Sciencemadness Discussion Board - Glow in near vacuum gas ampoules

Glow in near vacuum gas ampoules

woelen - 22-2-2016 at 00:25

I purchased a set of gas ampoules: http://www.ebay.nl/itm/261808224080?_trksid=p2060353.m2749.l...

I want to use them for my element collection. They have a nice size and I intend to put them in special vials with a conductor at both ends, to which I can apply a high frequency high voltage.

I did tests with these ampoules and it is fairly easy to make them glow. What I see though is that I do not get nice pure colors when I make them glow.

The neon tube gives bright orange light, but there also are places where it produces pale blue light.

The other tubes have similar issues, but much stronger. E.g. the nitrogen tube gives a pink/purple light with grey/white spots and pale blue spots at other places. The glow does not look as nice and colorful as on the picture on eBay.

What do I need to do in order to get a nice even glow in these tubes with constant colors?

I had the impression that any high frequency HV AC source of sufficient voltage would do the job, but apparently this is not the case. I tried the ZVS driver I made, but it is too powerful. I get arcs in these ampoules and I hear a strong hissing noise! I also have a small AC high frequency high voltage generator (according to seller: 2200 Vss output when input is 12 volts) with a max power output of a few Watts. This device, however, only is capable of letting the neon tube and helium tube glow.

Any idea of what kind of device I need for making these ampoules glow nicely? If you know of a (not too complicated) schematic which I can solder together, then that also would be very helpful.

[Edited on 22-2-16 by woelen]

j_sum1 - 22-2-2016 at 02:30

Glad you asked this woelen. I have a similar issue. I have ampoules of Kr and Ne and will in time get (or make) others.
I am awaiting arrival of one of these to make them glow. I will let you know how I get on.

(I wwas considering making my own power supply but figured that I would easily spend as much on bits and waste a lot of time.)

woelen - 22-2-2016 at 02:57

Making these ampoules yourself is not easy at all. You have to work at very low pressure and you need to preheat the ampoules in order to be sure that after sealing and drawing vacuum you do not have the issue of the glass gassing out into the vacuum. If e.g. you fill such a thing with 99.99% Xe and the ampoule gases out some N2, O2 and H2O, then your ampoule is spoiled and does not show the pure Xe-spectrum anymore when it is made to glow.

That power supply you show looks strange to me. I only see one electrical pole (the sphere). I'm quite sure that you can get a neon bulb glowing with such a thing (neon is excited easily) and a helium bulb may be borderline. I hope that you can get the krypton tube to glow with this device. To my experience, I need a much higher field strength for all gases, other than Ne and He, and for the others I need two poles, connected to the glass ampoule. I did tests with Al-foil, wrapped a few mm around the tips of the ampoules, with a wire sticking into the Al-foil for good contact. I will make special vials, with foam and Al-foil at both sides, wires sticking in this and then clamped between bottom and top, in such a way that the ampoule can be seen nicely. You also need high frequency AC, the glass does not conduct DC (if it does, it is punctured, due to arcing).

Maybe IrC can jump in, he has a lot of knowledge and experience with HV-electronics, much more than I have.

[Edited on 22-2-16 by woelen]

IrC - 22-2-2016 at 03:58

I had a friend in Phoenix back in the 80's that spent much time in his home lab building gas lasers and I think one of the solutions he devised applies. After getting his vacuum down to under 1 micron he would gently heat the glass tube and tap on it end to end with a plastic tool he created, he used the plastic handle of a small Xcelite screwdriver as the hammer and a fiberglass rod for handle. He only went to this extra bother for his HeNe lasers, not wanting any gas molecules to poison the laser. I never tried building a HeNe laser as in my opinion it is the hardest gas laser to get working well, if at all. But I think the problem of glass holding onto gas impurities applies in your case the same as it did for him. He said the holding the glass under high vacuum, hot combined with intermittent tapping was a mandatory step. To me it seemed the real trouble was all about getting the optical cavity aligned. This I know from spending long hours trying to get them working again after one was misaligned. The people at Meredith Instruments were friends so I had a decent supply of next to zero cost HeNe lasers to experiment with. Anyway as to your impurity issue I think you should try the heating/tapping under high vacuum to your ampoules before filling them. The mini Tesla coil idea of Jsum1 sounds interesting. I would think assuming the output is high enough it would work. A better idea using your end contacts would be build your own adjustable flyback driver. One where you could vary the voltage as well as frequency. Say from 15 to 50 KHZ since depending upon numerous factors you would see modes in the tube and a nice uniform glow is what you desire. The color impurities in my mind can only be contamination, no doubt you can do a better job than the tubes you bought off eBay. Forgot to mention he also repeated the glass cleaning steps by filling with the gas to be used, then pumping back down, repeating the heating/tapping steps before final filling.

