Testing nixie tubes
964 views
Skip to first unread message
Michel
Dec 7, 2012, 12:53:36 AM12/7/12
to neonixie-l
I wonder what is the best way to test nixie tubes to be fairly sure
they will work just fine.
I encountered a few issues, but this one is new to me. I had tested a
tube and made sure all segments would work. I put them into my circuit
and after a while noticed my number 2 digit didn't light up anymore. I
thought it would be a bad connection although I couldn't see anything.
I am using cathode side transistors so I soldered 2 digits together
and expected both of them to light up at the same time, but actually,
neither of them would light up. So I am thinking there is a short
between the cathode of number 2 digit and the anode. I took the tube
out, measure with an ohm meter and measure 1.6k ohm!! That explained
the problem, but how can that happen?
Michel
they will work just fine.
I encountered a few issues, but this one is new to me. I had tested a
tube and made sure all segments would work. I put them into my circuit
and after a while noticed my number 2 digit didn't light up anymore. I
thought it would be a bad connection although I couldn't see anything.
I am using cathode side transistors so I soldered 2 digits together
and expected both of them to light up at the same time, but actually,
neither of them would light up. So I am thinking there is a short
between the cathode of number 2 digit and the anode. I took the tube
out, measure with an ohm meter and measure 1.6k ohm!! That explained
the problem, but how can that happen?
Michel
threeneurons
Dec 7, 2012, 1:00:07 AM12/7/12
to neoni...@googlegroups.com
You just got unlucky. The metal that sputters off the cathodes, formed crude metal film resistors; partial short.
David Forbes
Dec 7, 2012, 1:13:11 AM12/7/12
to neoni...@googlegroups.com
turn up only after the watch is built and run for a few days.
The inter-electrode clearance inside these tubes is rather small, so
shorts can happen.
I don't think there's any way to test for such problems in advance.
That's why I use sockets for the tubes in my watches.
--
David Forbes, Tucson AZ
Michel
Dec 7, 2012, 2:23:12 AM12/7/12
to neonixie-l
Some time ago I read that it would be good practice to let each
cathode lit at maximum rated current for about 2 minutes, I can't find
that page no more so I am not too sure about the reasoning behind it.
If metal sputters are a cause of the problem, it could more likely
show up in these first 2 minutes of maximum drive I guess.
Putting the tubes in sockets would be an advantage in this case. I am
actually surprised how easy it is to change a faulty (or broken) tube,
it takes less effort than I had expected. It's quite easy to heat up
all the legs together, take the tube out and then use some wick to
empty the holes. Crucial here is to use a solder that has lots of
flux. My board setup is different from yours, so sockets would simply
not fit.
Michel
cathode lit at maximum rated current for about 2 minutes, I can't find
that page no more so I am not too sure about the reasoning behind it.
If metal sputters are a cause of the problem, it could more likely
show up in these first 2 minutes of maximum drive I guess.
Putting the tubes in sockets would be an advantage in this case. I am
actually surprised how easy it is to change a faulty (or broken) tube,
it takes less effort than I had expected. It's quite easy to heat up
all the legs together, take the tube out and then use some wick to
empty the holes. Crucial here is to use a solder that has lots of
flux. My board setup is different from yours, so sockets would simply
not fit.
Michel
Message has been deleted
Dekatron42
Dec 7, 2012, 3:24:28 AM12/7/12
to neoni...@googlegroups.com
I've had problems with swarf and small cut off pieces of electrodes inside some Nixies and also Dekatrons, this has mostly happened tubes manufactured by Mullard. I also read on another forum that this happened to a lot of the Anita Calculators which used Mullard manufactured ZM1080's, here: http://www.vintage-radio.net/forum/showthread.php?t=73142. In some Dekatrons the loose pins could be shaken around, but in some cases I had to hit them hard, after they had formed a short to get them loose again. Sometimes there were burn marks left which meant that there was an occasional flash over and in some cases there was a permanent path for the current to pass through. I've had this problem with Z504S and Z505S plus the ZM1050/Z550M.
/Martin
Michel
Dec 7, 2012, 3:44:12 AM12/7/12
to neonixie-l
I tried the trick with the back of a screw driver as mentioned in your
link, and indeed this works! The short is gone. Not sure what has been
the reason, could still have been some metal sputters I suppose. I
don't see anything floating around inside the tube, but of course it
could also be too small to see....
Michel
On Dec 7, 7:24 pm, Dekatron42 <martin.forsb...@gmail.com> wrote:
> I've had problems with swarf and small cut off pieces of electrodes inside
> some Nixies and also Dekatrons, this has mostly happened tubes manufactured
> by Mullard. I also read on another forum that this happened to a lot of the
> Anita Calculators which used Mullard manufactured ZM1080's, here: *http://www.vintage-radio.net/forum/showthread.php?t=73142*<http://www.vintage-radio.net/forum/showthread.php?t=73142>.
link, and indeed this works! The short is gone. Not sure what has been
the reason, could still have been some metal sputters I suppose. I
don't see anything floating around inside the tube, but of course it
could also be too small to see....
Michel
On Dec 7, 7:24 pm, Dekatron42 <martin.forsb...@gmail.com> wrote:
> I've had problems with swarf and small cut off pieces of electrodes inside
> some Nixies and also Dekatrons, this has mostly happened tubes manufactured
> by Mullard. I also read on another forum that this happened to a lot of the
Dekatron42
Dec 7, 2012, 8:29:24 AM12/7/12
to neoni...@googlegroups.com
Michel, I am glad that it worked for you to hit it with a screwdriver!
I have one Z504S which has four small pieces of wire inside the glass envelope, the longest one is 3mm and the shortest one is 0.5mm - I bought four Z504S's on eBay and they all had a few loose wires inside. These only work upside down as the wires then collect at the top of the glass envelope, any other way and the wires fall down between the pins and result in shorts or "neon" lamps if they fall in the wrong place.
/Martin
Michel
Dec 8, 2012, 6:25:32 AM12/8/12
to neonixie-l
Interesting Martin, it seems to me that after manufacturing, these
nixie tubes did not undergo the same failure tests as modern
electronic parts. I find quite a few tubes that have a piece of one
cathode segment not lighting up. Is that because they're so old or has
that never been tested when they were packed?
Michel
nixie tubes did not undergo the same failure tests as modern
electronic parts. I find quite a few tubes that have a piece of one
cathode segment not lighting up. Is that because they're so old or has
that never been tested when they were packed?
Michel
threeneurons
Dec 8, 2012, 10:58:31 AM12/8/12
to neoni...@googlegroups.com
"Sleeping Sickness". A form of cathode poisoning, from not being powered for some time. It appears that sitting idle for too long, effects the surface of the metal cathodes. This effects the work function, and they light up, unevenly. Its a problem for all cold cathode tubes. Did you run them for a few hours, or so, at elevated current ? This usually cures the problem. But not always.
... a few tubes that have a piece of one cathode segment not lighting up.
... a few tubes that have a piece of one cathode segment not lighting up.
Quixotic Nixotic
Dec 8, 2012, 1:56:20 PM12/8/12
to neoni...@googlegroups.com
On 8 Dec 2012, at 15:58, threeneurons wrote:
> "Sleeping Sickness". A form of cathode poisoning, from not being powered for some time. It appears that sitting idle for too long, effects the surface of the metal cathodes. This effects the work function, and they light up, unevenly. Its a problem for all cold cathode tubes. Did you run them for a few hours, or so, at elevated current ? This usually cures the problem. But not always.
I have today run a pair of new Burroughs B5092 round top view jobbies and they both had cathodes that refused to light. In one or two cases the legs would glow instead of the digit. A couple of digits looked plain dead. After poking about lighting various digits they have all woken up and started to fire normally, which is a relief. Trying to fire them later they were again sluggish for the first few attempts to light. I am sure they will start behaving better once they are run in.
> "Sleeping Sickness". A form of cathode poisoning, from not being powered for some time. It appears that sitting idle for too long, effects the surface of the metal cathodes. This effects the work function, and they light up, unevenly. Its a problem for all cold cathode tubes. Did you run them for a few hours, or so, at elevated current ? This usually cures the problem. But not always.
John S
Michel
Dec 9, 2012, 6:34:13 AM12/9/12
to neonixie-l
I haven't run them yet on elevated currents, I'll give that a try
later on. At the moment it is easier to just replace the tube and give
it a mark.
I also have 1 tube that shows another interesting phenomena. I can
light up the decimal point with any digit except number 1. The decimal
point fades to almost invisible when I light digit 1, while with all
others it shows just fine. I find it particularly strange because
digit 1 is I think the furthest digit away from the decimal point.
Michel
later on. At the moment it is easier to just replace the tube and give
it a mark.
I also have 1 tube that shows another interesting phenomena. I can
light up the decimal point with any digit except number 1. The decimal
point fades to almost invisible when I light digit 1, while with all
others it shows just fine. I find it particularly strange because
digit 1 is I think the furthest digit away from the decimal point.
Michel
glasslinger
Dec 9, 2012, 10:56:12 AM12/9/12
to neoni...@googlegroups.com
Hello!
Try lighting them up using AC for an hour or two. A small sign transformer with a variac works fine. This will fire the segments that don't light up using DC, and will bake off the mercury products that cause the poisoning. Do not run so much current through the segment that it turns red hot though. A few milliamps is enough.
ron
Try lighting them up using AC for an hour or two. A small sign transformer with a variac works fine. This will fire the segments that don't light up using DC, and will bake off the mercury products that cause the poisoning. Do not run so much current through the segment that it turns red hot though. A few milliamps is enough.
ron
Michel
Dec 9, 2012, 8:07:04 PM12/9/12
to neonixie-l
Thanks Ron,
Would you recommend doing this for all tubes or only the ones that
show a problem? I just wonder if perhaps it would be good to do this
procedure anyways, then I can make a test circuit using some triacs.
Michel
Would you recommend doing this for all tubes or only the ones that
show a problem? I just wonder if perhaps it would be good to do this
procedure anyways, then I can make a test circuit using some triacs.
Michel
glasslinger
Dec 10, 2012, 8:53:00 AM12/10/12
to neoni...@googlegroups.com
Hello!
If the tubes are working fine I would just use them. This is a way to salvage finicky tubes that you would be not using. I would bet that these tubes were the result of poor production quality control since there are so many that never develop problems. The thing to remember too is that all current run through a tube uses up some of its ultimate life. Thus whether regenerating tubes or using them in a clok or other project apply only what current is necessary so you preserve as much life as possible. I have found that it is not practical to rebuild nixie tubes. The ones I have rebuilt develop problems after a few dozen hours of use, and I have not found exactly why.
ron
If the tubes are working fine I would just use them. This is a way to salvage finicky tubes that you would be not using. I would bet that these tubes were the result of poor production quality control since there are so many that never develop problems. The thing to remember too is that all current run through a tube uses up some of its ultimate life. Thus whether regenerating tubes or using them in a clok or other project apply only what current is necessary so you preserve as much life as possible. I have found that it is not practical to rebuild nixie tubes. The ones I have rebuilt develop problems after a few dozen hours of use, and I have not found exactly why.
ron
Michel
Dec 10, 2012, 3:28:21 PM12/10/12
to neonixie-l
Hello Ron,
I see, so actually what you're saying is if they don't work properly
straight away, it is better not to use them at all. That's what I am
doing right now :-). I'll still give the AC a go with some of them,
and then see if the problem will return afterwards. It's quite easy to
test it that way, but then I won't need to design a complete test
circuit.
Michel
I see, so actually what you're saying is if they don't work properly
straight away, it is better not to use them at all. That's what I am
doing right now :-). I'll still give the AC a go with some of them,
and then see if the problem will return afterwards. It's quite easy to
test it that way, but then I won't need to design a complete test
circuit.
Michel
David Forbes
Dec 10, 2012, 3:31:42 PM12/10/12
to neoni...@googlegroups.com
On 12/10/2012 1:28 PM, Michel wrote:
> Hello Ron,
>
> I see, so actually what you're saying is if they don't work properly
> straight away, it is better not to use them at all. That's what I am
> doing right now :-). I'll still give the AC a go with some of them,
> and then see if the problem will return afterwards. It's quite easy to
> test it that way, but then I won't need to design a complete test
> circuit.
>
> Michel
I have a couple little drawers full of 5870 tubes that don't work properly, and
> Hello Ron,
>
> I see, so actually what you're saying is if they don't work properly
> straight away, it is better not to use them at all. That's what I am
> doing right now :-). I'll still give the AC a go with some of them,
> and then see if the problem will return afterwards. It's quite easy to
> test it that way, but then I won't need to design a complete test
> circuit.
>
> Michel
a few thousand 5870 tubes that do work. I can't bear to throw away the ones that
don't work, as they might be valuable to someone after the thousands that do
work are gone.
--
David Forbes, Tucson, AZ
Michel
Dec 10, 2012, 4:37:50 PM12/10/12
to neonixie-l
> I can't bear to throw away the ones that
> don't work, as they might be valuable to someone after the thousands that do
> work are gone.
>
> --
> David Forbes, Tucson, AZ
Yes, I know what you're saying, I even keep the cracked tubes :-)
> don't work, as they might be valuable to someone after the thousands that do
> work are gone.
>
> --
> David Forbes, Tucson, AZ
although they are completely useless to me, maybe one day someone
might like to have them.
Michel
dr pepper
Dec 11, 2012, 12:48:53 AM12/11/12
to neonixie-l
I've not heard of sleeping sickness till now, but I think I've seen it
on a couple of tube I have, they dont work untill I crank up the
volatge to the point where all the others are way too bright.
Like you I kept them in case they can be sorted, I'll put them on my
hv supply and cook them in and see if that sorts them out.
on a couple of tube I have, they dont work untill I crank up the
volatge to the point where all the others are way too bright.
Like you I kept them in case they can be sorted, I'll put them on my
hv supply and cook them in and see if that sorts them out.
Reply all
Reply to author
Forward
0 new messages