Thoughts on nixie lifespan and halogens...

[Nick]

The lifespan of a nixie is not a precise science - end of life may be considered as when luminosity drops by 50% (Weston), sputtering destroys a cathode, cathode poisoning renders a glyph unreadable (though this may be reversible), mechanical damage etc.

I was wondering about the luminosity and sputtering issue. In my experience, nixies rarely run at over 35 - 40C unless heavily over-driven. Would the introduction of a small amount of a halogen, probably chlorine in this case (maybe iodine?), allow a low-temperature halogen cycle to re-deposit any evaporated cathode? I'm well aware that mercury (Hg) is introduced for a similar reason, but a halogen may be safer (in today's H&S climate) if it works at all...

I'm not a physical/inorganic chemist, so thoughts welcome....

Nick

[Joseph Bento]

My bedside nixie clock was built with six IN-14's.  It currently has 44,562 hours of operation, and the nixies show no sign of dimming.  That time is low long the nixies have been lit.  I have the clock set to blank the display from midnight to 6am daily.

 

Joe

[Nick]

On Wednesday, 2 October 2013 03:53:11 UTC+1, Joseph Bento wrote:

My bedside nixie clock was built with six IN-14's.  It currently has 44,562 hours of operation, and the nixies show no sign of dimming.  That time is low long the nixies have been lit.  I have the clock set to blank the display from midnight to 6am daily.

Old nixies are not really relevant - obviously existing nixies (mostly) have Hg doping, but for new nixies being made now, its a potential issue - there is no exemption under the 2006 RoHS directive for nixies! Hg is one of the compounds explicitly tested for... ...

Nick

[Tidak Ada]

Halogens aren't useful in nixies, because of the working temperature is much to low.

It is not for nothing halogen incandescent bulbs use small bulbs of quartz glass to obtain a high temperature of the bulb!

 

eric

---

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Nick
Sent: woensdag 2 oktober 2013 10:22
To: neoni...@googlegroups.com
Subject: [neonixie-l] Re: Thoughts on nixie lifespan and halogens...

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send an email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/44f7fc31-d0b2-4a09-b369-783aaa8fa183%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Nick]

On Wednesday, 2 October 2013 09:25:37 UTC+1, Tidak Ada wrote:

Halogens aren't useful in nixies, because of the working temperature is much to low.

It is not for nothing halogen incandescent bulbs use small bulbs of quartz glass to obtain a high temperature of the bulb!

My understanding was that the quartz glass is not part of the halogen cycle - its only used because the presence of halogens allows the filament to run at a higher temperature and the bulb to have a higher pressure thus giving a whiter, more intense, light  - because of this and the smaller bulb, a higher melting point glass is needed. The fact that quartz glass is used is incidental.

Nick

[Dalibor Farný]

Thanks for info Nick, that is really pretty bad.. The RoHS directive
says that mercury in a product must be in a ration of less than a 0.1%
by weight (that is OK), and it cant be intentionally added - this is
the problem. But I found some information from 2011 that neon signs
are not limited by this directive, do You know any closer information?

That would be serious obstacle in producing real nixie tubes..

Thanks,

Dalibor

2013/10/2 Nick <ni...@desmith.net>:

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+...@googlegroups.com.
> To post to this group, send an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/44f7fc31-d0b2-4a09-b369-783aaa8fa183%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

--
Dalibor Farny
http://dalibor.farny.cz

[Dalibor Farný]

This is written in the directive:

-------------
Applications of lead, mercury, cadmium and hexavalent chromium, which
are exempted from the requirements
of Article 4(1)

1. Mercury in compact fluorescent lamps not exceeding 5 mg per lamp.
2. Mercury in straight fluorescent lamps for general purposes not exceeding:
— halophosphate 10 mg
— triphosphate with normal lifetime 5 mg
— triphosphate with long lifetime 8 mg.
3. Mercury in straight fluorescent lamps for special purposes.
4. Mercury in other lamps not specifically mentioned in this Annex
......
.....
...
--------------------------------------------

I guess that nixie tube could be considered as "other lamps not
specifically mentioned in this Annex" (number 4), what so You think
guys?

Dalibor

2013/10/2 Dalibor Farný <dal...@farny.cz>:

[Nick]

I've attached a copy of the latest RoHS exemptions annex in full (courtesy of Farnell/Newark/Element14).

The word "neon" does not appear :(

Nick

[Dalibor Farný]

Thanks,

there is:

4(a) Mercury in other low pressure discharge lamps (per lamp): maximum
15 mg per lamp
4(f) Mercury in other discharge lamps for special purposes not

specifically mentioned in this Annex

I am not english native speaker so I don't know what "lamp" exactly
means there as it is used for everything from electron tubes to
bulbs.. But nixie tubes fulfil the criterias of "low presure discharge
lamp"..

that will certainly need a consultation with someone involved into that stuff..

Dalibor

2013/10/2 Nick <ni...@desmith.net>:

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+...@googlegroups.com.
> To post to this group, send an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/aaf57e9a-502a-4a54-bcc9-7c889fcfa1b5%40googlegroups.com.

[Nick]

The halogen cycle for Iodine is, I suspect, at too high a temperature for nixies, which is why I was wondering if chlorine would do. Iodine works in car headlamps etc. as the filament there runs at 1000C or so...

Hence my question about inorganic/physical chemists - its their territory, not mine !

Nick

[Tidak Ada]

That is exactly what I said. The quartz glass is needed, because of the high temperature as well as the small bulb is needed to reach that temperature

 

eric

---

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Nick

Sent: woensdag 2 oktober 2013 10:39
To: neoni...@googlegroups.com
Subject: Re: [neonixie-l] Re: Thoughts on nixie lifespan and halogens...

--

You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send an email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/412b2e40-6c6f-4c53-9a0f-f121b7703282%40googlegroups.com.

[AlexTsekenis]

Hi Nick,

I am doing some experimental work on the subject of lifespan due to sputtering.

Could you provide the full reference, including if possible a chapter or page number, for Weston's 50% threshold please? You mentioned it in your first post.

Thanks in advance.

Regards,

Alex

[Nick]

On Wednesday, 2 October 2013 14:29:24 UTC+1, AlexTsekenis wrote:

Hi Nick,

I am doing some experimental work on the subject of lifespan due to sputtering.

Could you provide the full reference, including if possible a chapter or page number, for Weston's 50% threshold please? You mentioned it in your first post.

Thanks in advance.

I put it in the "nixie" article on WIkipedia some time ago - for reference, its:

• Weston, G.F. (1968), Cold Cathode Glow Discharge Tubes, London: ILIFFE Books Ltd, LCCN 68-135075, Dewey 621.381/51, LCC TK7871.73.W44.

on page 340.

Nick

 

[Dalibor Farný]

Hi Alex,

I hope You will share results of your experiments! ;-)

Thanks,

Dalibor

2013/10/2 AlexTsekenis <alexts...@gmail.com>:

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+...@googlegroups.com.
> To post to this group, send an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/384e33af-2754-4e97-a74a-b339a49d503c%40googlegroups.com.

>
> For more options, visit https://groups.google.com/groups/opt_out.

[AlexTsekenis]

Nick, thank you for the swift reply.

I've scanned in p340, here. The book is the exact one in the reference.

On this page Weston explains the impact of digit cycling on lifespan. The point of a gradual rather than abrupt end of life is also made. However I was not able to find the aforementioned 50% threshold. I also checked p240 and p140. Could it be that it is somewhere else in this chapter? I am conscious this would appear as pedantic, but the quoted end of life due to loss of luminosity comes down to this important threshold.

Dalibor, certainly. The work revolves around single-cathode tubes, but the theory and experimental setup are directly applicable to any indicator tube, handmade or not. We should discuss more.

Regards,

Alex

[greg...@hotmail.com]

From what I read about the Halogen cycle,  high-temperature is a key part of the cycle where tungsten dissociates from the halogen, in this case on the filament. Apparently the halogen present in the bulb reacts with the tungsten that essentially condenses on the inner surface of the bulb, thereby keeping the bulb's inner surface clean.

I think for nixies, the first step is to identify a metal that is resistant to sputtering. A few months ago when I was thinking about making my own tubes, I did a small amount of web searching and was not able to find any metals with notable resistance to sputtering. I found lots of information about improving sputtering efficiencies; not really what I wanted.

Before I gave-up, I was first thinking of using nichrome for my cathodes, because I believe it would oxidize much less than other readily-available metals during the high-temperatures encountered with glasswork. I was concerned about it's reflectivity; if you look carefully at 5092 or 6091 tubes (yes, I'm a Burroughs junkie...) you will see some back-reflection from un-energized numerals. I wanted to make really big tubes, and if they are made with overly-reflective metals, the appearance will suffer. And of course I need to chemically etch the cathodes; I dont think nichrome is as easily dissolved as, say, copper, so that puts restrictions on the etchant and the masking material used. 

Then I was concerned about how well glass would form a durable seal around nichrome, and also the differing thermal-expansion coefficients. Kovar is a better material from what I've read. Ugghh...now I have to spot-weld my electrodes to the pins. Even if I got all that figured-out, I still need to remove the impurities. With neon-sign tubes, this done during bombarding; I can only imagine what that would do to a nixie tube.....

So, that's where I buried the project. If I get blessed with eternal life and wealth, I'll take another stab at it.

[John Rehwinkel]

> From what I read about the Halogen cycle, high-temperature is a key part of the cycle where tungsten dissociates from the halogen, in this case on the filament. Apparently the halogen present in the bulb reacts with the tungsten that essentially condenses on the inner surface of the bulb, thereby keeping the bulb's inner surface clean.

I think the halogen reacts with the tungsten before it gets to the inner surface of the bulb. Then, when it gets near to the hot filament, it dissociates, releasing the tungsten back near the filament where it is likely enough to redeposit there.

> Before I gave-up, I was first thinking of using nichrome for my cathodes, because I believe it would oxidize much less than other readily-available metals during the high-temperatures encountered with glasswork.

Interesting idea. Chromium wouldn't oxidize much, but isn't mechanically stable by itself. Chrome-plated nickel would be amusing, but:

> I was concerned about it's reflectivity;

Bingo. Mirror electrodes might be distracting. Or they might be cool.

> if you look carefully at 5092 or 6091 tubes (yes, I'm a Burroughs junkie...) you will see some back-reflection from un-energized numerals. I wanted to make really big tubes, and if they are made with overly-reflective metals, the appearance will suffer. And of course I need to chemically etch the cathodes; I dont think nichrome is as easily dissolved as, say, copper, so that puts restrictions on the etchant and the masking material used.

Also, you increase the surface area (possibly greatly), increasing issues with processing and outgassing. It might also increase sputtering, I'm unsure about that.

> Then I was concerned about how well glass would form a durable seal around nichrome, and also the differing thermal-expansion coefficients. Kovar is a better material from what I've read. Ugghh...now I have to spot-weld my electrodes to the pins.

Nearly all nixies (and vacuum tubes) are made this way. Many tubes have three-section welded leads, with different thicknesses and alloys for the external pin, the lead seal, and the internal lead. These, in turn, get spot-welded to the actual tube elements. So each lead ends up being four different pieces of metal, with three welds.

> Even if I got all that figured-out, I still need to remove the impurities. With neon-sign tubes, this done during bombarding; I can only imagine what that would do to a nixie tube…..

Eddy current ("induction") heating and oven processing, basically. One of our members makes a commercial unit, also available as a kit, partial kit, and open sourced:

http://fluxeon.com

- John

[Instrument Resources of America]

Is there an exemption for Hg in C.F.L.'s and other Florescent lamps, since they are ALL made with it???   Ira.

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send an email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/44f7fc31-d0b2-4a09-b369-783aaa8fa183%40googlegroups.com.

[Nick]

On Monday, 7 October 2013 23:15:27 UTC+1, I wrote:

Is there an exemption for Hg in C.F.L.'s and other Florescent lamps, since they are ALL made with it???   Ira.

Nixes are not fluorescent lamps, therefore I cannot see how this exemption would apply. All exemptions are subject to continual review anyway, so as LED and other Hg-free technologies advance, RoHS have made it very clear that the existing exemptions will be cancelled.

Nick

[lai...@wcoil.com]

Couldn't you just use a piece of canned tuna fish in each nixie? Tuna has
mercury in it and I eat the stuff! Just joking of course.
Tim Laing

But I must add GREAT job on the nixies! Well not really great how about
outstanding job!

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+...@googlegroups.com.
> To post to this group, send an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/20cf3005df5038f4a704e84a2d07%40google.com.

[Nick]

On Wednesday, 2 October 2013 15:55:46 UTC+1, AlexTsekenis wrote:

Nick, thank you for the swift reply.

I've scanned in p340, here. The book is the exact one in the reference.

On this page Weston explains the impact of digit cycling on lifespan. The point of a gradual rather than abrupt end of life is also made. However I was not able to find the aforementioned 50% threshold. I also checked p240 and p140. Could it be that it is somewhere else in this chapter? I am conscious this would appear as pedantic, but the quoted end of life due to loss of luminosity comes down to this important threshold.

 

Finally  checked this - you are right - the actual references to end-of-life are in paragraph 2 of page 334 where he considers EOL to be either a discontinuous glow over the digit or the glow spreading to elsewhere in the tube. I know I've seen the 50% reference somewhere else, so I'll have to dig through my library.

I'd forgotten what a comprehensive book Weston is - lots of good stuff about sputtering in there, including detailed analysis of cathode materials, their performance and lifespan with and without Hg doping...

Nick

[AlexTsekenis]

Hi Nick, thanks for following up on this.

Yes, there are a couple of end of life statements criteria on that page, here is p334 for reference. As far as the end of life luminosity is concerned, I have adopted a threshold of 30% of initial brightness taking into considering the modern applications of these tubes.

Indeed it is a good text with many references to research of the time. The chapter on sputtering does cover two-three theoretical models but these fail at the higher gas pressures in commercial tubes (sputtering rate inversely related to pressure). There is however at least one author who took the experimental path, Stocker B. J. 1961.

I get the impression that research interest on sputtering, one of the most complex topic of this technology, faded away after the introduction of the LED in the late 60s mid 70s. Hopefully we will come to an agreement on the end of life of commonly used tubes using experimental methods, handmade tubes being of particular interest.

Alex

[Dalibor Farný]

Hello!

As for the sputtering.. I once bought a bunch of ZM1042, Z566M, ZM1040
and similar tubes.. Those tubes (ZM1042) doesnt have a mercury
capsule, so I think there is no mercury present. Notice, that those
tubes have a hole on the side of the anode. (look here:
http://www.jb-electronics.de/images/elektronik/nixies/n_zm1043s/zm1043s_aus_2.jpg)
I am not sure about the reason for making such a hole here, but I
think that it is there to show age of the tube by looking on a
deposited layer of metal on the glass just in place of the hole. Some
of the tubes were used, some a lot. And there alway was a layer of
metal on the glass next to that hole. Interesting was that there were
no visible layer deposited on the glass in the place of anode grid.
I later found an information (in Weston or Roth book) about relation
of size of the hole in anode and amount of particles able to go
through it. I didnt study it much as I solved different problems, but
I made a quick calculation and found that for our conditions, the hole
must be smaller than 2.5mm to prevent depositing the metal on glass.
That pretty well fits to a shape and form of the common anode grids.
And it also fits for that "service" hole that is around 4mm wide.

Another thing - is there anyone who saw a tube that would have
sputtered metal layer in front of the anode grid?

What do You think guys?

Dalibor

2013/10/10 AlexTsekenis <alexts...@gmail.com>:

> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to neonixie-l+...@googlegroups.com.
> To post to this group, send an email to neoni...@googlegroups.com.
> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/47d6e7af-f02d-46a7-8bf9-86a845292dc4%40googlegroups.com.

[Dekatron42]

I have several ZM1040, ZM1042 and Z566M where there is a lot of sputtered material on the front of the glass through the anode mesh and also several where there is a lot of sputtered material on the sides of the glass and none of those contain the hole, some have a Mercury pill and some have a getter and some not.

 

/Martin

[AlexTsekenis]

Hello Dalibor,

Regarding material being ejected, it is not charged, thereby its path is not affected by the internal fields of the tube. The trajectory of the atoms is affected by the angle it was ejected, mean free path, bouncing off surfaces and other such 'ballistic' factors. The idea is as the travel distance increases, the path will increasingly deviate from normal to the cathode. By forming the anode into a fine mesh, atoms at large angles adhere to the anode mesh and not the glass. In other words the anode acts as a collimator.

Having a look at the IN-13, we can see  the pattern on the anode (warning, tubes were harmed):

http://s13.postimg.org/irdgvxe7b/DSC04211.jpg

and here is the sputtered pattern on the glass:

http://s13.postimg.org/xxjimv487/DSC04215.jpg 

Notice the blurring. The reduction of the deposited material is a function of the angle of incoming atoms which is random (loose meaning) but with a distribution. It is that distribution that we see as the blurring. So I don't think you can eliminate sputtering completely by optimising the mesh geometry. You can greatly reduce it. For dekatron's tubes my guess is that the pressure is lower and atoms travel fairly straight. It could also be that they have so many hours that the varying thickness is no longer distinguishable. In any case I don't want to think how it would have been without the mesh anode, probably like this: http://www.radiomuseum.org/forumdata/users/14211/tuneon_brian_001_sml.png

Regarding that hole, I assume there is another one aligned on the other side? If so we can actively measure deposition by sending a laser beam down the path and measuring attenuation. But be careful of the simplification pitfall. The tube is not normally viewed from inside that hole so the relationship between the amount of material at the front (where it matters) and at the hole is crucial, and unknown :-) That's why a front-view, passive measurement method is preferred.

To add to the speculation of why is that hole there, it could be an alignment hole for the metal punch during manufacturing? A manufacturing mark like the 'nipple' at the bottom of injection moulded parts? If it was for life/endurance measurements it would have been used in the lab and maybe removed from mass production? Also if this was the industry standard we would have seen more in the textbooks? Difficult to say.

Alex.

[greg...@hotmail.com]

If iron was used, I wonder if a small magnet would collect sputtered material ? The magnet could even be external to the tube if it was strong enough, like those buckyballs that kids were swallowing. Then again, it might draw sputtered material towards a certain area and cause shorts.

I know for a fact that IN-1's develop internal shorts between cathodes when they run with the same digit continuously illuminated. Not only do you get 2 cathodes glowing simultaneously, the short can be measured with an ohmmeter. Zapping the shorts slowly causes incandescense, until they burn-thru. At that point the tube worked again, until it developed another short. I was tempted to break-open a failed tube, then decided against it because the mechanical shock would probably break the short and I'd never find it.

I assumed the shorts I experienced were tin whiskers. But maybe it was sputtering. I gave up worrying about it because I switched the IN-1's to 6091's and the clock has been running perfectly for months now; it used to fail within a few days.

[AlexTsekenis]

Hi Greg,

You are right, it would, but there is a catch. Ionised particles in the glow will also experience a force (Lorentz force) in the presence of a magnetic field. Here is a demo using my favourite tube:
http://www.youtube.com/watch?v=T1isdFu9Dp0&feature=youtu.be

Alex.