IN-3 Neon Bulbs

692 views
Skip to first unread message

Spencer

unread,
Mar 1, 2014, 8:17:46 AM3/1/14
to neoni...@googlegroups.com
Hello All!

I bought these for the nixie clocks, http://www.ebay.com/itm/251373901401, and they will work for a bit then start flickering all over in the tube, and lighting up in odd places and not where it should be lighting up!

An example can be found at, http://imgur.com/O32rVBi . Only the left side is light up, the little light has a mind of its own and has traveled to the back of the tube.

I am feeding the little guy +180V and the anode resistor is 220K. I am not sure if its too much current or what not.  Several have done this to me.

Suggestions?

Thanks!

Jon

unread,
Mar 1, 2014, 3:24:20 PM3/1/14
to neoni...@googlegroups.com, Spencer
Are you sure you've wired the IN3 the right way round? Unlike normal neon bulbs the internal construction is not symmetrical, and so they have specifically designated anode and cathode leads.

Jon.


Tidak Ada

unread,
Mar 1, 2014, 3:59:07 PM3/1/14
to neoni...@googlegroups.com
I am just occupied to develop a ac driver to supply neon bulbs symmetrical. Those simple neon's with only two rods as electrode are awful on dc.
 
eric

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Spencer
Sent: zaterdag 1 maart 2014 14:18
To: neoni...@googlegroups.com
Subject: [neonixie-l] IN-3 Neon Bulbs

--
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/1393679866.7567.YahooMailNeo%40web120901.mail.ne1.yahoo.com.
For more options, visit https://groups.google.com/groups/opt_out.

Quixotic Nixotic

unread,
Mar 1, 2014, 4:57:58 PM3/1/14
to neoni...@googlegroups.com
I bought 1,000 of these and conditioned a whole bunch of them for a ring counter clock. Many of them started to flicker and develop bad habits. I think they are just not very good neons.

John S

Frank Bemelman

unread,
Mar 2, 2014, 4:59:27 AM3/2/14
to neoni...@googlegroups.com
Hi Eric,
 
That’s a good point. I guess it would not be too hard to construct a H-bridge from 2x mpsa42 and 2x mpsa92
and drive that from the microprocessor, or even with one side using mpsa42 and mpsa92 and the other side
just 2 resistors of ~100k between HV and GND.
 
Cheers, Frank



Dit e-mailbericht bevat geen virussen en malware omdat avast! Antivirus actief is.


Dekatron42

unread,
Mar 2, 2014, 6:00:36 AM3/2/14
to neoni...@googlegroups.com
Some of the Russian users over at Radiokot.ru have discussed this and similar circuit arrangement, like for the IN-24 here: http://translate.google.com/translate?depth=1&hl=en&ie=UTF8&prev=_t&rurl=translate.google.se&sl=ru&tl=en&u=http://radiokot.ru/forum/viewtopic.php%3Fp%3D1268086 - I used Google translate so it would be easier for you to read directly. There are many discussions back and forth for which is the best solution, but the one that user Slvik has proposed at the top works very well for me too.

/Martin

Frank Bemelman

unread,
Mar 2, 2014, 7:03:42 AM3/2/14
to neoni...@googlegroups.com
Hi Martin,
Yes, the top one is in fact an H-bridge.
The 2nd circuit is as good as the first, it only consumes a tiny
bit more current. I’d favor the simple circuit, quite a reduction
of parts which makes the H-bridge look a bit over the top.
Cheers, Frank
--
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.

Tidak Ada

unread,
Mar 2, 2014, 7:28:12 AM3/2/14
to neoni...@googlegroups.com
Hi Frank,
 
That's even better as I have in mind. However, I am busy to make a dc - ac converter. I need it for an other project, where is no hv in use (meter clock). I want to use neon's to avoid LED's because of "old style".
 
Later more.
 
eric


From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Frank Bemelman
Sent: zondag 2 maart 2014 10:59

To: neoni...@googlegroups.com
Subject: Re: [neonixie-l] IN-3 Neon Bulbs

John Rehwinkel

unread,
Mar 2, 2014, 11:50:15 AM3/2/14
to neoni...@googlegroups.com
> That’s a good point. I guess it would not be too hard to construct a H-bridge from 2x mpsa42 and 2x mpsa92
> and drive that from the microprocessor, or even with one side using mpsa42 and mpsa92 and the other side
> just 2 resistors of ~100k between HV and GND.

Easier yet, just use two transistor pulldowns and two resistor pullups:

http://www.vitriol.com/images/tech/ne2both.gif

- John

Terry S

unread,
Mar 2, 2014, 3:42:00 PM3/2/14
to neoni...@googlegroups.com
I don't think that will work. The resistor will effectively short out the resistor/neon combination. It won't light in either direction.

Frank Bemelman

unread,
Mar 2, 2014, 3:51:44 PM3/2/14
to neoni...@googlegroups.com
Ok, I can understand that. You could try a royer converter. The fun is in the making
of the small transformer.
But there are many ways to skin a cat Knipogende emoticon
Cheers! Frank
wlEmoticon-winkingsmile[1].png

John Rehwinkel

unread,
Mar 2, 2014, 4:12:42 PM3/2/14
to neoni...@googlegroups.com
> I don't think that will work. The resistor will effectively short out the resistor/neon combination. It won't light in either direction.

One transistor is on at a time, pulling that end of the neon down to (approximately) ground, basically overwhelming the resistor on that end. The other end has the transistor off (effectively open circuit), and the resistor on that end simply acts as an anode/current limiting resistor. When you switch transistors, the same thing happens the other way. It lights just fine in both directions. I use this lashup with ordinary NE-2 type neons to have colon lights where the glow jumps from one electrode to the other every second for a little visual interest.

- John

Tidak Ada

unread,
Mar 2, 2014, 5:05:27 PM3/2/14
to neoni...@googlegroups.com
That's exactly what I intend to do, however a little bit more sophisticated:
 
 
groetjes,
 
eric


From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Frank Bemelman
Sent: zondag 2 maart 2014 21:52
wlEmoticon-winkingsmile[1].png

John Rehwinkel

unread,
Mar 2, 2014, 6:23:21 PM3/2/14
to neoni...@googlegroups.com
> http://en.wikipedia.org/wiki/File:Royer_Circuit1.gif

Usually when someone refers to a Royer oscillator, they mean a resonant Royer, which has a series inductor and resonating capacitor, and produces a sine wave output. The wikipedia article (like many such) is somewhat misleading, and mostly describes a push-pull blocking oscillator.

http://cds.linear.com/docs/en/application-note/an118fa.pdf

http://wiki.4hv.org/index.php/Royer_oscillator

- John

Joseph Bento

unread,
Mar 2, 2014, 6:56:04 PM3/2/14
to neoni...@googlegroups.com
What a great link!  It's fun to see what fellow hobbyists are doing, and the translation of the web page makes it so much easier!   Since I am a sucker for any new Russian nixie related stuff, I did a search on E-bay for the IN-24.  I found some, and therefore ordered a few.   Hopefully the current situation in Russia doesn't affect the mails.

Joe, N6DGY
Message has been deleted

NoCampersFluffy

unread,
Mar 3, 2014, 6:14:21 AM3/3/14
to neoni...@googlegroups.com, Spencer
Spencer.
As per Jon's post - there is a little arrow at the bottom of the glass envelope that helps to determine the polarity of he IN-3s.  Arrow points to the cathode.  Details are on the data sheet (in rRussian) on Dieters web site as to which is cathode and which is the anode.

Did you measure the strike and maintain voltages of the IN-3?  I've found wide variations in strike and maintain voltages even within the same box of 100 tubes. Typical operation in the data sheet is 0.8ma with a low of 0.2ma and a high of  1ma.  If your IN-3s maintaining voltage is at spec (55v), then by my maths you are dropping 125v over the 220k anode resister which is a 0.5ma current. This is within tolerances but on the low side.

I have tubes that have maintaining voltages as low as 36v and some other are as high as 70v.   If you run the math you will find that you can lower the anode resister to 180K, and even lower for tubes with higher maintaining voltages (probably best not to go below 150K!). I try and run all of my IN-3s as close to the 0.8ma spec as possible - but this is for consistency in a ring circuit where I am also trying to get all of the tubes strike/maintain voltages aligned.   Note even at 0.9 to 1ma I still got flicker and coverage variations in some tubes, but higher current did help with getting full plasma coverage.  

My own experience is that some IN-3s flicker and the plasma can even form at the back rather than the front of the tube before returning to the front again.  I have also had bulbs where the plasma forms on the wire lead at the bottom of the tube.  In some applications the flickering is endearing, it others, such as ring circuits its down right annoying negating the circuit.  Like John S,  I've also aged my IN-3s and it does not seem to make much difference - there are always some that just want to flicker.  If you age the IN-3s then after a couple of hours these "flickering" ones will become visible to you.  Having said this I have aged IN-3 with a couple hundred hours of ring circuit use that decided to flicker and become unreliable - go figure. From what I have observed, increasing the strike voltage or (minor) current changes has no impact, but higher current do get better glow coverage.  There also appears to be no correlation when it comes to flicker between the strike and maintaining voltage differences between tubes.  Having said this beware of tubes which have a high strike and maintain voltages as from examination I often found these to have some visible internal physical variation to a "normal" IN-3 and of course they were no good for a ring circuit.  From aging several hundreds of these IN-3s and building 2 ring-counter clocks,  to me it just looks like there are differences in the manufacture between tubes.  IN-3s look to have wide tolerances between and within the same box.  Ive discarded 80% of some boxes and less than 10% of others due to variations in performance - but again this is for ring circuits.

In your application you are using the IN-3 as a colon separator.  If it worries you replace it.  If not you can call it an extra feature.  One of my fellow workmates really likes the flickering of one lock I have at work,  which takes the form of the plasma pulsing, she says it has personality.

Spencer W

unread,
Mar 3, 2014, 11:26:49 PM3/3/14
to neoni...@googlegroups.com
Thanks everyone! I will go ahead and try measuring the voltage 

Sent from my iPhone
--
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 email to neoni...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages