Nixies: stupid questions.

Using Arduino Displays
#1

Firstly, I wasn't sure whether a nixie thread should go under displays or LEDs/multiplexing, so I plumped for here. I reckon I can get away with it, I've been here a while but right now I'm feeling like a complete n00b and that I haven't learnt anything in the eight or so months I've been messing around with arduinos. Anyway...

I have a nixie tube socket wired up to a breadboard (with a nixie in it). I have suitable 180v, 12v and 5v supplies. I have a 74141 (although it's russian so it's labelled differently. I have checked and the pin out is identical).

I have tested the nixie by running individual pins (not using the 74141) and it works fine.

I then wired up the chip as per the usual pinout (example), using 5v for the Vcc and popping the four logic pins into the arduino. I then knocked up a basic sketch which switched the pins low/high according to the number, aiming to go from 0 to 10 in sequence.

This is where I hit a problem. Numbers stayed illuminated. I'm afraid I wasn't paying enough attention to spot whether some or all of the previously illuminated numbers stayed lit up (they weren't going in the correct order but that shouldn't matter). I removed the arduino and some numbers stayed illuminated (powering off before doing anything). I set all the data pins to ground, no change. I removed the Vcc, no change. I tied the Vcc to ground, no change. I removed the data pins and Vcc, no change. I removed the ground pin - no output. Put it back, no change. The same digits were lit up.

Have I broken the 74141? Is there something I need to do to clear the chip when changing numbers?

#2

The nixie is running on 180v. The 12v powers the arduino and nixie PSU. The 5v is for chip logic (vcc input).

I have the nixie hooked up to the 180v + and the ten other nixie leads linked to the chip.

The chip has four logic pins coming from the arduin (5v), one 5v Vcc, ten nixie -ve connections and one ground connection.

I'd a tad confused - why would a nixie control chip not run at nixie voltages? Everything I read beforehand said I needed 120 - 180v to power a nixie.

#3

Nixies are gas discharge devices - when not conducting they are basically open circuit, and to conduct a minimum strike voltage is required across the electrodes. When conducting they need at least a minimum voltage to keep the discharge going.

The upshot is that driver circuit never gets to see the full supply voltage, so it is possible to drive with parts rated less than the supply as the off-state leakage current of the driver is plenty to keeps the nixie cathodes at a low enough voltage. The on-state voltage of the driver will be low if the drivers are low-impedance - again no problem.

The only problem is the switching transients as the nixies are conducting and the driver is not fully on - the difference of the supply and minimum nixie-on voltage will be present briefly across the driver - if this is too high the driver will be damaged. So if the supply is 180V and the min on-voltage is 80V, the driver needs to handle 180-80 = 100V...

So if you have datasheets for both tubes and drivers you should be able to work out if its a safe combination - the supply voltage can be lower than the nominal nixie voltage so long as its above the strike voltage - the tubes will be dimmer.

Also I seem to recall that you are supposed to use load-resistors to drive nixies, and these can drop some of the voltage and help protect the drivers.

#4

Are you using an anode resistor? That's crucial. It sounds to me like something's up with your driver IC. Can you post a schematic?

#5

wyager: Are you using an anode resistor? That's crucial. It sounds to me like something's up with your driver IC. Can you post a schematic?

No, what size resistor?

My circuit is as described - no resistors or caps or anything. The nixie plugs into the IC and 180v+, the arduino plugs into the four IC logic pins and the IC is also linked to +5v and GND. All the grounds are linked.

#6

OK, that's a problem, but I'm not sure it's causing your problem. What nixies are you using? Most people reccomend between 15 and 33k resistors, but it really depends. What you want to do is find a resistor (and you will have to experiment between those two values) where when you measure the voltage across the resistor and divide it by its resistance to find the current passing through it, and it should the recommended (you could go a little under and be fine) current for the nixie tube (small tubes tend to be around 1.5mA, medium tubes 2.5mA, and large tubes 4+mA. A model number would help). You have to experiment because of the variable nature of the nixie tubes, no one is exactly alike. So if you use a 15k resistor and find that the voltage dropped over it is 30V, you know that you are passing 2mA through your nixie tube which should be fine.

