Nixie Tube Reconditioning Possible?

[Alan Franzman]

I have recently acquired some scarce Nixie neon digital display tubes, Burroughs number B7032.  They are about 5 1/2 inches tall overall with a 14-pin bakelite base 2 1/4 inches diameter, displaying digits 2 inches tall.  Unfortunately they are rather darkened (though still functional) from either long use or excessive current.

I got an idea that sand or similar abrasive introduced into the tube may be used to clean the darkened glass (long story -- see below.)   Several technical questions remain, some of which may need to be answered experimentally:

1.  Would it be possible to use a torch to open a hole in the Nixie tube envelope without cracking the glass or damaging the internal elements or the bakelite base?

2.  Would opening the envelope to the atmosphere harm the internal workings of the tube due to oxidation, etc.?

3.  Would there be any harm in leaving sand inside the tube when it is sealed up again, to clean off the deposits next time the tubes darken?

4.  Is the gas inside a Nixie similar enough (pure neon, or some kind of a mixture?) to that used in neon signs that its functioning would not be adversely affected if it were refilled with this?

5.  How critical is the gas pressure to the Nixie's proper operation, and could a neon sign shop achieve the correct pressure?  What exactly is the pressure inside a Nixie tube anyway?


The long story of how I got the idea:

Recently I toured a museum in the Los Angeles/Hollywood area -- not sure which one it was now, perhaps the Hollywood Entertainment Museum, or the Museum of Radio and Television -- where they had on display a very old piece of film or television studio equipment, a lightbulb cleaning machine.  It is a large spidery looking machine which operated somewhat like an amusement park ride, having several arms which spun around in various directions.  The ends of the arms held old high-powered studio incandescent lightbulbs (not powered), 1000 watts or more as I recall, with large spherical glass about 8 or 12 inches in diameter.  These bulbs tended to blacken quite quickly when in use, so they were made with a small quantity of sand inside.  At the end of each day's work, the bulbs would be placed into this contraption and spun around slowly in all directions overnight and the sand would clean the black deposits from inside the glass, leaving the bulb looking shiny new again for the next day's shooting!

Recalling this gave me the idea that a similar process could be used to "un-blacken" the insides of my Nixies.


-----------------------------------------

Alan "A.J." Franzman

Email: a.j.franzman at verizon dot net

-----------------------------------------

[This message has been edited by Alan Franzman (edited August 19, 2001).]

[Alan Franzman]

I posted the above topic recently on a few usenet newsgroups.  Here are some excerpts from the more reasonable sounding replies.

 

quote:Originally posted by Don Klipstein (don@misty.com - Odd Light Bulbs ):

I would not use an abrasive - it would scratch the glass.  If I was going to try this, I would pour in hydrochloric acid and keep it in just long enough to eat the dark coating on the glass since it will eat the electrodes also.  Maybe have a glassblower separate the glass bulb and then have it cleaned with acid, or have a glassblower rebuild the darn thing.  Sounds expensive to me!

Well I have to disagree with this opinion, I think that acid would do far more harm to the works than sand would do to the glass.  Although I have learned that the material I saw in the museum bulbs was in fact not actually sand, thanks to Tim - Sand Bulb.  I don't think sand would scratch the glass anyway, considering it is only by gravity that the sand will be made to move around the inside of the tube.  As for the expense of hiring a glassblower, I thought that this type of operation would be almost routine for a neon sign shop.


   

quote:Originally posted by Don Klipstein:

I think neon lamp electrodes may have some coating that would oxidize.  Without this coating, the voltage drop is a little higher and the starting voltage may be a fair amount higher and the tube may blacken more easily.

I don't know much about this, perhaps only experimentation will reveal the answer here.


(With respect to leaving sand in the tubes when they are recharged and re-sealed):

 

quote:Originally posted by Don Klipstein:

No harm if the tubes stay rightside up as far as I know.

I thought about this, and it does seem to me to be possible that the dark material from the glass, being evaporated or sputtered from the electrodes, is probably fairly conductive.  Mix enough of it in with the sand and there may be a problem of short-circuiting, though this may also depend on just how substantial the deposited material is as well.  In the event of such a short circuit, the substance may simply re-evaporate thus clearing the short.  (We don't need to mention using tungsten granules instead of sand!)


   

quote:Originally posted by Don Klipstein:

Neon glow lamps traditionally have 99.5 percent neon .5 percent argon at a total pressure of 8 Torr.  Maybe a little less total pressure for larger sizes.  Maybe use a little more argon since argon ions sometimes hit the glass hard and bury themselves in the glass, reducing the argon content a bit.  High intensity neon lamps have pure neon but need more voltage.

I think neon signs operate at a similar pressure.  Pressure is not totally critical, but:
 
Too low - you get dimmer glow and you want less current.  Full current with underpressure causes blackening to occur more easily.
 
