Using NE2 bulbs as voltage regulators

Neons wre used as volt reg references in the valve days. Just beware that V is far higher until it strikes, so your following stage should ramp up. An d keep the neon in the dark to avoid erratics. And keep current down to avo id failure. And smooth their output well, they get get a bit messy years do wn the line.

NT

uy

eal

V is far higher until it strikes, so your following stage should ramp up. And keep the neon in the dark to avoid erratics. And keep current down to a void failure. And smooth their output well, they get get a bit messy years down the line.

Voltage overshoot could maybe be reduced wth high R resistors from V_in_+ t o various points in the neon chain.

NT

Apparently neon bulbs need 90V to start conducting and settle to about 60V at their rated current. Six or seven of them in series would give you 360V or 420V. You'd need better than 630V to drive a seven lamp string into cond uction, and your series resistor would be dropping 210V after they'd turned on. 100uA should be enough current to keep them glowing, so the resistor w ould have to dissipate at least 20mW, which should not be a problem. Gettin g the extra 200V might be a pest.

Zener diodes don't need extra voltage to get them conducting, but you do ne ed enough voltage headroom to cope with their temperature dependence - abou t 0.13% per degree Kelvin - about 5% over the 40C "room temperature range" or 21V in your 400V, on top of the +/-5% tolerance - another 20V - on the z ener voltage. The data sheets typically use 5mA tests currents (which is hi gh - 10W - for a 200V part). The 3W Motorola 1N5956B

at least specifies the zener impedance at 0.25mA and shows a plot of voltag e versus current.

You'd still need about 20V headroom above the worst case 440V for a simple series resistor based shunt regulator. You could do better with a PNP const ant current source to provide a constant current through the Zener string, which would work fine with only only a few volts of extra headroom.

Ever see the designof alotof 1990s DC regulators ? they use a Zener with a voltsage divided down reference.

I you want to stick with tubes, a constant current source is not all that hard. you could make one neon cancel out the other actually. From there it is a simple feedback circuit. I could do it, IF I understand what you want properly.

One thing though, I owuld suggest a lead cage for the neons and any other sensitive tubes. you might geet away with steel actually if you don't go digging around Fukushima, but it depends on how much accuracy you want.

The last time I tried to use a neon lamp as a voltage regulator, I managed to build to perfectly functional relaxation oscillator instead. Try googling for a GM power supply schematic: See any that use neon lamp regulators? That should be a clue.

My antique CDV-900 counter uses a 900v regulator tube made by Victoreen. There's a difference between these and neon lamps, but I don't recall the details. The designer of these replacements is someone that posts regularly to this newsgroup.

Sure, but why? Where are you generating the voltage from in the first place?

If it's from a big iron transformer and rectifier, or a hand-cranked leather and glass apparatus... it's probably a good idea to regulate (or at least filter heavily)!

If you're starting with battery power, you're much better off using an SMPS, and throttling that. Don't fight a dirty supply, bend it to your will and make it work for you.

As for the bulb themselves, they're kind of noisy, and sure to drift (especially over age), but as mentioned, they have been used from time to time, and given their operating envelope, do about as well as you'd expect.

There were some varieties that were specially selected or formulated or aged, if you wanted something smaller than a proper 0A3 or 0G3*. I'm guessing this isn't important in this case, as your pile of junk-box bulbs probably isn't the fancy kind.

Tim

-- Seven Transistor Labs Electrical Engineering Consultation Website:

All of these schemes mentioned are rather noisy; unless one uses a large current thru those zeners,they will be very noisy (look like they are oscillating). If you can afford it,how about a regulator that is quiet from 1uA to

Yes; i just mentioned that the Codatron(R) is very quiet and available in 50V steps from 50V.

Why bother shielding the NE-2s? A bit of radiation or UVA will improve the turn-on performance, and is pretty unlikely to be noticed in operation.

Cheers

Phil Hobbs

On Sunday, August 9, 2015 at 12:49:07 AM UTC-4, Geoffrey S. Mendelson wrote :

Many many years ago when I was in the Navy, the ship I was on had a radar r epeater that used a NE-2 neon bulb as a voltage regulator. The neon bulb failed fairly often, maybe about every three months. Often enough that I c ould remember that it had failed before. So I bought a NE-2 from Radio Sha ck and used it instead of one from the Navy supply system. And it did not fail again while I was on that ship. Just pointing out that old stock may n ot be reliable.

I would try to come up with a circuit that used one neon bulb across part of a voltage divider to do the regulating. I think trying to use 5 or 6 in series would lead to very erratic behaviour.

Dan

Indeed, I think some glow tubes were manufactured back in the day with a small amount of radioactive material to encourage ionization in dark environments.

Since you indicate they are "with resistors" (I presume built-in) they won't act as regulators... the resistor drop will ruin the regulation. ...Jim Thompson

I found a GE Glow Lamp Manual 2nd edition 1966 on my bookshelf. I was waiting for neon lamps to make a comeback. Chapter 4 includes using glow lamps for voltage regulation: If you search through the catalog pages at the end for "radioactive", you will find which lamps had something radioactive added to reduce dark effect. I couldn't find any documents listing what radioactive materials were added to neon lamps. If it was something like Krypton-86 (10.8 years) or Tritium-3 (12.3 years), the radioactivity is probably gone by now. Beware of old stock.

Don't know about those gases but at least one model used an isotope of nickel, it was loaded into a paint a dab of which was inside the glass envelope.

piglet

They make excellent lightning detectors :)

The GE document specifically says that the "mild radioactive" additive is added to the neon gas.

Neon John said (back in 2011) with reference to fl. starters and enon lamps.

"A long time ago a touch of Thorium oxide was added, the purpose being for the radioactivity to establish a uniform ionization environment and thus stabilize the strike voltage. As the Cold War progressed and the nation had Kr-85 running out its ears, that isotope was substituted for the Thorium. Easier to inject a little gas than to mix the thorium oxide in with the electrode material.

The same protocol goes for neon lamps. The only minor problem is that Kr-85 has a 10.7 year half life and so if the equipment at hand is very old, the starter or neon lamp may not be reliable."

--sp

If struck directly? Sure. So does a tree or, hopefully, you.

Thanks. I recall the discussion but was too lazy to search for it.

Neon lamps probably use the same radioactive isotopes as a fluorescent starter. Some possibles are listed in: Half-Life Emits Kypton-85 10.8 years Beta Promethium-147 2.6 Beta Tritium H3 12.3 Beta Thorium-232 14 billion years Alpha I can see why they used Thorium. Promethium seems too short. Krypton and Tritium are longer but I would still be worried about the usability of "old stock" neon lamps.

Back around 1961, while a student at MIT, I had a bad experience...

I had a tube amplifier which used 0A2's as regulators.

I located the amplifier on a shelf in the attic (I had a remote preamplifier... Germanium transistors, no less :-) and it was left on all the time, no problems.

Summer of '61, decided to vacation back home in WV, so I shut off amplifier power.

On my return, a month later, I flipped the power switch... KABOOM! A number of electrolytic cans blew out their bottom seals... over-voltage :-( ...Jim Thompson