Dekatron frequency counter and 6909 issue

[Chris Stephens]

some years back I found and restored a Erie Instrumation Dekatron based frequency counter.

http://www.xymox1.com/Projects/erie_instrumation_model_400.htm

Its a wonderful instrument.

Recently tho I noticed that one of the 6909 was degrading. It was getting stuck and could not reach 100khz operation any more. I checked all the signals to it and all were great. The tube was not performing normally. I know these 6909 tubes have life issues. The counter uses 2 of them, one in the 1's position and 1 in the 10's position. Swapping these brought the unit back to normal operation and careful adjustment of voltages got be to 100% perfect operation up to 120khz.

However..

It now painfully obvious the 6909's are dying. As are all 6909's I believe. I could alter the circuit to use 6802's but at a reduced functional max freq of 5Khz which is a long way from its 100khz. There is no doubt that one day there will be no more 6909's as they will all die.

Im looking to pick up some known good, and maybe really great shape 6909's. I need to replace 1 for sure, and then im going to store away the others for later hoping they will last on the shelf.

How should i store these ? If they are outgassing maybe a pressurized container to keep the gasses in the tube ? If the odd gas is reacting with materials inside the tube maybe placing them in the freezer might extend life ? Maybe both, pressurized container and freezing it ? I realize these are extremes but I REAALLY want to have working tubes in 20-30 years. Maybe longer.

I realize im like the only one who cares about these 6909's as I would think no one now needs to go 100khz but my unique frequency counter.

So maybe someone here has some nice 6909's ? and maybe you guys might know of a way to preserve them for 30-100 years from now ?

Also of course, as always, if anyone happens to have a manual for the counter im VERY interested in it.

[Tristan]

I have one 6909 and have been wondering the same thing. There is a section on gas permeation in the book Vacuum Sealing Techniques by Alexander Roth and from reading it would seem that while the hydrogen can permeate the glass envelope, the low pressure within means that gasses (helium and neon in particular) can more readily permeate into the glass envelope from the outside atmosphere. Lower temperatures will certainly reduce the rate of permeation but increasing the pressure (with compressed air at least) would probably just lead to increased contamination of the internal gas fill with atmospheric gasses. I would think the best way to preserve the fill would be to have the same atmosphere external to the envelope as there is internally (not so easy). Failing that, keeping them cold should help but I wouldn't expect miracles.

[threeneurons]

Chris,

You're in a real pickle. I'd say, "I feel your pain", but I've never been a fan of the 6909. I do, however, understand, that you want your instrument to be fully functional, at its full specs. I really don't know what can be done with a 6909 to extend its shelf life. To do that we need to know how they're really failing. Don't know if its from gas leaking out, or air leaking in. That way we'll what's better. Storing it under elevated pressure, or lower pressure. Cooling will probably slow down any reactions, but by how much. It may only be a few percentage points. And if we're not careful, we might accidentally introduce condensation which will corrode the wires, and the rusting wires may crack the glass seal. There are some 2 guide high speed dekatrons other than the 6909, 6910, and Z505S. I think the A103 and A106 meet that standard. They are a little shorter than the 6909, so maybe they'll work using a socket adapter. The Russians kept making dekatrons well after the west stopped. So that may buy an extra 20 years.

[Nick]

As others have said, the 6909 is prone to leakage - more accurately "ingassing" than "outgassing" as it starts off with a Penning mixture below atmospheric pressure,

The Sylvania ones are certainly problematic, but the Raytheon electrical equivalents (CK6909) are mechanically very different - I wonder if they also suffer the same problems?

Nick

[Tidak Ada]

Isn't it just the loss of hydrogen (helium) that makes the trouble? The absolute pressure inside the tube is lower then outside, but the partial tension of the hydrogen (helium) is higher than atmospheric as me is told....