j_sum1 - 22-2-2016 at 04:38

Interesting IrC. What exactly does the tapping achieve?
Making the ampoules is probably a project for much further down the track. I have more than enough projects right now. It might be that I short-cut and just buy what I want.

As far as power supplies go, I have little experience with building my own electronic devices and so picked something that would at least give me a starting point. My main concern with what I have chosen is that it lacks power. But I think that having a play with a small one is a sensible precursor to building something larger. I have seen some YT vids. Some of those sparky devices look positively frightening.

I also have a CRT to disassemble and retrieve the flyback. I am probably going to need a tutorial to follow so that I don't do something stupid.

woelen - 22-2-2016 at 04:48

@IrC: I also am not in the process of making my own ampoules, but I could make my own driver with adjustable voltage and frequency. For variable voltage I could use a variable voltage power supply (with ZVS driver you have output more or less proportional to power supply voltage, isn't it?). Variable frequency sounds somewhat more difficult for me. That would require a variable high voltage capacitor or some adaptive inductor. Do you have schematics of low power (fairly safe) variable voltage variable frequency power supplies?

@j_sum1: If you start doing high voltage things, know what you are doing. A mistake with high voltage can easily be fatal. To my opinion, HV is more dangerous than nearly all home chemistry. A single touch at the wrong place can kill you. I never feel comfortable when working with HV, e.g. working with the ZVS-driver I made always makes me somewhat scary that I accidently touch something which I should not touch. For this reason I want a low power device for my gas ampoules (I think that a few hundreds of mW is enough for making the gas glow).

j_sum1 - 22-2-2016 at 04:53

Ok. So 1 Watt might be enough.

Good to know.

woelen - 22-2-2016 at 06:38

Yes, glowing a small cm-sized gas ampoule only takes a small power. What is more important is field strength and having sufficiently high frequency. The glass of the ampoule works as isolator, and hence DC and very low frequency AC cannot pass through the glass. For high frequencies, however, the glass allows a current to flow, positive charge builds up on one side of the glass and negative charge builds up on the other side of the glass (it works like a capacitor) and during the AC-cycle the charge buildup, on each side of the glass, alternates. The net effect is an alternating current on each side of the glass, while inside the glass there is no current. The higher the frequency (at least up to several 100 kHz, above that other effects may kick in), the less the glass is a barrier for the current. The standard "high frequency" high voltage power supplies, like your plasma globe, and my ZVS driver, operate at something in the tens of kHz, maybe 20 kHz, but it can also be 10 kHz, or 40 kHz, but this is the order of magnitude for the frequency.

IrC - 22-2-2016 at 15:48

I assume he was creating vibrations in the glass walls that he thought would help impart motion to anything loosely bound to them but that is just my guess. I was relating what I watched him do while hanging out experimenting with lasers he built. If there is scientific validity to the idea I think I would use ultrasonics.

http://uzzors2k.4hv.org/index.php?page=flybacktransformerdri...

I cannot think of anything simpler or cheaper than a 555 flyback driver where a pot was used to adjust frequency. The change I would make to the examples in the link is to feed the flyback primary with an adjustable DC voltage. This would allow the power level to be adjusted. A ZVS circuit is not a good idea in this application for multiple reasons. The result would be dangerous and difficult to adjust. Better to have a supply that could bite but not actually hurt you.

I drew up a quick circuit and scanned it to give you an idea of what I was thinking on the flyback driver. Cheap, low power and pretty safe. The fet can be what you have handy if its ratings are adequate. I did not do a lot of math on the circuit so no doubt there is room for improvement. At low power levels the snubber may not be needed although many people lose fets with these circuits by not having a snubber circuit. I did not do any calculations here either just added what I have used before. The PDF is a good read on the subject if you wish to improve the protection for the mosfet.