Too high - you need more current to cover the electrodes with glow - possible excessive heating of the electrodes.

Too much pressure and you may get a piece of main discharge column (glow filling up space) in addition to the electrode glow.  This increases the voltage requirement - at least this feature is just a bit of dim purplish fog if the mixture is 99-99.5 percent neon, .5-1 percent argon.  The "fog" will be a much brighter orange-red with pure neon.
 
  The cathode glow in a neon-argon mixture is, strangely enough, a non-reddish to maybe very slightly yellowish orange.  This glow is reddish orange in pure neon.

Well I don't know anything about what the gas(es) may be.  I noticed that he didn't mention mercury.  This is good, since a neon sign shop won't even touch something that they suspect may contain or once have contained mercury.  (I don't understand why, after all they still put mercury vapor INTO sign tubes, don't they?)  And his description of the color of a neon-argon mix sounds right.  Nixie glow is definitely not as red as the main discharge of a neon sign tube.  Spectral analysis, anyone?


   

quote:Originally posted by Mike Harrison (mike@whitewing.co.uk):

If the black deposit is resting on the glass surface (as implied by earlier comments about scratching), it could be worth trying using ultrasonics to rattle it off from the outside without the need to open the tube.  I'm sure it wouldn't hurt to try a tube in an ultrasonic cleaner for a few hours, or maybe use some sort of hard U/S probe on the glass surface - the hard glass should transmit the vibrations well.

An interesting idea...  anyone want to give this a try, or know anyone I could go to in the Los Angeles area?


*****

Well, that's all I have so far.  Still nothing definite on getting the (sans-nipple) tubes open in the first place, though of course the non-invasive ultrasonic method deserves to be the first one attempted.

Not much on oxidation, though I have a possible way around that if it turns out to be a major concern... simply open the tube, put in the sand (possibly along with some kind of a getter?), then quickly evacuate and reseal the tube.  After the black stuff is cleaned off the glass, reopen the tube, dump out the sand, evacuate the tube, fill it with gas and reseal it.  This will minimize the electrodes' exposure to the air.


-----------------------------------------

Alan "A.J." Franzman

Email: a.j.franzman at verizon dot net

-----------------------------------------

[This message has been edited by Alan Franzman (edited August 19, 2001).]

[James]

Hi Alan,

First of all, regarding tube opening this is not too difficult.  Take a piece of pyrex glass rod about 5mm diameter, melt the end of this and push the soft glass over the end of the tip-off on the nixie.  Then snap off the pyrex rod and it will take some of the original nixie glass tip with it.  If this doesn't open the tube to atmosphere, do it a few times until you break the seal.

The black deposit is most probably either iron or nickel metal which has been sputtered from the internal assembly and onto the glass.  This could be removed with a fairly strong nitric or hydrochrloric acid, admitted via the opened tip, and since the coating is so thin I think you could eat all of that away without appreciably dissolving the electrodes.  Throrough washing with distilled water afterwards is absolutely essential to remove all traces of acid from the glass-metal seals which will be attacked over time if any acid remains.

If this doesn't work, try the sand or tungsten chippings scouring technique (I can assure you they won't leave scratch marks on the glass).  Be careful not to get electrically conductive tungsten flakes caught in the electrode assembly (that killed many old studio lamps by short-circuiting some of the filament coils), and if you use sand make sure that every last grain is tipped back out of the tube.  Most things you use to do the cleaning will contain impurities that will outgas and destroy the gas quality when you come to re-pump the tubes.  It would be a good idea to rinse the tube out with distilled water again to remove the sand and scraped off nickel coating particles since these could also outgas.

Assuming this works, any glassblower should be able to re-attach a new glass exhaust tube over the tip-off which was removed earlier.

Use this new tube to flush "forming gas" into the nixie (10% hyrdrogen in nitrogen) and torch the glass to heat everything up to about 300 celsius.  The presence of hydrogen will remove any oxidation of internal components, and using it diluted in nitrogen will reduce the risk of a hydrogen explosion.

Then, still filled with forming gas, attach the new exhaust tube to a vacuum system, pump the device down to the best vacuum you can get, and heat all the glassware to about 380-400 celsius.  This is absoluetly essential to prevent the tube re-blackening quickly again.  Then admit about 8 torr of the 0.5% Ne in Ar fill gas, and strike a discharge in the tube.  If you can do this at several times the rated current, almost to the point of making the electrodes glow dull red hot, you will get better tube life and considerably lower striking voltage.  Pump the tube back down to high vacuum, admit the final gas filling when it has cooled to room temperature, and then tip it off.  Hopefully if you have outgassed the glass and metal components by heating them sufficiently, you will get a long life from the tube again!

I do not think that you need to worry about internal oxidation, but using the forming gas cleaning technique is an added precaution which will reverse any oxidation reactions that may have happened.

Hope this helps, let me know how you get on!  BTW I am sure if you don't find a local neon sign shop who can help, that pretty much any University Glassblower would be pleased to help you out on an interesting project like this.


[This message has been edited by James (edited October 08, 2000).]

[Alan Franzman]

Thanks for your input, James.

I'm sorry I didn't make it clear enough that the Nixies I referred to do NOT have an exposed tip-off nipple.  If they do have any, it would have to be under the bakelite bases.  I do not relish the idea of trying to remove the bases considering that I would have to somehow break the cement bond between the bakelite and glass and unsolder the (concealed) wiring to the 14 contact pins, then somehow later reverse the process!

I think the only way I could heat all 14 pins at once to remove the base would be to use a solder pot of some kind.  However, these tube bases have a key-splined center post (part of the bakelite) that is actually slightly longer than the contact pins, so to wet all the pins with molten solder in order to remove the connection wires, this post would have to be submerged in the solder as well.  How much heat can bakelite withstand anyway?  I saw that you recommended heating the whole works up to 400 C as part of the process.

Here's an image of 2 of these Nixies:

           
These are the 2 best-looking of the 5 I have.  Sorry about the glare and poor positioning; I didn't take this image since I don't have a digital camera.  It was taken, presumably, by the person who sold me the tubes on eBay.  Therefore, it is set up to conceal rather than expose the darkening of the glass, though you can just see a few splotches on the upper side of the tube on the right, and slight darkening of the right side of the tube on the left.  Note that the left tube is actually upside-down; you can just make out the "7" which is the second digit cathode from the front.  The "3" which is in front of the "7" did not catch as much light so it is harder to see.

I don't even know where I might find a "university glassblower" here.  The closest university to me is Cal Poly in Pomona but I don't think they make their own glassware.


-----------------------------------------

Alan "A.J." Franzman

Email: a.j.franzman at verizon dot net

-----------------------------------------

[This message has been edited by Alan Franzman (edited August 19, 2001).]

[Alan Franzman]

More input from the usenet newsgroups:

       

quote:Originally posted by Kenny Greenberg (kenny@neonshop.com - Krypton Neon):

>Unfortunately they are rather blackened (though still
>functional) from either long use or excessive current.

The blackening is likely sputtered metal from the electrodes which advances over time.


>1.  Would it be possible to use a torch to open a hole in the Nixie tube
>envelope without cracking it or damaging the internal elements or the bakelite
>base?

If it has a tubulation nipple you could score it and touch it with a molten tip of a small glass rod. That way it will not suck in.  But otherwise just using a torch will work. You will need to splice a tubulation to that hole for reprocessing.


>2.  Would opening the envelope to the atmosphere harm the internal workings of
>the tube due to oxidation, etc?

Not really, but the activation coating on the metal may already be played out.  That will probably raise the operating voltage when reprocessed. In any case you will want to properly heat the electrodes to remove the atmospheric 'impurities' under vacuum. This can be done with an RF induction heating coil.


>3.  Would there be any harm in leaving sand inside the tube when it is sealed
>up again?

If it is fairly clean, no, but the entire tube must be heated to approximately 250'C.  It might create some interesting crackling effects if you add a lot of sand.


>4.  Is the gas inside a Nixie similar enough (pure neon, or some kind of a
>mixture?) to that used in neon signs that its function would not be altered?

I'm pretty sure you could use a normal neon. The originals probably mixed with either argon or krypton to help reduce the ionzation potential.


>5.  How critical is the gas pressure to the Nixie's proper operation, and could
>a neon sign shop achieve the correct pressure?  What exactly is the pressure
>inside a Nixie tube anyway?

Probably a higher fill than that used for neon but still within range for a neon shop. I'm totally guessing here but you might try 40 torr.  Connect the power supply to the tube while filling and vary the fill pressure until you are close to the normal operating voltage.

Well, some of this seems useful, especially that last bit about finding the correct pressure by reading voltage drop across the tube as it is filled.

Any second opinions on opening the tube and splicing on a tubulation?  These tubes have no visible tubulation nipple, though there may be one under the bakelite base.  Of course if the bakelite can't withstand the heating necessary to rejuvenate the tube, the point is moot since I would then be forced to remove the bases anyway.  See the next quote for some possibilities regarding base removal.

I think I've decided to try to maintain as much as possible the original appearance of the tubes so I will not be leaving sand inside.

I don't have a clue about using an RF induction coil to heat the electrodes so I think I'll stick with the overcurrent method mentioned in a previous post by James.  These Nixies have a listed normal current range of 10-16mA (presumably only applicable to the normal condition of one cathode in use at a time), so I wonder how much current it would take to sufficiently heat them up?  And what harm could be done to either of the electrodes by this overcurrent?  How about heating all the cathodes at once, which would put an extreme overcurrent into the common anode?

His opinions about pressure and gas composition are different from those I got from Don Klipstein though Kenny admits he's only guessing.


A few ideas regarding tube base removal:

       

quote:Originally posted by Bob Weiss:

I have had good luck immersing the entire bottom of the tube in a container of acetone overnight. Softens most of the old basing cement right up. Doesn't hurt the bakelite base, but it may remove any paint labels, though.

       

quote:Originally posted by Martin Ackroyd (martin.ackroyd@virgin.net):

Alcohol (industrial ethanol)["methylated spirits" in Britain, dunno what it's called in USA] has always worked fine for me.  Plus use of a soldering iron to unsolder the wires.

Once the glass has been removed, heat each pin with the iron, then blow into the other end of the base to clear the solder from each pin.

       

quote:Originally posted by Ralph Day (gecko@gypsytejas.com):

A trick to make unsoldering the pins a litttle quicker and easier is to make a tube desoldering tip for your soldering gun. Replace the gun tip with a bare piece of copper house wire bent in a loop to wrap around the pins and touch all of them at once.  After you've used your solvent to loosen the base glue, wrap this tip around the pins and fire it up.  When the solder melts in all of the tips, pull the base off.

So for base cement release I have ethanol and acetone (perhaps isopropyl alcohol would work also) as solvents.  And the wire loop in the soldering gun trick might be just what I need to unsolder the connecting wires from the pins.  #14 solid copper should be about right.

-----------------------------------------

Alan "A.J." Franzman

Email: a.j.franzman at verizon dot net

-----------------------------------------

[This message has been edited by Alan Franzman (edited August 19, 2001).]

[James]

Hi Alan,

The solvent used in the basing cement during manufacture is industrial alcohol (methylated spirit).  If you soak the base in this for a few days it will soften slowly and you should be able to remove it.

To remove the solder, I usually heat all the pins up simultaneously by passing the flame of a small hand torch over them.  When its all molten you can then just pull the whole base off.

This will give you access to the hidden exhaust tube inside the base.  You could crack the end off that using the molten pyrex rod method I mentioned earlier, and attach a new tube to it.  Then rinse out the tube inside with acid or the metal chippings.  But before you pour in any powder, make sure the contour of the glass around the base is such that you'll be able to tip them out again!  Some tubes have a re-entrant shape glass neck which would prevent this.

I like the idea of lighting the tube while you fill it with gas to get the pressure right, definitely do that if you can!

After its finished, you'll need more base cement to stick that back on.  Don't just use any ordinary adhesive because that may exapand and contract at a different rate and crack the glass.  2-pack epoxies are usually okay, but if you want to do it properly, find a lightbulb factory near you and ask them if they can stick a lump of base cement in the mail to you.  I think your closest lamp plants are probably GE at Mattoon, IL, and Philips at Salina, KS.

[Ed Covington]

You guys sure are wordy! I shouldn't be replying because, frankly, I haven't read all your comments; however, that never stops me. I have not heard of putting sand into a lamp but I am aware of the use of tungsten powder to scrub lamps. A colleague of mine, Roy Leighton, now deceased, patented this idea along with the use of collector grids above the filament for use in motion picture lamps. Indeed, he was given a special Oscar for this invention.

[Alan Franzman]

I did a web search and found out what "methylated spirit" or "industrial alcohol" is called in the U.S. (Thanks James and Martin).  On this side of the pond we call it "denatured alcohol".  What it is, is ethanol mixed with methanol, benzene, or another solvent.


------------------

Alan "A.J." Franzman

Email: a.j.franzman at verizon dot net

------------------

[This message has been edited by Alan Franzman (edited August 19, 2001).]

[Alan Franzman]

Long time between posts, but...

I have obtained some basing cement and removed the base of one of the B-7032's using denatured alcohol and a loop of solid copper wire inserted into a soldering gun to heat all the pins of the tube base simultaneously.  Interestingly, the base did not seem to be attached using lamp base cement originally.  It appeared to be some kind of a rubber material, transparent and light amber in color.

Reinserting the tube leads into the base pins after repair may be a bit involved but I think I can do it by soldering a length of magnet wire onto each lead, then threading each of those "guide wires" through the pins and then keeping tension on them all while pushing the base back onto the bottom of the tube.

One possible problem area I see is the size of the tubulation nipple and thickness of the glass there.  I think the "score and snap" method will either break the whole tube envelope or open a very large hole (about 1/2 inch diameter) where the nipple is now.  I think a better method may be to heat and stretch this thick-walled tip of the nipple until it becomes more conventional in dimension, then open it.

Now if I can just find someone on these boards, or a scientific glassblower, who is willing to attempt the repairs without charging me more than I think the tubes are worth!

A.J.F.