Also, what are you using as a power supply? If your 180v power supply does not have a built-in output filter cap, you will want to install one. 1µF @ 250v is probably enough to adequately filter the voltage.

Also, one possibility is that your tube has a severe internal short or something (but this is very unlikely). I recommend (after installing the resistor) individually testing each digit (without a decoder IC) to make sure the tube works well.

Good luck with your nixie project, as I think you know I am doing the same thing. ]:D

#7

CowJam:

wyager: Are you using an anode resistor? That's crucial. It sounds to me like something's up with your driver IC. Can you post a schematic?

No, what size resistor?

My circuit is as described - no resistors or caps or anything. The nixie plugs into the IC and 180v+, the arduino plugs into the four IC logic pins and the IC is also linked to +5v and GND. All the grounds are linked.

For what I know this should work. I have a couple of nixies here and I am starting to work with nixies those days. I will give a try to this setup (that should work, in my opinion) F

#8

CowJam:

wyager:
Are you using an anode resistor? That’s crucial. It sounds to me like something’s up with your driver IC. Can you post a schematic?

No, what size resistor?

My circuit is as described - no resistors or caps or anything. The nixie plugs into the IC and 180v+, the arduino plugs into the four IC logic pins and the IC is also linked to +5v and GND. All the grounds are linked.

The resistor depends on the specs of the Nixie in question. For instance I have some sub-miniature Russian IN-17 nixie clones that are speced for 170V and 47k series resistor. The sheet I have says they will work at lower voltages and upto 82k series resistor. The series resistor goes on the anode to avoid having resistors for every cathode (which can directly connect to the driver).

That said I haven’t tried interfacing these tubes yet, I’ve merely verified they light up.

#9

Thanks. They're number 12s. I got a dozen of them, sockets and the ICs from Ukraine (gotta love ebay!) - they look Russian (cyrilic writing on them).

I'll have a play tonight and see what else I can break ;)

#10

I've got them working, thank you everyone (especially wyager).

I slapped a 33k ohm resistor on the anode and it's working fine now using a fresh IC (I've permanently damaged the first one).

#11

Excellent! It was probably just too much current that damaged the old IC.

#12

Yeah. I'll get the hang of this electrickery stuff eventually!

#13

Spoke too soon! http://arduino.cc/forum/index.php/topic,50274.msg358237.html

#14

No! That's awful! =(

#15

I am following your experiment and as soon as I finish my pov experiment I will apply them to nixies... Where did you find the datasheet for yours? I have just tried to light up my IN-14 from my DCDC power supply but I am quite sure that the others little blulbs that I have, IN-3 (just neon bulbs) can't work at 170v...

#16

I googled for it. Not sure I had a datasheet for my actual nixies, but I got the datasheet for the chips (which came with the nixies) by googling, and this is it: http://cowjam.co.uk/arct/k155id1.pdf

I've got some IN-35s which are just neon bulbs and they run fine at 180v (or whatever my circuit is running at) and ran fine with or without a 33k ohm resistor. These ones: http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=390206146006&ssPageName=STRK:MEWNX:IT#ht_864wt_905

#17

We bought nixies from the same vendor :slight_smile:
Mine are those
http://cgi.ebay.it/ws/eBayISAPI.dll?ViewItem&item=400174245756&ssPageName=STRK:MEWNX:IT
I am going to try to drive them at 170v and see…
After reading about your experience I was thinking that multiplexing and optocouplers should be the safest way…

#18

Yeah, optocouplers are the way to go!

If I figure out what happened I’ll report back on here.

#19

BTW guys, small neon bulbs should have at least 100K resistors (some use 200K) at 170+V or else they could burn out themselves or your circuit. I am able to multiplex them at 170v with a 100K resistor. Remember that the gas inside them will change its resistance until the voltage across the lamp is the sustaining voltage. A 200K resistor (which is probably better for long life) makes them look a little sputtery.

My A1As regulate themselves at 60V, which means that 110V gets dropped over the anode resistor. 110V/100000?=1.1mA. Almost twice the recommended amount, but I figure since I'm multiplexing... Still probably not healthy.

#20

wyager, thanks a lot for the advice. I am going to follow it!