 

eric

---

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Nick
Sent: vrijdag 21 juni 2013 10:56
To: neoni...@googlegroups.com
Subject: [neonixie-l] Re: Dekatron frequency counter and 6909 issue

As others have said, the 6909 is prone to leakage - more accurately "ingassing" than "outgassing" as it starts off with a Penning mixture below atmospheric pressure,

The Sylvania ones are certainly problematic, but the Raytheon electrical equivalents (CK6909) are mechanically very different - I wonder if they also suffer the same problems?

Nick

--
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/cdef96e1-09f7-440d-ab13-f2b20d355c9e%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Nick]

On Friday, 21 June 2013 11:35:13 UTC+1, Tidak Ada wrote:

Isn't it just the loss of hydrogen (helium) that makes the trouble? The absolute pressure inside the tube is lower then outside, but the partial tension of the hydrogen (helium) is higher than atmospheric as me is told....

Whilst many fast dekatrons contain hydrogen (and some helium) I'm not sure the 6909 does - its glow is very orange, indicating a standard-ish Penning mixture with no hydrogen - also, I can find no documentation stating that this tube contains hydrogen - it is, however, a reasonable assumption that it *might* as its a "higher speed" (for a dekatron) tube...

Hydrogen gives a pink-ish glow... not orange. The only real way to be sure would be to find a good 6909 and do a spectral analysis of its glow... I'm sure we have members who could do that...

I'm still interested in the relative performance of extant Sylvania vs Rattheon tubes...

Nick

[Tidak Ada]

That will be difficult, because of lack on the  (expensive) equipment..

I have here a Sylvania 6909 and a 6910 as well the Raytheon equivalents. Also two Sylvania 7155's. Should be fine if I could test them.

Tha tubes are for the collection, except the 7155's. They are bought to make a clock divider.

 

eric

---

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

Sent: vrijdag 21 juni 2013 12:55
To: neoni...@googlegroups.com
Subject: Re: [neonixie-l] Re: Dekatron frequency counter and 6909 issue

--

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/d8bd589a-62a7-46c5-a249-7755b6b04864%40googlegroups.com.

[Chris Stephens]

Wow great responses.. I have to run to work. I will be back later and comment more..

The spectral emissions is a interesting thing :)

So im also after some 6909's. As they are not prized because of pinkish glow maybe you guys have some you might want to part with :)

Working a Russian substitute into the instrument is a great idea. Im interested in doing that.

Gotta go, be back later..

[threeneurons]

Here's a photo of one of my Sylvania 6910 dekatrons, glowing:

The color is the pink/orange on the fringe. That's the color of 3 of my 6909s, both of my 6910s, and a Z505S, when it was working. I did get another 6909, as a swap for a 6802, in the last year, that has a glow that is more of a standard dekatron's orange. Its still a little more yellowish, than the standard speed orange, but closer to that orange, than to the pinkish /orange,in this photo. 

I did try a simple spectral analysis with a cheap plastic spectrometer, with inconclusive results. The gas mix has lines all across the spectrum. I need at least a spectrometer, of the grade, that I used in college, so I can get some actual measurements of each of those lines. These tubes definitely use some complex gas mix.

As you can see its not the blue/violet that you see in 3-guide ("single pulse") tubes, like the OG-3 & GC10D. 

[Dekatron42]

The only working specimens of Z505S that I own have a purplish/blueish glow with very little orange in them. The ones that have more orange only counts below 50KHz and not well at all, they count erratically. The other ones count perfectly and also a bit beyond 50KHz.

 

My 6909/6910 from various manufacturers also have slight blueish glow to them but are more orange than the Z505S's. all of my 6909/6910 count up to the specified 100KHz and a little more if the surrounding circuitry is trimmed.

 

It would be interresting to check magazine articles on these Dekatrons as you usally get a lot more information from those than from manufacturer datasheets/manuals, especially when it comes to details the manufacturers themselevs didn't want to reveal.

 

/Martin

[Chris Stephens]

I tell you what. Its nice to have this group !. Where us could such geeky neerdy guys like us hang out !

Hmmmm.....

Well ive a 6909 coming from ebay, who knows what state it is in. Im going to buy a number of 6909's as I can find them.

As far as preserving them, hmmmm.... Lots of good discussions here. So as we dont know the pressure in the tube its impossible to say what pressure outside the tube would be best - even if that concept was good at preserving it. Do you think keeping them in a easy gas to obtain might help ? Like nitrogen ?

So it seems like cold might be a good idea ? maybe a freezer bad with a bunch of desiccant gel packs to reduce the moisture ? or would the desiccant also absorb the gas mix ?

You know, cold will shrink the tube glass ever so much and might change things and allow more gas in/out.

Man this is a difficult topic. A good one to know tho as given 100 years all the dekatron tubes, and neon tubes, will be a issue. So good to know to preserve these things as im sure they are fading, if not now for sure in another 50 years.

A bit off topic. Ive been playing with the counter today. I have a rubidium freq source from ebay and hooked that to a modern ( 70's ) HP freq counter as its external freq source. I was calibrating the Dekatron freq counter. Amazingly accurate. BUT I cant trim out that last little bit because i run out of trim room. The parts are all in spec, its the crystal aging. Over 50 years it aged out of the trim range. I needed to add a 22pf cap to get it back in range. But it reminds me this gear is decaying. A week ago I replaced the electorlytic caps with wet tantalum as they dont age much. But even carbon comp resistors do age. I think at 100 years old this counter is going to start to reach a point its just not going to be serviceable without total overhaul of all the original parts. Which takes the fun outta it. Except for the caps its currently all orginal parts.

I was testing today and it can count at 120khz and not miss one count, over and over and over. The counter has a built in check, and it counts for EXACTLY one second and shows what it counted, so for 120khz you can get a display of (1)20000 over and over and over. Never (1)20001. Dekatron tubes are actually amazingly accurate as long as the circuits around them are well designed and trimmed. Yea I gotta adjust voltages during calibration for best performance at the highest freq and this needs to be done when you replace the ones tube.

Im letting the TXO warm up and then im going to calibrate it against the rubidium source and after I do that I will do a vid showing how it does not miss counts and is amazingly accurate.

[Jon]

This is fantastic - the group is relatively quiet for weeks, and suddenly we get this brilliant thread on a most interesting subject! :)  So, some thoughts...

Chris might want to approach ebay seller electroware regarding 6909 supplies. He has sold quite a few of them from some kind of bulk supply box - they seem to be NOS, but not boxed, and usually go at reasonable prices. He is also aware of the problem of duds with this tube type and usually offers them with a guarantee not to be DOA - my guess is he's doing a simple striking test.

 

With regard to the gas fills in high speed tubes with orange glows, I have not come across any definitive statement tying a particular mix to a tube model. There is literature discussion of a Ne/H/Ar mix - my guess is it's something like that here, but as Mike said, we need a spectrometer to be confident. Hydrogen seems to be a common feature of high speed dekatrons as it reduces the deionisation time of the gas and allows higher frequency stepping - the fastest tubes are pure hydrogen (blueish glow).

 

A couple of observations on orange high-speed dekatrons (by which I mean anything quicker than 4kpps). First, most show a blue/purple glow associated with the anode. You can just make it out round the rim of the anode disc on the 6910 photo which Mike posted, but it's most clearly seen on auxiliary anode tubes (GSA10G, Z572S). Here's one of mine in action - the blue glow is around the auxiliary anode pins.

 

Secondly, although many of these high-speed tubes have one or even two standard D shaped getters, I have never ever seen the standard mirror of flashed barium on the wall of the tube. Anyone got any experience that runs counter to this? From this observation  I've been thinking in a couple of directions. First - perhaps the gettering material used in these tubes isn't that standard stuff, even though the getter units look absolutely the same as on neon tubes. Quite possible. Second option, the tubes as manufactured did have standard getter flashes, and over time these have been consumed and lost.

 

For a number of reasons, I'm currently leaning towards the second explanation. Not least because it ties up with the observation that all these tubes have a significant dead rate now, regardless of manufacturer or model - something you might expect if there was some process at work which was slowly (over decades) exhausting the getters and then poisoning the tube. I don't think we're looking at gas leakage in the sense of the envelope or pin seals failing though. More likely it's chemistry going on - hydrogen reacting with something in the tube.

 

To Nick's question about Raytheon 6910/6909 and whether they were any better than Sylvania's flavour... A couple of years back I did think the Raytheons were better, but looking at my collection now the numbers seem to have evened out. They're both running about 50% dead :(. The Mullard Z505S is the biggest nightmare though - a large majority of them seem to be dead or dying. Good job Chris' lovely counter hasn't got those in!

 

Cheers,

 

Jon.

[Chris Stephens]

Isnt this all so much fun :)  we are truly geeks.

I would think that if we take a pic using a digital camera with NO OTHER LIGHT of the tube glow we would pull that image into something and get a spectrum from the software ? Not exactly a spectradiometer but this might provide clues to the spectra and gas mix.

Hmmmm..... getters.... Thats a whole different subject.. And how does cold effect them.

Yea my 6909's have no getter. None what so ever. Good point on that one.

Indeed I will contact that ebay seller. Thanks for the tip :)

Ive been playing with the counter more.. I have fixed the issue where I could not trim out the main osc properly. Required a 22pf cap of extra range to trim the aged crystal. I was able to trim it to 100000.050hz. It drifts around .01hz after a 30 min warm up. That seems quite good. My brain is failing me to determine ppm drift for it but its impressive.

Good to have it calibrated against a rubidium atomic standard.

Thats the actual osc circuit. It then divides it down to 10khz, 1khz,100hz,10hz,1hz. It does this in a analog domain using some kind of harmonic process I dont quite understand. Each 5965 tube somehow divides the input by 10. These produce perfect division with the 1 hz at the end of the chain of 5965's being 1.00000050  This one hz reference is the BIG deal as its the /sec reference that the counter uses count incoming cycles to. IE cycles per sec.

BUT a error is creeping in. The counter tubes, the 6909's in question, is missing a few counts. So in the self test mode about 10% of the time it misses 3 counts out of 100,000.. Thats counting to 100,000 in 1 sec mind you and it sometimes misses exactly 3. It misses these exact three, rarely, all the way out to 120khz where then it just falls apart.

Im letting it cool off and im going to do a little work on the TXCO ( oven based crystal ) because its case is a little loose.

Remember its using 5 dekatrons to count to 100,000 in 1 second and it almost never misses one count. Dekatrons are stunningly reliable counting devices. Below 100khz it seems 100% accurate.

I will try using a camera and see if I can catch a spectrum.

[Dekatron42]

Maybe the 5965's are hooked up as Phantastrons?

 

/Martin

[Chris Stephens]

The 2 6909's i have now are different.

I have a Sylvania 6909 and a Raytheon ck6909. The Raytheon is failing and will not clock at 100khz. The Sylvania works out to 120khz and sputters out and stops at 145khz.

They are driven by 5964's.

Yea I need a set of NOS 6909's that are the same but its that first one that sees the real speeds. The tens digit will only see 10khz max.

[threeneurons]

The 5965 is the industrial version of the 12AV7 dual triode. Same pinout as the 12AT7, 12AU7, and 12AX7. Its probably hooked up as a one-shot, which is basically what a phantastron is. Non-retriggerable one-shots and phanastrons can be used as timebases. Not really dividers, but will sync with a reference signal, so they can be used in place of a divider chain.

Yes, we are a bunch of geeks ! I love seeing so many people interested in dekatrons. They are such a unique class of tube. And its nice to see that GCA10G in full glow ! I saved that photo !

[Chris Stephens]

"Phantastrons"

What a awesome name. I believe that is exactly what the circuits are.

I got in my new 6090. Its a Sylvania 6909 but has different layout of logo and has PAD on it. It looks completely NOS but no box.

I put that into the ones posistion as I can fully test its top counting speed there. Right away I reached 181khz before its sputtered out. It was of course not reliable at that speed. It does look like I can hit 150khz now reliably but I will burn it in and test more. I will determine a top speed where it does not miss a single count. It appears how fast it can count is related to its eventual failure. So as it fails it gets slower.