Attachment: AN-4147.pdf (152kB)
This file has been downloaded 473 times

"Ok. So 1 Watt might be enough."

Actually with efficient power transfer I think one or two tenths of a watt would light something the size of these ampoules.

Just realized I forgot to label the regulator, should be along the lines of a 2N3055 or TIP3055 and should have some sort of heat sinking.

[Edited on 2-22-2016 by IrC]

woelen - 22-2-2016 at 15:53

I have a working ZVS circuit lying here (based on the IRFP264 transistors you suggested me to use one and a half year ago). If I put a 20 kV/ 10 MOhm resistor in series with the output of the transformer, what would you expect with the gas tubes? Would it be "safe"? In such a setup I can use the ZVS driver as is for other more powerful experiments, and for lighting my bulbs. By adjusting the input voltage between 12 V and 18 V I can adjust the output voltage, while staying well inside the range allowable for the ZVS driver. What is your opinion on that?

I read quite a lot of negative comments on the 555-drivers. They seem to be unreliable and easily break down if I am to believe diverse comments on high voltage forums. They need very good cooling and the 555 can easily be blown out by capacitive coupling between the high voltage on the flyback transformer primary side and the gate of the MOSFET. Is this really true? I can imagine that adding a zener diode can protect against this effect.

IrC - 22-2-2016 at 16:20

Too many unknowns to answer your ZVS/high ohm resistor question but I firmly believe for what you are doing here it is overkill, too powerful, and unneeded danger for a project requiring such low energy input. As to would it work, obviously so well it could melt the tube if pushed but control over parameters would be limited and harder to implement. Certainly not as convenient as dialing two pots to the right setting for each type tube desired.

No doubt the many 555 complaints stem from poorly skilled in the art experimenters. I would bet none of them even considered snubber circuits for the flyback and this would have greatly reduced bad transients hitting both the fet and the 555. Trying to reach high drive levels and thus severe inductive kickback which were better suited for a ZVS or other power circuit. Yes you could blow the fet or 555 if you crank the circuit I posted way up but the point is at the low energy required for this project the circuit shown is cheap, safe, reliable. If you study the link I gave the 2nd 555 circuit used a complementary pair of 2N3904-06 transistors to help isolate the 555. You could take that snippet out and add it to the circuit I drew, or use their circuit and just rob the adjustable DC source to the flyback primary in my circuit. I have used circuits similar to the one I drew today trouble free for years to power gas lamps larger than your ampoules without blowing things but as I said I was not going for high power. For that I would use something like the ZVS unless adjustable frequency was also desired. If so I would design a circuit around something like the TL494. However for your needs in this thread I fail to see the need for high power.

I should add I assume your ZVS question is due to you already having it and the high value resistor on hand and want to try it out. Do so but start low on voltage, keep in mind below 10 volts that eBay ZVS board (or the similar circuit we have been building and using in the ZVS thread) may not start reliably resulting in overheating fets so pay attention to input current draw (and lack of output high voltage). Not knowing the flyback voltage nor the gas lamp characteristics I am not sure how to answer your resistor question. If I had to pull numbers out of a hat I would guess 80 volts more or less, 0.3 to 2 ma, Considering a gas tube of relatively small size with terminals, as with a Neon lamp brightly lit (no radioactive source inside the lamp). Other gasses no doubt will vary widely and of course your tubes contain more gas than my Neon lamp example. Also obviously things will vary with no connection through the glass, as well differences exist with the mini Tesla coil approach. You will have to experiment to find the right values for each tube. Using the circuit I drew you can merely start at the bottom on the voltage adjust with frequency set at say 17 KHZ, slowly bringing voltage up to where the tube lights properly. The flyback is likely going to have peak efficiency around 15 to 20 KHZ. Next step experimenting with frequency and voltage adjustments to get a uniform glow. Make a scale for both pots and record the settings for each type gas tube. Possibly you may find a single frequency which works well for all tubes and can replace that control with fixed values leaving you with a single voltage adjust control with a best setting for each type gas you could record. That way when you wish to try each different gas you can preset the level for it. I imagine variations in how you make each ampoule would also affect the setting. I was trying to keep the duty cycle near 50 percent but this may not be the best value for what its worth.

[Edited on 2-23-2016 by IrC]

woelen - 22-2-2016 at 23:57

Thanks for your clear answer. I indeed asked my question on using the ZVS driver, because I already have it and I have a big resistor 10 MOhm / 20 kV (its almost 10 cm long!).

You convinced me, however, that building the simple 555 circuit at the top of your link is the best option. I will not use the ZVS circuit for these lamps, I value my life and I also do not want to destroy these beautiful ampoules.

Indeed, all the horror stories about blown MOSFETs and even exploding 555 chips are from people who want to squeeze maximum power from the little circuit.

I will build the first 555 circuit from your link and I will do some research on snubbers in order to make it even more reliable. I still have three IRFP264s left over from my ZVS project, so I will put one of them to good use in that simple 555 circuit.

Thanks again for your valuable input. If I have results, then I certainly will post them here.

j_sum1 - 28-2-2016 at 04:46

Quote: Originally posted by woelen

That power supply you show looks strange to me. I only see one electrical pole (the sphere). I'm quite sure that you can get a neon bulb glowing with such a thing (neon is excited easily) and a helium bulb may be borderline. I hope that you can get the krypton tube to glow with this device. To my experience, I need a much higher field strength for all gases, other than Ne and He, and for the others I need two poles, connected to the glass ampoule. [Edited on 22-2-16 by woelen]

Device arrived. It gets little neon bulbs to glow. Ditto a little argon(?) bulb that came with it. It lights up compact fluorescents rather well which is quite impressive.
It didn't do anything visible with the gas ampoules I have. I guess I am back looking for another solution.

woelen - 28-2-2016 at 07:33

Actually what you describe is what I expected. From a small USB-device I do not expect sufficient field strength to get the ampoules glowing. The main issue is that the glass of the ampoules is a strong barrier. You need a flyback circuit like IrC posted, with some snubber circuit added to assure that the transistor is not destroyed easily.

careysub - 6-6-2016 at 08:17

I think the advice about being very cautious working with HV is good, and indeed I would much prefer to use ready-made units, like neon-sign transformer or a sufficiently powerful Tesla coil.

In the Tesla coil category vacuum tester probes look interesting, they produce up to 50 KV at 500 KHZ, 40 watts and are available from $150 (used) to $250 (new). I might get one to try.

It appears that modern NSTs have an output frequency in the low 10s of kilohertz range, though not all of them specify an output frequency I assume all modern switching based ones have similar output frequencies.

This listing:
http://www.ebay.com/itm/Complete-Set-of-noble-gases-sealed-i...

is interesting, since it shows the ampoules he provides glowing with a 2 KV NST, although he asserts that his ampoules are specially made to glow.

NSTs with output voltages of up to 12 KV are easily found on eBay and elsewhere.

Since NSTs are relatively inexpensive and would be straightforward to wire up a display, using suitable high voltage wiring obviously, I am considering experimenting with using one to illuminate gas ampoules off of eBay, there are some quite inexpensive ones.

Any advice about what voltage I should try and any other safety advice?

The 2 KV seemed low to me, I was considering trying 9 KV. There are also "self adjusting" transformers with outputs listed as (for example) 2 KV - 12 KV. Do they vary the voltage based on current drawn?

[Edited on 6-6-2016 by careysub]

Heavy Walter - 6-6-2016 at 09:12

Fellows
There are two aspects about those ampoules.
Neat colors are related mostly with gas purity. As woelen and IrC stated, ampoules need to be well evacuated, with degassing procedures and filled with spectroscopic quality gases.
Glow discharge for those electrodeless ampoules depends on HV and HF and pressure filling. I doubt you will be able to ignite them with a NST.
Vintage "violet" electric massage/therapy devices are a good power source for igniting low pressure ampoules.

careysub - 6-6-2016 at 13:57

Quote: Originally posted by Heavy Walter

...
Glow discharge for those electrodeless ampoules depends on HV and HF and pressure filling. I doubt you will be able to ignite them with a NST.
Vintage "violet" electric massage/therapy devices are a good power source for igniting low pressure ampoules.

The properties of the "violet ray" devices (high voltage, high frequency, low current) sound like the operating properties of Tesla coils. Sounds like those Tesla coil vacuum testers are the best bet.

I guess I believe the "the-element-shop" guy on eBay when he states that his gas tubes are specially prepared to work with an NST (especially a low voltage like 2 KV).

woelen - 6-6-2016 at 14:31

Quote: Originally posted by careysub

This listing:
http://www.ebay.com/itm/Complete-Set-of-noble-gases-sealed-i...

Interesting link, but only for people outside of the EU

This seller does not sell anything to EU-countries, although he himself is located in the EU.

I purchased one of these a few months ago, besides the small ones I already had purchased earlier:
http://www.ebay.nl/itm/Complete-Set-of-noble-gases-sealed-in...
http://www.ebay.nl/itm/Complete-Set-of-Noble-Gases-Sealed-in...

They are twice as big as the ones from the other link, but they are MUCH harder to ignite. The glass is thick. This makes the ampoules robust and good-looking, but the downside is that making the glass glow requires much higher voltage. I only managed to light the Ar, Kr and Xe tubes with my ZVS-driver at 20 volts input, which probably is well over 10 kV output. The Ne-tube and He-tube are easier to light, but still require the use of my high power ZVS-driver.
The tubes do not take much power from the ZVS-driver, they only require high voltage, not much power.

[Edited on 6-6-16 by woelen]

careysub - 6-6-2016 at 15:27

Quote: Originally posted by woelen

Quote: Originally posted by careysub

This listing:
http://www.ebay.com/itm/Complete-Set-of-noble-gases-sealed-i...

Interesting link, but only for people outside of the EU

This seller does not sell anything to EU-countries, although he himself is located in the EU.

There are ways around this. EU people can U2U me if interested.

[Edited on 6-6-2016 by careysub]

careysub - 6-6-2016 at 15:37

Quote: Originally posted by woelen

...

I purchased one of these a few months ago, besides the small ones I already had purchased earlier:
http://www.ebay.nl/itm/Complete-Set-of-noble-gases-sealed-in...
http://www.ebay.nl/itm/Complete-Set-of-Noble-Gases-Sealed-in...

They are twice as big as the ones from the other link, but they are MUCH harder to ignite. The glass is thick. This makes the ampoules robust and good-looking, but the downside is that making the glass glow requires much higher voltage. I only managed to light the Ar, Kr and Xe tubes with my ZVS-driver at 20 volts input, which probably is well over 10 kV output. The Ne-tube and He-tube are easier to light, but still require the use of my high power ZVS-driver.
The tubes do not take much power from the ZVS-driver, they only require high voltage, not much power.

Looking at the image of the argon tube here:
http://www.ebay.com/itm/Argon-gas-ampoule-purity-99-999-elem...

It looks to me like they are lighting it with a Tesla coil, notice the single electrode and the spark current jumping to the tube.

I will procure one along with some Chinese gas tubes and run some tests. Stay tuned.

woelen - 6-6-2016 at 22:53

I can recommend this yaolihong2013 seller. She sells good materials and when something is wrong (I once had an ampoule with 2 grams of Li from her, which had a small crack), then you get a replacement without any hassle.

She sells many other interesting things. Most noticeably, 25 grams of very pure and shiny lithium metal for a very good price, and other element samples in ampoules.

careysub - 7-6-2016 at 06:31

I went with oneworldonedream who I have done business with before. It looks like the same product.

j_sum1 - 7-6-2016 at 23:03

I have nearly a full set of these that I have bought one at a time. (I don't have Xe yet.) I have not yet got a power source that works.

For something that looks like an empty glass tube they actually look quite nice.

woelen - 9-6-2016 at 03:59

I now found a circuit, based on a simple astable oscillator and using TIP2955 and TIP3055 as driver transistors:

http://tmms.co.jp/EFD/Lab_&_experiments/Corona_Lab/Power...

I will use BDX53 and BDX54 for these, I just ordered two sets of these. This circuit allows limitation of current to tens of uA.

I will replace the astable oscillator with a NE555 circuit, the driver part will use BDX53/54. I have a good flyback transformer lying around, which only has a secondary. I wind my own primary.

This should be the final answer to my quest. Relatively safe, low power, and robust.

careysub - 9-6-2016 at 05:57

I see quite a number of new-old stock TV fly-back transformers on eBay. There are no specs on any of them.

Are they good choices for a project like this?

[Edited on 9-6-2016 by careysub]

woelen - 9-6-2016 at 12:04

Most flybacks are not suitable. You need one without built-in rectifier. I do not do the rectification step from the web-page I linked to, I use the AC-output. Glowing discharge tubes made entirely of glass without electrodes only is possible with high frequency AC.

I purchased a transformer with only a secondary winding (2400 turns) and then I can make my own primary of appr. 10 turns.

careysub - 10-6-2016 at 16:59

Looking for a semi-OTS solution for lighting up the gas tubes in an element display, perhaps all at once (but not necessarily).

Analog HV electronics are not my forte, and I would rather not try to build a solution, and am not well equipped to evaluate offerings. Safety of any solution especially concerns me, making me wary of trying to implement my own solution.

I received the industrial Tesla coil and it is very nice, 6 cm sparks! This is a form of AC I understand, so perhaps I could use this to energize tubes with external electrodes by having one going to ground (and using suitably high breakdown wiring)? Still waiting on my gas tubes.

Current limiting an NST with a resistor looks like a non-starter since the output power is so high (200-300 watts).

Alternatively, this site has a lot of HV products of various kinds:
http://www.amazing1.com/hv-hf-power-supplies.html
http://www.amazing1.com/transformers-high-voltage-high-frequ...
but I am unable to tell if there is anything usable for the desired application. To the extent that they describe the output of most of their devices, it seems to be DC. The have "easily disassembled" fly-back transformers, which sounds promising, presumably if they have rectifiers they can be removed.

[Edited on 11-6-2016 by careysub]

woelen - 13-6-2016 at 10:32

NSTs already are current limited by design. They are specified with open loop voltage and with short circuit current. The heavier the load, the lower the voltage. This behavior is beneficial for gas discharge loads, because initially, when there is no current flowing, the lamps need a high voltage to ignite, and once they ignite, their current tends to grow without limit, but the transformer limits the current to something less than the short circuit current.

The site amazing1.com has quite some interesting power supplies. The PVM12 is an AC power supply with adjustable output frequency and adjustable output voltage. It looks nice. For me it is too expensive though, because it also must be shipped to my country and I'll have to pay 6% taxes plus EUR 20 or so declaration fee.
The GRADRIV10 also looks like an AC output device.

Twospoons - 13-6-2016 at 15:19

The power sources used by ozone generators would probably work quite well.
http://www.aliexpress.com/af/ozone.html?ltype=wholesale&...
These all operate using a high voltage high frequency dielectric barrier discharge. Specs are a bit sparse though.

careysub - 13-6-2016 at 17:04

This one:

http://www.amazing1.com/products/plasma-power-generator-7kv-...

for $40 might be right-sized for the element display, 7 KV at 10 mA.

A $40 voltage multiplier add-on is available to boost the output to 35 KV at 2 mA if needed (probably not).

Tubes to be driven: H, D, He, N, O, Ne, Ar, Kr, Xe; possibly Cl, Br and I. So up to 12 tubes.

j_sum1 - 14-6-2016 at 17:37

Quote: Originally posted by careysub

This one:

http://www.amazing1.com/products/plasma-power-generator-7kv-...

for $40 might be right-sized for the element display, 7 KV at 10 mA.

A $40 voltage multiplier add-on is available to boost the output to 35 KV at 2 mA if needed (probably not).

Tubes to be driven: H, D, He, N, O, Ne, Ar, Kr, Xe; possibly Cl, Br and I. So up to 12 tubes.

As far as i can tell, this one is similar at a better price.
http://www.ebay.com.au/itm/401122166917

If someone can confirm that it is likely to work then I will grab one or two. I am much more comfortable with buying something than attempting to construct my own.

Twospoons - 14-6-2016 at 17:56

There's no indication if the output is DC or AC. From the description of the purpose (ignition) I'd guess its DC, from an internal diode multiplier, which is not what you want.

[Edited on 15-6-2016 by Twospoons]

j_sum1 - 14-6-2016 at 18:15

Ah. Thanks for that compatriot.

woelen - 14-6-2016 at 23:26

Quote: Originally posted by j_sum1

As far as i can tell, this one is similar at a better price.
http://www.ebay.com.au/itm/401122166917

This is a pulse generator. It generates sparks, once per second or so. These sparks can be used to ignite something, e.g. gas in a burner. I myself have such a thing, it produces 40 kV sparks with a 6 mJ energy content. It charges a small high voltage capacitor (7 to 8 pF) until a spark occurs.

careysub - 18-6-2016 at 16:05

Results of testing my Chinese set of noble gas tubes plus hydrogen with commercial Tesla coil vacuum tester (an Electro-Technic Products BD-10), which has a variable voltage from 20 to 50 KV at a frequency of 500 KHz and a maximum current of 0.1 mA.

I tested them by just touching the coil electrode to the tube, and by also grounding one end with an 8 foot length of 100 KV wire.

They all produced some visible glow in a well lit interior room, the neon was far and away the brightest, faint for hydrogen. In a fairly dark room they all glowed quite visibly, though some much brighter than others.

Grounding one end while touching the electrode to the other produced the strongest, most even, display. But it was only modestly better than ungrounded.

20 KV lit up the tubes nicely. 50 KV produced much stronger arcing, and a slight increase in output, but this voltage is excessive since arcing is an undesirable effect.

I was able to clearly light up the four easiest tubes stacked side by side all at once (neon, krypton, xenon, and helium) with 20 KV.

Trying to transfer the current from the electrode to the gas tube over the 8 foot wire produced a feeble effect, only slightly enhanced by grounding one end. This is more or less what I expected since a Tesla coil is an odd-ball AC source, without a regular neutral/ground line.

The xenon and krypton tubes produce brightly glowing conducting channels, not just a glow, unlike the others.

My conclusions.
A Tesla coil of 20 KV is fine for displaying an electrode-less glass gas tube by electrode contact, and even energize multiple tubes in physical contact if the coil is strong enough. It cannot be used to light up a display of mounted tubes via wiring.

All of the gas tubes will put on a useful glow if mounted in a dark recess to maximize the effect.

My recommendation for powering a mounted display (to be tested, once I get the power supply) is a 15-20 KV high frequency AC supply, with grounding. Since I got a good effect at 20 KV, I think it is safe to say that 15 KV will do well also.

0.1 mA is plenty of current for one tube (although the dimmer tubes might benefit from a higher current?). For a multitube display, lit all at once, 1-2 mA should be fine.

14 KV neon sign transformers are readily available, but put out 20 mA, which is about ten times more current than needed for this.

I think making a tube holder for each tube out of a copper wire coil on each end (which are the electrodes) is a good display design.

I think the MINIMAX40 (NEONX80), $27, and the MULTI50 voltage multiplier, $40, combination from the source below which together produce 20 KV at 1 mA (25 to 35 KHz) is about right for this:

http://www.amazing1.com/hv-hf-power-supplies.html

I am going to order one of these and set up a display prototype.

[Edited on 19-6-2016 by careysub]

Dan Vizine - 18-2-2017 at 08:43

It seems this very interesting thread just kind of dried up. Anybody have any later results which might indicate the most cost effective power source to illuminate gas tubes such as those sold by yaolihong2013 on eBay?

j_sum1 - 18-2-2017 at 15:53

Nope. Not from me. This is one of my many shelved projects. But in the interim I have collected nearly a full set of inert gases in nice ampoules from yaolihong2013. Just Xe to go.

Dan Vizine - 21-2-2017 at 13:03

I also purchased the set except argon. A 2.5 kV transformer with wires wrapped around the tube did nothing, as expected.

Too bad about the lack of results, I was hoping to ride on your and the other's coattails!

[Edited on 2/21/2017 by Dan Vizine]

woelen - 22-2-2017 at 13:16

No, 2.5 kV is too low for such thick-walled ampoules. You need at least 10 kV or so for decent light output.

Dan Vizine - 22-2-2017 at 15:29

With 10 kV @ 50 or 60 Hz (a reasonable neon sign transformer), is it your experience that a coil at top and another wrapped around the other end will give a reasonable glow?

The 2.5 kV transformer (from a junked hplc to power the UV detector) makes a little neon lamp glow like a red LED! No surprise, this little negative resistance trinket heats up fast.

[Edited on 2/22/2017 by Dan Vizine]

wg48 - 22-2-2017 at 15:57

Wrapping wire round the end of the ampoules creates a capacitance to the gas of a round 20pF and there is two of them in series. At 50/60Hz that is an impedance of about 320 x 10^6 ohms at 10kV that will produce a current of only about 35uA which will not be very bright.

To get more current into the gas and hence brighter you need a much higher frequency 20kHz and up to MHz.

A mains HV transformer is not the brightest option LOL

You need something like this

http://www.ebay.co.uk/itm/DC-High-Voltage-Generator-Inverter...

and here is a 500kV one (an example gross inaccurcy but it would mean you could claim your money back)

http://www.ebay.co.uk/itm/DC-6v-12V-to-500Kv-High-Voltage-Co...

[Edited on 23-2-2017 by wg48]

[Edited on 23-2-2017 by wg48]

Dan Vizine - 23-2-2017 at 07:13

Thanks wg48,

I had absorbed just enough of what was said earlier to be hesitant to think 60 Hz would be good enough.

Dan

wg48 - 23-2-2017 at 09:20

Dan:

I should also add that you need a higher voltage (about X2) to start the gas conducting (breakdown) than to maintain it glowing. The actual voltage is dependent on the gas, its pressure and the distance between the electrodes. You may not have the optimum gas pressure. From memory it’s a round a few torr of pressure.

There are things you can do that will increase the chances of getting the glow going and making it brighter. Increase the capacitance of the connections, say by silver plating the tube ends to increase the surface area. Perhaps put a conductive strip on each side of the tube to decrease the separation. Of cause of a white or reflective background may make it look bright or a black background if the ambient light level is higher than the glow. I don’t know the output impedance of your mains transformer but its probably much lower than the impedance of the tube so the tubes should be connected in parallel.

You could also use your 10kV transformer to build a small Tesla coil. That will make your tube/s glow.

Dan Vizine - 24-2-2017 at 07:33

Wg48,

Thanks.

I don't actually see anyone who just outright states "I have yaolihong2013's sample tubes and here's how I made them glow". Maybe I'm reading too much into that, but I'm not as sanguine about the prospects of success as I once was.

j_sum1 - 2-3-2017 at 12:41

Here's one.
https://www.youtube.com/watch?v=SD07uTBlmJk

But few details given.

O.K here's how i did it...

NeonPulse - 4-3-2017 at 01:42

Quote: Originally posted by Dan Vizine

Wg48,

Thanks.

I don't actually see anyone who just outright states "I have yaolihong2013's sample tubes and here's how I made them glow". Maybe I'm reading too much into that, but I'm not as sanguine about the prospects of success as I once was.

I actually have the gas tubes from this seller and actually got the neon one and argon ones to glow. the neon one obviously glowed very nicely orange as expected but the argon one is the only other one i actually got a result from.
I didn't think my experiment would even work but was pleased when it did.
The power for this was one of these cheap discharge capacitors:
http://www.ebay.com.au/itm/ultra-high-1000KV-High-Voltage-Pu...

i attached a 18650 battery holder and a momentary power switch to it and when it is powered up it fires quite loudly depending on the distance between the end wires.
This is what i used to make the Neon Pulse..... since keeping it on for too long would probably fry the capacitor i had to make it pulse instead.

I simply attached the end contact wires to either side of the Neon gas tube and pressed the button. the brightness of the glowing was dependent on where the wires were placed on the tube but offsetting them either side gave the best results.
I did try with the Oxygen, helium and krypton tubes with a barely perceptible glow coming from the helium tube. the best results were the Ne and Ar tubes.

j_sum1 - 4-3-2017 at 02:22

Thanks for that NP.
I tried the same procedure with a 400kV unit without success. Somehow I missed the 1000kV ones. Just how grunty do these things get? I haven't seen anything bigger than 1000kV in my searches. It would be nice to have them glow along the entire length. It would be nice to get the other gases to gow. And it would be really good to be able to leave them switched on for a while.