DIY Neon Lamp bulb / Nixie tube power supply
Neon lamp bulbs (the high output ones I have at least, which are C2A-ND) need about 95vAc or 135vDC to ignite. Placing a 30k Ohm resistor keeps the bulb from exploding. Later on, in the oscillation circuits I'm building I'll be using some NE2-A1A bulbs which only need 65AC or 90DC volts to light up ("strike")...I'll be using those with my new Agilent 0-120V DC power supply.
I didn't have a 30k resistor handy, so I just put two 15k ones end-to-end, which totals 30k. You only need 0.25 watt resistors, the huge blue ones I used can handle 2 watts, which is why they are so huge.
When you plug a neon lamp bulb into an AC power source both the cathode and the anode glow.
When you plug a neon lamp bulb into a DC power source only the negative side will glow. Swap wires and the other side will glow.
Getting 95v of AC power is easy: just hook a variac into a wall outlet and ramp up the juice. Some people just plug the neon bulb (with a resistor soldered to one leg) directly into a wall outlet, but that's dangerous because itsi easy to touch or cross the legs.
Finding a 135v DC source is pretty hard-unless you make your own.
I took a disposable film camera with flash and took the circuit out of it. This is very dangerous because there is a HUGE capacitor inside which can kill!
Before I took apart the camera I fired the flash, then I removed the battery, then advanced the film ratchet inside the camera forward to recock the shutter and fired the flash. This helped to bring the voltage in the capacitor down from its 350v level.
Once the camera was open I removed the circuit board without touching the components because they could still shock! Luckily with I put a voltmeter on the two wires coming out of the capacitor it read 30 volts, here is where I put a large resistor across to bleed away the rest of the voltage-sometimes I put the blade of an insulated handle screwdriver to discharge a capacitor-but that makes a loud bang and huge spark and can actually blast away some of the screwdriver metal!
Once the capacitor was reading only 5 volts I put a screwdriver blade across both it's wires and nothing happened: safe!
Then I cut out the capacitor (two snips) and then attached a wire to each spot that they connected to the board: those wires then when to my neon bulb (and resistor) legs.
At this point I put the battery back in a used a pair of insulated pliers to hold the circuit while I used an insulated screwdriver to depress the "on" button on the circuit board: success!
My cheap voltmeter couldn't measure the high frequency voltage going to the output wires. The readings just bounced all around, but the high output neon bulb lighted right up.
After removing the battery the board is still dangerous: even though the huge death capacitor is gone there is still at least one more capacitor on the board that holds enough energy in it to give a nasty shock! For safety I'm going to put this whole thing in an insulated container; once I figure out a way to either press the on button from outside without getting shocked, or solder a wire to across the on switch to make it permanently on.
Since I have a bunch of low voltage DC supplies I soldered a wire to each of the AA battery holder. I'll run these wires to a 1.25vDC power supply to power this setup. That'll save on batteries. That will give me a power supply that gets 1.5vDC input but outputs 135vDC+.
Next, for fun I cut out the actual xenon flashtube. It didn't hurt the circuit and I hooked it up to my 5000v transformer and variac: at about 1000v there was a purple thunderstorm looking action in the bulb: many jagged strings of lavender lightning in the little tube.
Of course AC power is better for Neon Lamp Bulbs: both filaments (actually electrodes) are glowing, the voltage is lower and alternating at 60hz so the "wear and tear" on the filaments is spread out. When only one electrode is glowing because of DC power it is: turned "on" 100% of the time (not alternating) and all the (higher) voltage is sent to a single electrode. This is why neon lamp bulbs last about 50-60% longer when fed by AC power.
Neon Bulbs and Root Mean Square AC Voltage
120v AC house current fluctuates between -170v and +170v. So
why don’t we say that the AC house current is 170v? Because if you average 170v AC you get zero!
-170 + +170 = 0 0/2 gives you an average of 0v. So a meter that reads an "average" voltage would always show the wall outlets in your house as having "0v", which would be mathematically correct, but obviously weird!
What if the meter showed you exactly the voltage? Your meter would show -170, -169, -168...+168, +169, +170 volts; all cycling 60 times per second up and down, so your meter would be a blur of numbers on the screen.
So, instead of taking an average voltage, even cheap multimeters are designed to take the:
Root Mean Square (RMS) of the AC voltage
+170 squared = 28900
-170 squared = 28900
The square root of 28900 = 1.414
170 divided by 1.414 = roughly 120.
The average AC output of a light socket in your home is 0v.
The root mean square of a light socket in your home is 120v.
The "120v" number is more useful and less deadly than the "0V" number would be. Think about it: let's say you have an AC current that was 2 million volts, that would be +2M and -2M which would still average out to be "0v". This is why you sometimes need to plug "170v" into an equation instead of 120v so you don't blow up your circuit.
So, what does this have to do with neon lamp bulbs?
Well, my particular model of neon lamp bulbs need 95v to light. So my bulbs (on AC power) only are lighting at +95v to +170v and -95v to -170v.
The red areas are when the light bulb is on. So, the lamp is only working about 44% of the time. This works out to close to the 50-60% longer useful life when running on AC.
When a high brightness version of a neon lamp ages it: sputters (sprays) metal from the filaments, which darkness the glass until it turns black. This process is constantly going on. I suppose if you had a neon lamp running on DC only one electrode would be sputtering--so once half the bulb's glass was blackened you could just switch the legs of the neon and that would turn on the other electrode, giving you a longer time. However, the electrodes also end up needing higher and higher voltages to light--once a neon bulb running on AC needs about 160v to ignite the bulb's useful life is over because that's too high to reliably ignite. In the meantime the bulb would have: started flickering instead of showing a steady light; and it would have significant blackening from the vaporized electrode metal inside the bulb.
Neon bulbs need ionization already occurring in the neon gas to be able to start. There are two methods in use: one is to put a little bit of radioactive material into the bulb (an isotope of Krypton); or having the bulb exposed to room light! Yep, neon bulbs can't light up in a completely darkened room (or at least not very easily). That is why new bulbs sometimes flicker in a dark room, but if you shine a flashlight on them--or just flick on the room's light to get some regular old light photons ionizing the neon in the bulb. This is called the dark effect.
There are a ton of different things you can use neon bulbs for: oscillators, tone oscillators to make electric organ sounds, instrumentation lights in rugged environments that would cause an LED to fail, a bistable flip-flop circuit for use in very basic computers as and/or/majority gates, ring counters, a sort of radiation sensor very similar to a geiger counter using the lamp's photosensitive (dark effect) properties, triggering and pulsing other circuits, delay circuits, an Is it AC or DC? meter, voltmeter which lights corresponding to various voltage ranges, etc.
You can duct-tape them to a stick and point them near equipment to see if there is RF or high voltage--you can even hold the bulbs near spark plug wires in your car to see if each plug is actually firing! Because it lights up under outside influences (static, ionization, RF, high voltage fields, etc.) you can use it as a touch control of sorts too.
At this point I'm just happy my little neon lamp (and 30k resistor) spurred me on to: make a DC power supply; light one or the other electrode via DC; and lighting both electrode via AC.
Here is a video of the xenon flash hooked up to my 5000v transformer. Like bolts of lightening in a jar.
Here are two schematics, well one is more of a drawing and one is a circuit layout. They go with the photo: 3 blinking neon bulbs.
This circuit is super easy to build. The parts were:When a high brightness version of a neon lamp ages it: sputters (sprays) metal from the filaments, which darkness the glass until it turns black. This process is constantly going on. I suppose if you had a neon lamp running on DC only one electrode would be sputtering--so once half the bulb's glass was blackened you could just switch the legs of the neon and that would turn on the other electrode, giving you a longer time. However, the electrodes also end up needing higher and higher voltages to light--once a neon bulb running on AC needs about 160v to ignite the bulb's useful life is over because that's too high to reliably ignite. In the meantime the bulb would have: started flickering instead of showing a steady light; and it would have significant blackening from the vaporized electrode metal inside the bulb.
Neon bulbs need ionization already occurring in the neon gas to be able to start. There are two methods in use: one is to put a little bit of radioactive material into the bulb (an isotope of Krypton); or having the bulb exposed to room light! Yep, neon bulbs can't light up in a completely darkened room (or at least not very easily). That is why new bulbs sometimes flicker in a dark room, but if you shine a flashlight on them--or just flick on the room's light to get some regular old light photons ionizing the neon in the bulb. This is called the dark effect.
There are a ton of different things you can use neon bulbs for: oscillators, tone oscillators to make electric organ sounds, instrumentation lights in rugged environments that would cause an LED to fail, a bistable flip-flop circuit for use in very basic computers as and/or/majority gates, ring counters, a sort of radiation sensor very similar to a geiger counter using the lamp's photosensitive (dark effect) properties, triggering and pulsing other circuits, delay circuits, an Is it AC or DC? meter, voltmeter which lights corresponding to various voltage ranges, etc.
You can duct-tape them to a stick and point them near equipment to see if there is RF or high voltage--you can even hold the bulbs near spark plug wires in your car to see if each plug is actually firing! Because it lights up under outside influences (static, ionization, RF, high voltage fields, etc.) you can use it as a touch control of sorts too.
At this point I'm just happy my little neon lamp (and 30k resistor) spurred me on to: make a DC power supply; light one or the other electrode via DC; and lighting both electrode via AC.
Here is a video of the xenon flash hooked up to my 5000v transformer. Like bolts of lightening in a jar.
Neon Oscillation Circuit
Here are two schematics, well one is more of a drawing and one is a circuit layout. They go with the photo: 3 blinking neon bulbs.
3 ne-2/a1a neon bulbs
3 1uF ceramic capacitors
3 1M resistors
Not needed but for fun: piezoelectric buzzers.
The blobs of solder were from rearranging things to see what would happen. For that purpose I started crimping the ends of the ceramic capicitors so I could add or subtract them to the circuit without soldering:
Also shown is a piezoelectric buzzer. I used it in place of a capacitor as an experiment. Here are some results:
No capacitors or piezo: the lights turn on and the brightness varies a little as voltage varies.
1 piezoelectric between two of the bulbs: all 3 bulbs blinked quickly. They blinked faster with more voltage.
1 piezoelectric and 1 cap: the bulb with the piezoelectric blinked quickly and the other two bulbs alternated slow blinks. More voltage sped all blinks.
1 piezoelectric and 2 capacitors: two bulbs quickly blinked alternatively and third gave a longer blink (longer time bulb was lighted).
2 capacitors: quick blinks from bulb with no cap, bulbs with cap had longer blinks. No cap bulb blinked more than each capped bulb.
3 capacitors: bulbs blinked at different times. Higher voltage only slightly sped up the blinks.
At no time did the piezoelectric buzzer make a noise, probably a change in resistor values might be needed. Possibly a cap used with a piezoelectric would have made noise.
So what is happening? The capacitor charges up. Due to hysterisis the neon bulb doesn't conduct until the capacitor reaches the bulb's striking (trigger) voltage. Only then will the bulb conduct which lights the bulb and discharges the capacitor. When the capacitor voltage drops, the bulb stops conducting and the capacitor begins charging starting the whole thing all over again. The circuit charges, fires, then relaxes. It's a relaxation oscillator.
Well, more specifically: with just 1 bulb, cap and resistor using DC power this gives out a sawtooth wave. It would be a Pearson-Anson Relaxation Oscillator.
