How are Nixie Tubes triggered?

Thread Starter

Rocky_circuits

Joined Nov 1, 2011
57
Just a quick question for my nixie tube clock project, how do the tubes operate?
It has the one cathode, or anode I forgot which. Then it has 10 pins for the rest of the numbers. If I am not mistaken, these pins must be operated with high voltage (150-200 maybe) depending on the data sheet. So that means I would be able to use 7400 series chips then.. right? Since I believe those can only operate at about 5 volts.

I wish they could be operated by being powered by whatever high voltage they needed, and then having 0-9 triggered by the 5 volts... Can anyone lead me in the right direction with this?
 

Wendy

Joined Mar 24, 2008
23,415
The anode is high voltage (+180VAC) with a current limiting resistor. The cathode is normally open, a transistor collector is connected to them, and when the transistor is on, that lead is connected to ground.

You could look up magnet18 circuit, and see how he did it. As long as you aren't lifting the design wholesale there is nothing wrong with referencing to previous art.
 

shortbus

Joined Sep 30, 2009
10,045
Here are some links to Nixie tubes and clocks found while looking for something else. Actually the whole site is very informative. http://mysite.du.edu/~etuttle/electron/elecindx.htm

This one explains the tubes and what is needed for them. Along with other displays. The Nixie part is about 3/4 of the way down the page. http://mysite.du.edu/~etuttle/electron/elect46.htm


This one is about clocking circuits and a Nixie clock circuit. Again about 3/4 down the page. http://mysite.du.edu/~etuttle/electron/elect48.htm

Hope it helps.
 

Thread Starter

Rocky_circuits

Joined Nov 1, 2011
57
Hmm.. It appears a lot of people like to use alternating current as their oscillation. I was hoping to do something a little different.
I'd like to take a crystal oscillator and hook it up to the CLK of a 74161 chip (4 bit binary counter)

This would then technically go to a decoded 7 segment display, and set up so forth with a few AND and NOT gates to make it all work. But of course, I do not want to use an LED 7segment display. I'd like to use nixie tubes.. Now, is it at all possible to take this 5 volt circuit and have it magically turn into 180 volts before it gets to the nixie tube? Maybe through use of transistors or something of the sort. Solid State relays would work great for this, but it's not very practical to buy 40-60 of them.... Rhose 10 pins are ground pins I guess though. But then I could just wire some things backwards for that.

Does what I am saying make sense?
 

SgtWookie

Joined Jul 17, 2007
22,230
You connect 180v to the anode using a 15k to 24k resistor.
Then you short the cathodes to ground using high-voltage NPN transistors in order to select the digit.

You can use the TTL outputs via resistors to turn on the high-voltage NPN transistors. Since little current through the Nixie tubes is involved (maybe 2 to 3mA), your transistor base current can be very small (1/10 that amount).

Have a look at these STPSA42's that Digikey is discontinuing:
http://search.digikey.com/us/en/products/STPSA42/497-2407-ND/603432
$0.02/ea. Fifty for a buck - such a deal!

They are similar to the MPSA42/MPSW42, but those are ~$0.40/ea if you're buying 50. Might as well buy 100, and get them for a bit less due to the discount. Would you rather spend ~$1, or ~$19? Your choice.
 

Thread Starter

Rocky_circuits

Joined Nov 1, 2011
57
That is quite the deal, I may just get those even if I don't use them all for this project. Wait, can't i just use a 74141 decoder? Would that not act as those high voltage transistors.
 

SgtWookie

Joined Jul 17, 2007
22,230
The 74141's output maximum voltage is 60v. You're using 180v.

I don't know why you're using a 74161 anyway. Why don't you use 74HC4017's? You will be able to use far fewer parts that way.
 

Thread Starter

Rocky_circuits

Joined Nov 1, 2011
57
Those chips would be better.. I'll consider using them. It's just I already designed the circuit using 74161 binary chips in my clock circuit and it all works (YAY!)

About that 74141.. I don't doubt your knowledge, but those chips should work great for this. I realize that its maximum voltage is 60 volts, however the 74141 has 60 volt zeners on the output. Let me copy over some information I found on this dutchforce forum:
"Bingo, 74141 has 60v zeners on the outputs.

So. say our supply is 180V.
With 74141 "on", it's output transistor's collector is essentially shorted to ground. 180V is across the tube & resistor.
When off, the collector is open, so we have120V across the tube and resistor, 60V across the zener. (this should be enough to extinguish the nixie)." -Colt45

"I found that with +180 volts connected to the anode through a 33k resistor, and all of the cathodes open circuited, and using a voltmeter with 100 Meg input impedance, I wasn't able to read more than about 60 volts on any cathode. So the dropping effect across the Nixie seems to work even when there's virtually no current flow. " -Village Idiot
Along with a lot of comments saying that the data sheet for this chip isn't very good..
So with that in mind, I believe I have this whole circuit worked out. I'll post back again if it all works (few weeks till then) and see what you guys think :)

Thank you so much for all your help!
 

SgtWookie

Joined Jul 17, 2007
22,230
OK, but... keep in mind that the '161 circuit would be great for an LED clock driven by a 7447, and you can use the 4017 with the Nixies - then you'd have two clocks. ;)
 

Thread Starter

Rocky_circuits

Joined Nov 1, 2011
57
Haha I suppose it would be even more :p. This whole nixie deviation should be worth quite a bit of extra credit though considering the original project is to design a clock circuit powering some simple 7-segment LED's. But with time permitting (hardest part getting the parts for the nixie since everything is in russian or europe apparently) I'll make two clocks.

I do have another question about this 74141 chip.. but I'll ask it in the other thread since Magnet18 found the other one and as you have mentioned, he has a lot of experience making these clocks.
 

Wendy

Joined Mar 24, 2008
23,415
Like I said, the 4017 is a mainstay, I use it for many, many projects. It can be modified for a lot of different patterns, uses a 3V - 18V power supply, a typical CMOS chipie.

I've even thought about using it for a 60 LED counter, like the hands on a clock.

For the master oscillator you can not beat the CD4060. It will use a 32.767Khz watch crystal to generate 2Hz, which can be converted to a precision 1Hz with a simple flip flop.
 

Thread Starter

Rocky_circuits

Joined Nov 1, 2011
57
Oh I definately will be using 4060 for the oscilator. I made the crystal oscilator in a lab before so I should be golden there.

I may be converting to the 4017 side now.. I just realized a flaw in my circuit that for some reason is allowing the hours to count 0A-0F. But only when I have the condition of when the hour displays "12" to reset. There's to much logic in logic D: soo time to explore the world of this decade counter :D
 
Top