I moved my other good sylvania 6909 into the 10's position where its max speed will need to be like 15khz.

I still need to adjust the voltages for the new tube, but this is a impressive top speed right away.

I also need to correct something i said in a previous post. There is getter on the Raython tube, the bad one, its all around the counting pins on the inside of the tube. The Sylvania tubes, the good ones, do not have any anywhere on them.

[chuck richards]

Hello Chris,

I've been reading all about your very neat dekatron
freq. counter! The pending troubles with the aging 6909s
is an interesting one, and congratulations on your recent
Ebay purchase of a fast 6909 for the "ones" position.

This reminds me somewhat of a much later freq. counter
I have that I bought at a hamfest some years ago.
It was outside, in the parking lot, at the huge Orlando, FL
hamfest. This guy had tons of old test equipment stacked
on the ground and I spotted a few pieces that had nixie
readouts in them.

I bought a few of those pieces basically for scrap value
and paid $2.50 each for a few of the smaller pieces of
test equipment, and I also paid a hefty $5.00 for this
CMC 12-digit nixie tube freq. counter.

Originally I was just going to scrap the counter to get
the NL-5853 nixies, and their old Fairchild driver ics.
But, before I started scrapping it, just for the heck of it
I powered it up, and, amazingly it started working straight
away!! It's only problem was that somewhere along the line
it turned up missing its left-most tube in the readout.
That was an easy fix, I just dug out a NOS NL-5853, and
put it into the socket. The driver ic was still there.

This freq. counter, made by CMC, looks to be made sometime
in the early 1960s. It has a crystal oven, and once warmed
up, it is amazingly accurate! Needless to say, I never had
the heart to scrap it. It works!

It only triggers and operates reliably though if it is
fed a nice square wave. It does not do very well with
a sine wave, nor much of any other waveform, for that matter.

Anyone out there know if this is a common trait of these
old counter-timers?

Great work on the dekatron frequency counter though!
You are very lucky and fortunate to own it.

Chuck

>
>
>---- Original Message ----
>From: pub...@xymox1.com
>To: neoni...@googlegroups.com
>Subject: RE: [neonixie-l] Re: Dekatron frequency counter and 6909
>issue

>>--
>>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/5535459e-2ed2-4d1e-aa15-
>833f94b780f7%40googlegroups.com.

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

$4.95/mo. National Dialup, Anti-Spam, Anti-Virus, 5mb personal web space. 5x faster dialup for only $9.95/mo. No contracts, No fees, No Kidding! See http://www.All2Easy.net for more details!

[Chris Stephens]

Ive done a new video.. Showing it in operation. The tubes are clearly different color even tho they are both sylvania's. Its counting perfectly at 150khz. I will test further to see what the upper limit is, but I think its around 165khz. This tube is a LOT better then the other one..

http://www.youtube.com/watch?v=HIVz9bH6gL4

[Chris Stephens]

Chuck... I just love vintage test equipment. I have lots of 70's - 80's gear. Its cheap and you can get really good gear. I LOVE my Tektronics 465B scope. I also have a tube based Tek as well that works perfectly, but it just gets too hot to use all the time and the 465B is a better scope.

Yea I just love vintage test equipment. I have a fully functional tube based VTVM from the 1940's..

Pics of some of my fully functioning vintage stuff..

[Chris Stephens]

OK Ive had a very interesting thing happen.

I got the new 6909 nas i mention above and all was good.

The bad tube and a spare i have both light up and count just not fast like they should, even tho they worked like a year ago. So some decay is happening..

All fine and good..

So I was using the slightly slower tube in te tens digit. I did a lot of use. Like prob 20-30 hours over 3 days. All worked perfectly.

It was running and spinning away and suddenly i noticed that the tens digit was not even lit up.. I investigated. I used my spare 6909 which i had also been using over the previous 3 days as well testing to see which was faster. So it was known good. No luck. So it seemed my counter had developed a issue. I spent a bunch of time with test gear and checked everything around that tube and nothing was wrong. The 520Vdc before the 270k resistor was correct BUT there was no drop on the other side. The tube was not drawing any current.

So I was stumped. Both tubes could not be bad..

So i decided to try my new tube in the socket. IT WORKED.. Back to the other 2 tubes, they did not work..

SOMEHOW BOTH THE TUBES HAD DIED AT THE SAME TIME. Even tho one was not even in the unit. There is no way they both could have died at the same time !.. But indeed further testing revealed both were now dead. No glow at all. When the day before both had glow. INSANE how could this be ?!?!

Here is my guess and it might explain some things..

Maybe my use of the tubes set off a reaction in the gasses and materials in the tube that resulted in both tubes dying at the same time. I started using them both at the same time 4 days before. Even tho one got WAY more use maybe just the act of using them could have triggered some complex chemical reaction that took so many days to complete. Turning on a tube does ionize the gas. Anything in gas will also be ionized. So maybe after 40 years some chemical of something in the tube had mixed with the gas. When i started using the tube this ionization caused some complex reaction that ultimately corrupted the gas in the tube. It started a reaction that ended with some gas being depleted or some reacted gas that wont work. Maybe it deposited some film on the electrodes. I dont know.

As crazy as that sounds..

I just cannot explain what happened any other way. Both tubes dying on the same day even tho one got almost no use and the other got tons of use BUT both got used for the first time in years 4 days before.

So this led to a really stupid, but fun event.. I wondered if there was any neon gas left in the tube. Or ANY gas in the tube.. I applied lots of voltage. Nothing. Tried some HV from a CRT TV, nothing.  It was like there was not any gas left.. So in a moment of complete stupidity... I decided to pop it into the microwave.. The tube was toast. Why not.. If there is any gas in there the microwave is sure to ionize it.. It WILL light up...

Wow.. Dont try that at home folks..

So the microwave kicked on. The tube IMMD glowed like a supernova. Not orange, but mostly white. Then something happened inside the tube, most likely arcing.. Then the most amazing thing happened. It changed color to kinda a purple and then the top of the tube popped off, cracked all around the getter. There was little if any pressure difference so nothing really popped. It kinda just cracked and fell open. Then the most amazing thing occurred. The gas escaped as a cloud and STAYED LIT AND IONIZED. It held together for a few seconds and made this awesome 60hz kinda deep hum. Quite loudly in a menacing way. It was not the microwave, it was the gas humming obviously acting as a speaker being modulated by the microwave under duress with a 60hz hum in the signal. It was all kinda a purple cloud. It was rather unexpected and scary so i shut it off in a scramble. But OH MY what a display.

So obviously there was still some gas in the tube. But for some weird reason I could not get it to glow even providing lots of voltage. Of course the microwave could have literally ignited some of the gas. No doubt if there was hydrogen in there when it arced there would been havoc.. But there was something in there. It stayed ionized even outside the tube.

It was awesome... Im going to explore the remaining tube that does not work. Im interested to know what you guys think and any tests I might be able to do to test it to isolate whats going on with the decay of these tubes... In the end tho, I suppose i will have to microwave it and this time use a camera to get the gas plasma fusion ignition. It was pretty impressive. The hum from the gas was awesome. I know you can use a gas as a speaker, Plasmatronics did it back in the 1980's...

But yea... SOMETHING odd happened inside the tube. Im thinking its some sort of chemical reaction I set in motion by using them.. This could be bad.. It might mean all 6909's will suffer this same fate. So use might be bad..

[Chris Stephens]

Further research...

Taking the tube apart I think I found something...

The ceramic spacer that separates the pins has a odd coating on the top side. Its conductive. Like a thin film resistor. Its 1 Mohm from side to side across any section of it. Its some thin metalized film deposited on the ceramic. This seems odd to me considering the nature of whats going on here with the pins being rather electrified. But if you look none of the pins actually touch this part of the ceramic spacer. But the entire surface has a conductive coating of some kind. This also appears true on the 6802's as well.

http://www.xymox1.com/Misc/Dekatron/IMG_0306a.jpg

Now the interesting part is the film seems to be decaying or changing.. Its got a crackly surface while the ceramic is super smooth. You can see that in the above image. In fact the new 6909 I have has FAR less of this crackly surface. Much smoother.

Close up http://www.xymox1.com/Misc/Dekatron/IMG_0306.jpg  close up 2 http://www.xymox1.com/Misc/Dekatron/IMG_0311.jpg

After trying quite hard i could not scrape any off. So its very well deposited onto the ceramic. Its not flaking off so these crackly patterns are not caused by that.

So im thinking maybe a few things..

1. The gases are interacting with the thin film and maybe then when the tube is powered this surface gets some type of charge and some very fine dust from the corrosion of the thin film mixes into the gas. Maybe even there is some small amount of air movement inside the tube when its operating and this moves this conductive dust around. Then maybe this accumulates someplace and causes the tube to not work.

2. Maybe something in this material is becoming vaoprized by the electrical currents flowing thru the film as the counter dot spins around creating very tiny arcs inside the film and over time these vaoprize this material and again there is some deposition on something else and this causes the tube to fail.

Looking at the other failed tube I can see FAR more crackly surface then on this one. This tube was worse performance. So the crackly surface might indicate its decay state. I also noticed much more pronounced and longer connected cracks in the crackly surface. Thet started from pins and went radially outwards. Like arcs would away from a pin. There is no doubt some current must end up in this film from each pin. So maybe the current takes the same path over the years and etches out these cracks in the surface all the while vaoprizing the conductive material that then mixes with the gas and moves around the inside of the tube ending up in the wrong places. Using the tube and electrifying it might cause a vast amplification of this process and a few days of settling later might cause the tube to not work.

These are of course just gueses..

All the other parts of the tube look fine. I doubt the mica spacers are a issue and the metal bits dont seem likely to cause a issue.

I suppose the ceramics might be absorbing something or outgassing something but this does not explain the very short period of operation i saw. Vaporizing this conductive film in use and then having it move around and redeposit might well explain what i saw..

So... MAYBE I can measure something on the remaining tube that might show more then normal conductivity.. But I dont have a easy tester of conductivity at high voltages..

[John Rehwinkel]

> The gas escaped as a cloud and STAYED LIT AND IONIZED. It held together for a few seconds and made this awesome 60hz kinda deep hum. Quite loudly in a menacing way. It was not the microwave, it was the gas humming obviously acting as a speaker being modulated by the microwave under duress with a 60hz hum in the signal. It was all kinda a purple cloud. It was rather unexpected and scary so i shut it off in a scramble. But OH MY what a display.

Yeah, a microwave oven can keep a plasma ionized. There are many ways to start such a plasma, and you found one. The most common one is to put
a small lit candle in the microwave.

Check out

http://www.youtube.com/watch?v=vgiaKTEZBCM

Or just search for "lit candle microwave" or "microwave plasma balls" on youtube.

- John

[threeneurons]

Just to remind you, that dekatrons, and other cold cathode tubes (nixies) are subject to sputtering, where the ions knock off bits of metal from the cathodes, and they deposit on any surface. Its actually used as an industrial process in vacuum deposition:

http://en.wikipedia.org/wiki/Vacuum_deposition

Sputtering is one of several methods used. The coating tends to adhere quite well, as you have found out.

Thing is, this also occurs with the neon filled standard types, and they definitely last much longer. So:

(1) This has nothing to do with the high speed dekatron ills. Or:

(2) The gas reacts with the deposited metal, but only the particular gas, in high speeds.If this reaction occurs, it could occur while the metal is a vapor, moving from cathode to surface.

If one of the gasses used in the 6909 is hydrogen, then it can form a hydride with metals. It will do this mostly while the tube is energized, but I suspect some sputtered metal may stay in vapor for sometime afterwards. Also, the sputtered metal surfaces are probably porous, which will expose a lot hydride formation surface, even when the tube is off.

On Thursday, June 20, 2013 10:43:37 PM UTC-7, Chris Stephens wrote: