Good Nixie Tube books

[Dominic Scardino]

Hello, I am new to the site and am looking into to making my own nixie tubes. Before I get to far into it I would like to do as much research that I can on how they work and best ways to make them work and so on. 

So I was just wondering if you guys had any good suggestions on what books to look to for some background knowledge and if you have a PDF of it if you would share it.

P.S. I already am reading "Cold Cathode Glow Discharge Tubes" By Weston

[gregebert]

On a whim, I did a search on helium leak-detection and came across this document link which is helpful info:  https://www.leyboldproducts.com/media/pdf/90/c7/87/Fundamentals_of_Leak_Detection_EN.pdf

Leaks will kill nixies, so be sure to understand them.

Several years ago I was researching neon art, and I decided against it until after I retire because I dont have the time to work on it with my demanding day-job.

The most-concerning step was bombarding, which involves kilovolts at several hundred mA of current, which is instant death if you contact it. Nixie tubes dont require bombarding per-se, though if you want to make large tubes you might need to in order to cleanse the glasswork. Other options I've heard are to bake the entire assembly and flush with gas. Basically, tubulate both ends, rather than single-ended. If I can find a way to do neon work without bombarding, I will give it a try. I will not, under any circumstances do any bombarding. If flashover can happen thru your manifold into your vacuum pump, it will find a way of hitting you as well. I've even seen a video on youtube where an experienced electrician got killed opening the grounded door of some equipment even though his body was not anywhere close to being the best ground-path.

Baking-out the metalwork seems easy enough with an induction heater; I think the challenge is finding metals that seal nicely against glass (Kovar), and cathode materials that resist sputtering which and can be etched.

Working with mercury (toxic) is not something I want to do either, but apparently it's necessary inside nixies in order to reduce sputtering. Assuming I can get this far without bombarding, I will continue researching materials with low sputtering yields. Most industry research I've seen seems to be finding materials with HIGH sputtering yields, for use in semiconductor wafer processing.

There's a fair amount of info on the web about glasswork; I have not seen a lot of info about annealing nixies but I'm fairly certain you will want to do that for reliability reasons to minimize stress around pins and joints.

[gregebert]

Here's the book list

• Neon techniques (Strattman)
• Neon Engineer's Notebook (Crook & Fishman)
• Neon Superguide (Caba)

Not nixie, but overall very informative and will help you understand the process & techniques

[John DeArmond]

On 2/11/19 6:55 PM, gregebert wrote:
> The most-concerning step was bombarding, which involves kilovolts at
> several hundred mA of current, which is instant death if you contact it.
> Nixie tubes dont require bombarding per-se, though if you want to make
> large tubes you might need to in order to cleanse the glasswork. Other
> options I've heard are to bake the entire assembly and flush with gas.
> Basically, tubulate both ends, rather than single-ended. If I can find a
> way to do neon work without bombarding, I will give it a try. I will not,
> under any circumstances do any bombarding. If flashover can happen thru
> your manifold into your vacuum pump, it will find a way of hitting you as
> well. I've even seen a video on youtube where an experienced electrician
> got killed opening the grounded door of some equipment even though his body
> was not anywhere close to being the best ground-path.

What you've seen on the net is the traditional method which uses a 25kV
bombarding transformer.  This is a horribly bad practice but it
continues to be used 'cuz nobody wants to change.  Here's the problem.

The setup of a 25kV stray flux transformer and a choke is essentially a
constant current source but with 25kV of compliance voltage.  If you
look at the curve of voltage drop vs pressure in a discharge tube, it
looks like a check-mark.  Once a bit of vacuum is formed, the tube will
fire.  As the pressure drops, so does the voltage across the tube until
a critical pressure point is reached.  About 2.5 Torr (don't take that
to the bank - my memory isn't so good anymore) when the voltage starts
rising as the pressure is further dropped, culminating with no discharge
at about 0.01 Torr.

Here's the problem.  If the operator mistakenly keeps the high voltage
on as he pumps down, the voltage will rise to preserve the set current. 
It will rise all the way to 25kV.  The bends are much thinner than
straight glass so if two bends are near each other separated by a few
feet of other bends, about the time the tube starts flickering, there
will be some fraction of 25kV between bends.

A characteristic of glass furthers the problem.  Glass is electrically
conductive when the glass is hot enough to be extremely stiff.  Bends
tend to get hotter than straight line tubing. So the glass is thin, hot,
maybe enough to go be conductive and with the flickering, the bombarder
is outputting the full 25kV, the result is punch-through and a ruined tube.

When I decided to take on neon as a hobby business, I built all my tools
from scratch and I approached things from an engineering perspective. 
For the bombarding transformer I got a 14.4kV 20kVA pole pig from my
local utility (they owed me a favor :-).  This one had a switchable
primary voltage.

The switchable primary voltage transformer is used for distribution
capacity upgrades.  The REA-installed rural distribution all used
7.2kV.  With rural growth, the distribution lines became overloaded. 
The solution is to hang dual primary voltage transformers.  On the
appointed day, a lineman climbs each pole and flips the primary switch
to 14.4kv while the transformer tap at the substation is changed from
7200 to 14.4kV.  The tragically funny part of this is they used to
invariably miss a home or two which put 440 on their 240 drops, usually
setting the house on fire.

Back to neon, this type transformer is so useful because one can dial in
just the max voltage needed for the given tube.  Your typical
hobby/business bender is very unlikely to have dozens of feet of glass
in one unit.  More typical would be something like a stylized OPEN
sign.  I once did a window hanger for a health food store that had a
seafoam frame, a red OPEN word and "health" and "food" done in purple. 
The open unit was on a timer, the rest ran all the time.

I made quite a number of this kind of sign and my curiosity got the best
of me so I took some measurements.  The OPEN unit in 13mm tubing took
around 7kV to strike and about 3500 volts running.  Given that data, I
knew that the 7200 volt setting would satisfactorily bombard that unit
without any risk of flashover.  I did an experiment of pumping the unit
on down to a high vacuum with the bombarder on.  The internal discharge
went out but no flash-over or burn-through.  I've only ever had one
burn-through and that was because I forgot and bombarded at 14.4kV.

A raw pole pig, being a constant voltage transformer, isn't ready for
bombardment yet.  I used a large external inductor to make the thing
close to a constant current device

As far as safety goes, I designed my station with maximum safety in
mind.  A grounded metal manifold reduced the risk of flashbacks to 0. 
Flashbacks would not happen in glass manifolds except for that 25kV
transformer.

I have an electrician's insulating mat at the control point.  To start a
cycle, one first inserts the key into the keyswitch.  Then he presses
the "arm" light which illuminates a large green light on the wall.  This
is a time gate.  If the pneumatic foot pedal isn't pressed within 10
seconds, the gate resets and the green light goes out.  I used a
pneumatic foot switch so I would be completely isolated as the
bombardment continues.

I used refrigeration leak-proof diaphragm valves.  This type of valve
has a large plastic knob that is a good insulator, though that was
really overkill since the manifold is firmly and redundantly grounded.

I can go into more details about my safety controllers if there is interest.

> Baking-out the metalwork seems easy enough with an induction heater; I
> think the challenge is finding metals that seal nicely against glass
> (Kovar), and cathode materials that resist sputtering which and can be
> etched.

Remember that my ex-partner is still selling the Neon Roy, an
inexpensive induction heater designed (by me) specifically to convert
'trodes and flash getters.  I recommend 2 so that both trodes can come up

>
> Working with mercury (toxic) is not something I want to do either, but
> apparently it's necessary inside nixies in order to reduce sputtering.

Metallic mercury isn't particularly toxic.  I still have part of a 72 lb
carboy of mercury that my parents bought me for Christmas.  I've played
with it.  I accidentally swallowed a bit (poison control was amazingly
rational - they said to just let it pass)  I've suffered actual acute
mercury poisoning (a homemade mercury still with a "steam" leak) so I
have a pretty good handle on the stuff.  I'm not at all careful in my
shop with little droplets scattered all over my bombarding table.  Yet
my annual heavy metal blood test comes out lower than the average.  I
bought a home mercury tester and, as instructed, placed the sample
container on the table.  It barely registered anything.  I'm not
careless with mercury but I'm not chemophobic either.

> Assuming I can get this far without bombarding, I will continue researching
> materials with low sputtering yields. Most industry research I've seen
> seems to be finding materials with HIGH sputtering yields, for use in
> semiconductor wafer processing.
>

I have a book called Gas Tube Discharges or something like that.  It has
a chapter on sputtering and how to avoid it.  Basically use high work
function metals.  As one would figure, the higher the work function, the
higher the price.

John