anode attached to ground with 20kOhm resistor. All numbers lit up, so the anodes are not exclusive to their respective digits (e.g. one to evens, one to odds). Tried placing 20kOhm resistors on both anodes, and the current jumped to 18mA at either anode. I'm really not sure what the correct wiring procedure is for this thing to get it stable with a 20kOhm resistor. Really seems like it needs the much larger ones to keep the current at a reasonable level, but the datasheet says otherwise.
 

ebp

Joined Feb 8, 2018
2,332
Since the tube isn't designed for multiplexing, I suspect the two anodes are probably intended to improve brightness uniformity.

You didn't answer my question about how you were measuring current - in the anode circuit or in the cathode circuit.
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
I was measuring the current in the anode circuit, as shown in the image in one of my previous posts. Haven't measured the current in the cathode circuit and comparing them. Will do later today.
 

ebp

Joined Feb 8, 2018
2,332
For each test you also need to measure and report
  • anode to cathode voltage
  • voltage across the current limiting resistor (just the resistor, nothing else)
  • supply voltage
I simply cannot rationalize your previous measurements.
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
Alrighty, here are the results from the latest tests:

Supply voltage: 181V
Anode voltage: 142V
Cathode voltage: 181V
Resistor voltage: 37V
Current across tube: Still ~9mA

All voltages are +/- 1.0 due to multimeter rounding / indecisiveness.
 

ebp

Joined Feb 8, 2018
2,332
Again your numbers make no sense.

How is the cathode voltage 181 V?
What is the value of the resistor? IF I'm interpreting your numbers correctly, they mean you were testing with a 4100 ohm resistor.
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
Woops, just measured again, and it's actually 46.2V. I somehow measured that incorrectly the first time.
The resistor is 20K. To be specific, it is two 10Ks in series. Though that shouldn't matter.
 

ebp

Joined Feb 8, 2018
2,332
You're really making this difficult!
What is 46.2 V?

If you have a 20k resistor, a 181 V power supply and an anode voltage of 142 V, with the resistor between the positive of the power supply and the anode, then the current is (181 - 142)/20k = 1.95 mA.

So I return to my question:
EXACTLY how have you measured current? It would appear to me like you are connecting a current meter from the anode to the negative of the power supply.
 

Thread Starter

TheMobius40

Joined Sep 13, 2018
28
Hoo boy, I should've named this thread stupid questions. Forgive me, and thank you for your patience.

I'm afraid, yes, I have been measuring the current from the anode to the negative of the power supply.
I suppose if it wasn't clear before, I don't have a lot of experience with the hardware aspect of things yet, just components on paper.
What is the proper way to measure this current?
 
Top