New set of IN-9 nixie tubes

301 views
Skip to first unread message

rkvs

unread,
Mar 12, 2013, 10:33:45 AM3/12/13
to neoni...@googlegroups.com
Hello experts,
I recently bought a set of IN-9 tubes for a diy audio spectrum analyzer.

Here are links to the pictures of one of the tube:

Is this really new one? I noticed that the backside of the tube has black deposits like it was used?
Any ideas?

I'm totally new to nixie tubes.
-rkvs.
Message has been deleted
Message has been deleted

rkvs

unread,
Mar 13, 2013, 12:09:46 PM3/13/13
to neoni...@googlegroups.com
Hi,

I took out one tube from the lot and tested it as follows:

HV supply : 230VAC-to-110VAC transformer -> full wave rectifier (using 1n4007) =>> When i measured using multimeter, it read approx 110V dc (i believe that this is average value; peak voltage in rectified output is 155vdc.)

I used circuit shown in "Figure C. Single transistor current sink" in the following document:
(Thanks to the author of the document)

Vin input to base of MJE340 is made variable (0V to 5V) using potentimeter -> power supply for this is a nice regulated DC.

Emitter resistance I used was 330ohms.

After power ON of the setup and Vin is set to 5V, then only 50% of the tube length was glowing. I followed an advice from one of the member regarding intermittently making it glow using double the current. So, i reduced emitter resistance to 150ohms approx and the length of glow became 75% of tube length ! I changed back emitter resistance to 330ohms and now the glow length is 75%.

The following picture shows the glow in the tube when Vin was 5V approx.:

From other posts, I followed these two steps:
1. Not using filter capacitor for smoothing High voltage DC
2. Intermittently operate the tube with double the current 

Should I really perform Step #2 for several hours till the glow reaches 100% length?
Please advice. I shalll perform any experiment any one suggests.
Message has been deleted

rkvs

unread,
Mar 13, 2013, 12:39:45 PM3/13/13
to neoni...@googlegroups.com
I performed the test again after 10 minutes after posting the previous message. I repeated Step #2 in Previous post again few times (using 150ohms emitter resistance).
Now, the tube glow reached 100% i think.

I changed back emitter resistor to 330ohms and glow is now 95% of full length. I'm wondering why it didnt work when I performed step #2 some time back? Will 10 minutes gap inbetween do the trick???

Alex

unread,
Mar 13, 2013, 4:08:12 PM3/13/13
to neoni...@googlegroups.com
Hi,

The tubes do normally have dark deposits, though usually located in one place where it was evacuated of oxidizing gasses by igniting some combustible material (i believe). Either way, they are never test tube clean. This darkness is usually on the back though, not obstructing the viewing side. To be honest though, if they were used one would not expect the leads to be full length. They do poison up and a quick blast with some current will usually clean them up. IN-9 tubes are quite fickle and tend to misbehave quite a lot. My post a month or two ago about using ripply DC does help a lot with them jumping about. Have you got a Potentiometer you can use straight in line with a fixed resistor and your DC supply to vary the current around? Excellent for testing the linearity and looking for iffy spots? 

Good luck on your spectrum analyser project, I hope to do the same soon too.

- Alex
Message has been deleted

rkvs

unread,
Mar 29, 2013, 9:19:40 PM3/29/13
to neoni...@googlegroups.com
Hi,

I used a driver similar to th one shown here:

But, I think i was not able to get any voltage output out of lm358 opamp.
If i drive the base of mje340 directly using a pot, in-9 seems to work. Any ideas?

threeneurons

unread,
Mar 29, 2013, 11:45:10 PM3/29/13
to neoni...@googlegroups.com
Unless you need precision, the op-amp stage, and its feedback isn't needed. The resistor, in the emitter leg of the MJE340 will provide enough feedback, all by itself. Just note that the IN-9 (12mA) uses more current than the IN-13 (4.5mA). But with a minimum hfe of 30, the base current into the MJE340, should be no more than 400uA. If you make the emitter resistor 330 ohms, it will have a drop of close to 4V across it, when 12mA is flowing thru the tube. That'll be ~4.7V from the base of the MJE340 to gnd. It will also look like a ~10K load at that base pin. That's all assuming an hfe of 30. Typically, it will probably be twice that value, and the load will look closer to 20K. 

AlexTsekenis

unread,
Mar 30, 2013, 7:51:53 AM3/30/13
to neoni...@googlegroups.com
Hi rkvs,

Warm welcome to the world of nixies.

The darkening of the tube is part of the manufacturing process and is not an indicator these have been used. Shorter leads are also not an indicator, sometimes they are trimmed for shipping. In brief, its purpose is to clean impurities from the cathode electrode and gas. The front of the tube is also affected, but it has been partially shielded by the mesh-type anode. The very early and excellent-named 'tuneon' bargraph tube had no such mesh and look at it before and after prolonged use.

Alex I think you are referring to traditional 'flashed' getters that leave an unmistakable reflective surface on the glass.
I believe the 'non-evaporable' type is used in the IN-9 and IN-13, in the form of a sintered metal alloy 'pill' at the very top of the tube. Here is a close-up of that 'pill' at the top of the tube, showing the granular surface. For the getter to operate it must get warm/hot. The getter is also connected to the cathode which warms it up during operation. