Mercury not all it’s cracked up to be?

[Paul Andrews]

I recently saw a post that suggested that the addition of a mercury dopant to Nixie tubes does not confer the protection that we have all been led to believe - apparently the Russians made a study suggesting this to be the case.

Regardless of whether this is true or not, I was wondering if there is a way to determine the presence of mercury in tubes experimentally in a non-destructive way, for example by using a spectrometer. If so, it might be interesting to test a sample of tubes, including some that were developed later, to see if they really do contain mercury. The presence of mercury wouldn’t be conclusive evidence of its effectiveness, but the absence could lend credence to the argument. It would be interesting information either way.

I know that some of you will say that you can see a tell-tail blue glow, but I have quite a few nixies, including later examples, that do not show this. I wouldn’t take this as proof that those tubes don’t contain mercury, it could just be smaller amounts.

This also got me thinking; if it isn’t mercury that confers a longer life, then could it be something else? Could it be fine-tuning the cathode material? Could it be fine tuning of the gas mixture and/or pressure? Which got me wondering if there would also be a way to determine the pressure in the tube? For example by examining the width of spectral emission lines?

[Terry Kennedy]

On Friday, July 17, 2020 at 1:27:39 PM UTC-4, Paul Andrews wrote:

This also got me thinking; if it isn’t mercury that confers a longer life, then could it be something else? Could it be fine-tuning the cathode material? Could it be fine tuning of the gas mixture and/or pressure? Which got me wondering if there would also be a way to determine the pressure in the tube? For example by examining the width of spectral emission lines?

I expect Dalibor knows since he has tubes that have been in service for years and aside from a small contaminated batch some years ago he has an infinitesimal failure rate. He also provides a 10 year warranty on his tubes. He may or may not consider this information to be a trade secret, though.

[Dekatron42]

There is some information in this article, presumably first hand information from a Burroughs engineer, on the extended life when mercury was added: https://spectrum.ieee.org/tech-history/dawn-of-electronics/the-nixie-tube-story-the-neon-display-tech-that-engineers-cant-quit (just search for mercury).

I also know of a few patents whoch describe this, but don't have the numbers at hand, I do however think these were by Burroughs.

/Martin

[Nicholas Stock]

Here's the text in question...

In a 2011 email to me, Roger Wolfe, a Burroughs engineer, recalled the team’s first fragile attempt: “We put the tube on life test overnight. When we came in the next day, so much cathode material had sputtered onto the dome of the tube that the numerals were no longer visible. We had invented a tube with a 24-hour life!”

After some tinkering, Wolfe wrote, they discovered that the addition of mercury vapor would greatly extend the tube’s life span. The sputtering had been caused by the accelerated neon ions striking the cathode. But when the neon ions collided with the heavier mercury molecules, their energy dropped below the point where they could damage the cathode.

“We secured a tiny ampule with mercury sealed inside, wrapped a few turns of resistance wire around the ampule, [and] connected the ends of the wire to two of the [tube’s] pins,” Wolfe wrote. The tube was then sealed, and the team ran current through the wire, which heated and broke the ampule, releasing the mercury.

In August 1955, Burroughs unveiled its new indicator tube at Wescon—the Western Electronic Show and Convention, in California—which was for many years the leading U.S. electronics event. Soon after, it began shipping the first tubes to customers. That December, the company filed for a patent on its “glow indicating tube” [PDF] The devices were mechanically superior to the numeric display tubes still on offer from National Union: They had dedicated anodes made from wire mesh, and instead of hand-bent wires, the cathode numerals were etched out of thin sheet metal. The addition of mercury prolonged the tubes’ life span, eventually to more than 200,000 hours.

--
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 view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/cf14f263-3901-4fef-8dbe-b80c9b66583ao%40googlegroups.com.

[Nicholas Stock]

I should have prefaced that by explaining that it was written by our very own Jens Boos, who's a member of the group! :)

Nice job Jens!

[Dekatron42]

Where did you read about this by the way?

/Martin

[Paul Andrews]

My point really is, is there a way to determine the composition of the gases inside a Nixie tube? It would be interesting to know, as this is something that is lacking in datasheets.

I’ll try and dig up the Russian paper.

On Jul 17, 2020, at 9:37 PM, Dekatron42 <martin....@gmail.com> wrote:



--
You received this message because you are subscribed to a topic in the Google Groups "neonixie-l" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/neonixie-l/4QWf5akhjCU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/f2063952-e927-4bc0-acfb-94508f8dda56o%40googlegroups.com.

[Toby Thain]

On 2020-07-17 10:00 p.m., Paul Andrews wrote:
> My point really is, is there a way to determine the composition of the
> gases inside a Nixie tube? It would be interesting to know, as this is
> something that is lacking in datasheets.

Get a sufficiently precise spectrogram?

--Toby

>
> I’ll try and dig up the Russian paper.
>
>> On Jul 17, 2020, at 9:37 PM, Dekatron42 <martin....@gmail.com> wrote:
>>
>> 
>> Where did you read about this by the way?
>>
>> /Martin
>>
>> On Friday, 17 July 2020 19:27:39 UTC+2, Paul Andrews wrote:
>>
>> I recently saw a post that suggested that the addition of a
>> mercury dopant to Nixie tubes does not confer the protection that
>> we have all been led to believe - apparently the Russians made a
>> study suggesting this to be the case.
>>
>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "neonixie-l" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/neonixie-l/4QWf5akhjCU/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to
>> neonixie-l+...@googlegroups.com

>> <mailto:neonixie-l+...@googlegroups.com>.

>> To view this discussion on the web, visit
>> https://groups.google.com/d/msgid/neonixie-l/f2063952-e927-4bc0-acfb-94508f8dda56o%40googlegroups.com

>> <https://groups.google.com/d/msgid/neonixie-l/f2063952-e927-4bc0-acfb-94508f8dda56o%40googlegroups.com?utm_medium=email&utm_source=footer>.
>
> --
> 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

> <mailto:neonixie-l+...@googlegroups.com>.

> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/24A83C80-822E-4DEF-A34B-2825C1950315%40gmail.com
> <https://groups.google.com/d/msgid/neonixie-l/24A83C80-822E-4DEF-A34B-2825C1950315%40gmail.com?utm_medium=email&utm_source=footer>.

[Nicholas Stock]

I wonder if anyone ever tried adding a few iodine crystals to a nixie? I know it’s added to some plasma tubes to give a certain plasma color (i have one....nice blue tinge to the plasma glow)....

Hmm....

Sent from my iPhone

> On Jul 17, 2020, at 20:35, Toby Thain <to...@telegraphics.com.au> wrote:

> To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
> To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/e7b3c0ae-3a55-5a80-9610-721fad69b12b%40telegraphics.com.au.

[Dalibor]

Hi!

Paul - could you please share the source of that information? I would like to look at it.

My experience is that Hg is absolutely necessary in nixie tubes.

Few tests I made:

1. two tubes - one with, one without Hg. Both running at 5 times rated current on cathode 8. After few weeks, Hg free tube sputtered so much metal that it caused internal short between 8 and anode, test ended. The Hg free tube was heavily poisoned on unused digits, they were all practically black. The 8 had 50% metal sputtered away.
The Hg doped tube was visually just like a new tube, no black deposit, the 8 was slightly more matte than other digits. Light poisoning on unused digits, quick healing reverted it all back to 100% condition.

2. Comparison of poisoning development between Hg and Hg free tubes. Two clocks, 6 tubes with and 6 tubes without Hg, no anti-poisoning routine, rated current. I wanted to see whether the Hg only slows down poisoning development or it has also other impact on the tubes. After 12 months, the Hg free tubes were all black inside, even tubes (2,4,6) had no poisoning on cathodes, 3,5th tubes had heavy poisoning on 6-9, other digits ok. 1st tube had cathodes 0,1,2 thinner, 1 being half eaten away. Other digits heavily poisoned.
The Hg doped tubes had poisoned 1st tube (3-9 digits), odd tubes had slight poisoning on 6,8,9.
What is interesting - the poisoning patterns (location of the poisoned segments) were different between Hg free and Hg tubes - this suggests some other differences happening inside - I’ve set up more tests that are being run right now.

My conclusion is that a Hg free tubes are difficult to keep without poisoning and also recovery from poisoning at rated current is usually not possible.

I have photos and more tests - will post next week..

Regards!

[Robert G. Schaffrath]

In terms of longevity provided by mercury one just need look at the Chinese QS27-1 and QS30-1 tubes that contain no mercury. They are known to have very short lives. I also understand that Russian tubes prior to the IN-7, which does contain mercury, also have very short lifespans. I see a lot of the IN-1, IN-2 and IN-4 tubes on eBay at relatively low prices compared to the higher numbered tubes. I am guessing the demand is not there because of their failure rate or they have to discount them because they will fail.

On Friday, July 17, 2020 at 1:27:39 PM UTC-4, Paul Andrews wrote:

[Jeff Walton]

Thanks for the information, Dalibor. Appreciate the findings with actual supporting evidence. It would be nice to know the actual role of the mercury in the process and how it protects the cathodes. Can you tell the gas mixture in a tube with a spectrograph while the nixie is operating or does that only show which gasses are ionizing?

I had a 6 digit clock with NL-6844A top view nixies which ran continuously and didn't last a year before the digits became unreadable and couldn't be healed. I believe that these are a non-mercury style nixie tube, but I didn't know that when I bought them. I replaced them with slightly larger NL-6091 tubes that have been running fine ever since (>4 years) with no degradation or modification to the clock.

While my experience with this clock and these tubes is not scientific, I've certainly come away with a belief that whatever role the mercury plays, it seems to be essential to long life and cathode protection. Not sure why anyone would want to make a tube without mercury or other element that helps protect the cathodes and provides long life.

Jeff

[Kevin A.]

I found an IEEE article which contains an anecdote from a Burroughs engineer who worked on nixie R&D. 

Link: https://spectrum.ieee.org/tech-history/dawn-of-electronics/the-nixie-tube-story-the-neon-display-tech-that-engineers-cant-quit

To quote that article:

"Roger Wolfe, a Burroughs engineer, recalled the team’s first fragile attempt: “We put the tube on life test overnight. When we came in the next day, so much cathode material had sputtered onto the dome of the tube that the numerals were no longer visible. We had invented a tube with a 24-hour life!”

After some tinkering, Wolfe wrote, they discovered that the addition of mercury vapor would greatly extend the tube’s life span. The sputtering had been caused by the accelerated neon ions striking the cathode. But when the neon ions collided with the heavier mercury molecules, their energy dropped below the point where they could damage the cathode.

“We secured a tiny ampule with mercury sealed inside, wrapped a few turns of resistance wire around the ampule, [and] connected the ends of the wire to two of the [tube’s] pins,” Wolfe wrote. The tube was then sealed, and the team ran current through the wire, which heated and broke the ampule, releasing the mercury."

--
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 view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/5f133c30.1c69fb81.225bd.6a51%40mx.google.com.

[Dan Hollis]

Does the gas make any difference? What about heavier gases like Kr, Kr+I, or Xe?

https://www.youtube.com/watch?v=l3M0cY29Pq0

-Dan

> To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/CALcVLKKakeYS%2B99Ae5myyP-cDfikMsg2uriZB4A3eePxqkZy9A%40mail.gmail.com.
>

[David Speck MD]

Nicholas,

Iodine (and all of the halogens) produce excimers (excited dimers) with the noble gases.  Excimers do produce unusual colored discharges, but they have a problem -- there is nothing more chemically reactive than an ionized halogen. 

Tubes containing metal electrodes and a halogen "clean up" rapidly, as the ionized halogens react with and permanently bind to the metal electrodes.  After a relatively short period of operation the halogens are depleted.

Some plasma sculptures are made with halogens, and do produce interesting colored displays, but the halogens will even attack the silicon dioxide of the glass envelopes, and clean up after some period of operation.  I've seen a traffic light red plasma globe that used neon plus a halogen, as well as a lovely turquoise globe that used a halogen plus one of the heavier noble gases  Neither had a long lifetime.

Dave

[Nicholas Stock]

Thanks David, it had occurred to me that the halogen may eventually react with the cathodes, but wasn’t sure at what rate....sounds like it’s pretty fast. 

Cheers.

Sent from my iPhone

On Jul 18, 2020, at 17:13, David Speck MD <dr.s...@davidspeckmd.org> wrote:



[Dalibor]

Hi!

I am sending the photos from the Hg-free and Hg-doped comparison test. There are photos for each cathode in the Dropbox directory, "before" means photo from the beginning of the test, "after" from the end of the test.

https://www.dropbox.com/sh/f4oisg371cu0q9q/AAB4ww3AZFp2UhusmMgzjOd9a?dl=0

The idea behind this test was to find a possible way how to set up accelerated tests for lifespan and poisoning study. Currently, we need to wait a long time before we get results from the test as everything with Hg takes soooo long.. I thought that Hg free tubes might have the same behavior as Hg doped tubes, just the processes inside are faster. With Hg free tubes we would get results very soon.

Test conditions:

- two clocks, one with 6 standard Hg doped tubes, one with 6 Hg free tubes - all other parameters identical (material, gas filling, manufacture)

- standard clocks (170 VDC PSU with 8k2 anode resistor) with cathode poisoning prevention routine fired every 10 minutes for 10 seconds duration.

- there is a camera taking a photo each night

- on top of this routine, each digit is displayed for 2 seconds each night for taking the photo  

- the clock with Hg free tubes has been running 24/7 since 26 Nov 2019 (5700 hours), last photo from today

- the clock with Hg doped tubes has been running 24/7 since 25 Dec 2019 (5000 hours), last photo from today

Conclusion - the Hg free tubes wear much faster than Hg doped tubes. Even with the cathode poisoning prevention routine, the standard tubes get some little poisoning. This has been already taken into consideration in the latest clock firmware release (a month or so ago), the customer's clocks are already updated (thanks to cloud connection!). The modified routine exercises the unused digits for a longer time.

I can't wait to make time-lapse videos from the photos :-)

The camera rack picture attached..

Have a great day!

Dalibor

[Alex]

That is an impressive test wall and camera setup! I like the single red tube, is that an original 568 for a compare or a painted tube of your own creation?

Looking at those comparison photos it seems that some tubes suffer fairly significant poisoning and others very little, though the effect is quite nostalgic in some ways, like a flickering neon sign above a cheap motel, static overlay on a CRT TV or a sticky digit on a filling station pump, it almost looks period to the technology.

Either way, your dedication to the process is very inspiring, thanks for sharing your test results and I look forward to your future work / tube designs!

- Alex