More Dekatron Foolery

[threeneurons]

I just got a 6909 hi-speed dekatron in the mail, today. I hate hi-speeds, but it was a trade, for a standard speed 6802, so the other guy's spinner would work. But this one is a strange one. My other 6909s and 6910s, have this almost indescribably hideous pinkish/yellowish glow. This one glows almost orange. Not 6802 or GC10B neon orange, but the color of a Japanese DK23 dekatron, with the hint of blue. That's not a problem, because I can look at the lit tube, now, without cringing.

Well it has a strange operational problem. When I stick in my spinner kit spinner, it runs upto a certain cathode, then stops. This is at the preset current ~350uA. If I raise it to ~1mA, the high speeds normal run current, nothing changes. I move it over to my dekatron pendulum jig, and it runs fine, but only at the lower current (~350uA). If I raise it to within its specified operating range, it stalls, at that very same cathode.

For now, I'll let it burn-in 24/7 for a few days, at that lower current, and see if it clears up. Will report back later, if anything changes.

Plus, I still have to check the issues with those two 6879 dekatrons, I got a few weeks ago, when time permits.

[Jon]

How odd.

 

Sylvania or Raytheon manufacture? Is the sticky cathode one of the guides or a main cathode? Is it in the immediate vicinity of the index cathode?

 

Most 'sticky pin' problems I've seen were persuaded to go away if I could just get the tube spinning. Leaving a dekatron running for a day or so seems to sort out most things.

 

Jon.

[Dekatron42]

Does it stick on the same pin if you run it both directions? I usually run them backwards (counter clockwise) as they have mostly been running clockwise.

 

If I get a dekatron that is hard to get to run I sometimes use the guide electrodes as the cathode and the true cathode that is sticking as an anode and let it sit with just enough voltage/current to start the glow for some time, usually not more than a few minutes, leaving the true anode without any connection. Sometimes you can get a good glow and sometimes not, but it has had enough effect on some dekatrons to remove enough of the material that is making it stick. On some dekatrons there are also deposits sputtered onto the ceramic base which surrounds the pins which make it much harder to regenerate them into a working condition as there is no electrode connected there, sometimes dekatrons that have this "tarnish" will work for a while after being regenerated and then stop again.

 

/Martin

[threeneurons]

- Does it stick on the same pin if you run it both directions? I usually run them backwards (counter clockwise) as they have - mostly been running clockwise.

-

- If I get a dekatron that is hard to get to run I sometimes use the guide electrodes as the cathode and the true cathode that - is sticking as an anode and let it sit with just enough voltage/current to start the glow for some time, usually not more than 

- a few minutes, leaving the true anode without any connection. 

Sticks in both directions. Same spot. Weird is that it sticks at the higher current, but okay at the lower current.

Never tried rigging a "main" cathode as anode. But its novel scheme for "spot cleaning".

[Dekatron42]

I've regenerated somewhere around twenty GS10H's this way and a few 6909/6910/GC10D's, less rare dekatrons I just put in a box for future regeneration, and they have not stuck while they have been running, but I only run them for short periodes like a few weeks at most so I really don't know how they will fare for longer periods than that. Most have been used in rate-meter or batch-counter instruments and were not working when I rceived them.

 

/Martin

[threeneurons]

Its working, properly now. Both at low current (~350uA) and at its specified higher operating current (1.5mA), in both the pendulum and spinner. Never got around to trying the "alternate anode" trick. Just left it  in the pendulum circuit for the last 3 or 4 days, 24/7. I'm keeping that "anode" trick under my hat, just in case. Thank you.

Glow color is still different than the other tubes I have of this type. Its a Sylvania 6909. Green "Sylvania" text silk screened on a black base. The "6909" is on the glass. Other text on the base is "952" vertically, and "PAZ" horizontally. Just in case someone knows the Sylvania date codes.

For reference: The difference between the pendulum and spinner circuits, other than the obvious, of the pendulum switching directions, is that the guide pulses on the spinner are transient, while on the pendulum, the pulse duration on the guides are uniform. 33mS, each for G1 & G2, in succession, and also a 33mS pause, at the main cathodes. Circuits can be found at my site: http://threeneurons.wordpress.com/dekatron-stuff/

[Jon]

Martin's 'anode trick' is indeed a neat one. I've always shied away from reversing the polarity when playing with dekatrons because I read somewhere vaguely authoritative that it rapidly destroys the quality of the cathode surfaces. However that's probably not an instantaneous event, and anyway if the tube is otherwise unusable, there's nothing to lose. Will definitely give it a go next time I find a sticky tube which resists the usual methods.

 

I think the 6909 is probably an early one - my sense is that Sylvania tubes labelled in green ink predate the yellow ink ones. Can't read the date code though, and nobody I've asked seems to have the scoop on it. Or maybe they do and they're not telling. Week 52 of 1959?

 

With regard to the glow colour, I wonder if it's a partial outgassing effect which has effectively changed the gas mix in the tube? The more exotic fill mixes often contained hydrogen to reduce the effecive deionisation time, and this may be preferentially leaking out or reacting with tube contents over time. I have seen something which may be similar - most of my 7978 dekatrons show a orange peach/blue glow, but one has a much more typical neon orange colour. If this explanation is right, then I would predict:

a) the odd-glow colour 6909 cannot be operated as quickly as a 'normal' glow colour 6909

b) ithe deionisation time should be longer in the odd 6909.

 

I've never tried to do this, but I guess you could look at deionisation by striking a glow and then watching the decay of tube current following a sharp step of the anode voltage to below the tube maintaining voltage (eg ground). Would have to be careful that the measurements don't get confounded by stray capacitance differences between tubes. Sounds like a fun job for a rainy afternoon with a scope!

 

Cheers,

 

Jon.

[Dekatron42]

You'll of course have to be careful when doing something that a dekatron is not designed for, as Jon says there is always the risk of wrecking it by destroying the electrodes or accidentally creating a conductive path between electrodes you don't want a path between.

 

Always put a current limiting resistor in the circuit, calculated for the current you expect the gap to ignite at and turn the voltage up slowly and then back somewhat when you get a glow. If you don't get a glow where you expect it it can sometimes have ignited inside the structure as there might be a point inside/between the ceramic wafers where the path forms. Doing this with low lighting conditions also helps as you can then see more easily if you get a glow somewhere else than expected.

 

Trying to regenerate a dekatron is not an exact method but you'll have to rely on what has been described for other valves and also what can be read about how to prime neon lamps and similar elements.

 

I have the same experience as Jon with the orange glow, but then for the Z505S - outgassed ones which almost don't run at all get a more orange glow but ones with proper filling have a blueish/prupleish glow.

[Leigh Klotz WA5ZNU]

I recently built threeneuron's Dekatron spinner and it worked great, but when I picked it up, the tube (Hivac GC10B) got stuck.

Remembering some previous work with delivering RF into Neon bulbs to kick them up to the next energy level, I used my 2-meter ham radio (2 watts) right near the stuck pin and it started right up.

I can get it to stick again sometimes by turning the speed down very low, but it starts up again when I turn the speed back up; perhaps there's a slight affect on the current from the speed adjustment?

Anyway, try this trick if yours gets stuck.  It will probably work just as well with an unlicensed 460 MHz FRS (or PMR in England) radio.

Leigh/WA5ZNU

[threeneurons]

Neat, a little RF to energize things. But that means things are a bit marginal, so it may get stuck in the future. I currently attenuate the guide pulse, with R12. At 150K, it limits the pulse to ~130V. But its biased by R16, to about 60V over ground, so the guides idle around +70V, and pulse down to -60V relative to the cathodes. R16 is currently 33K. If you make R16 39K, it should raise the bias ~10V, and see if it still sticks.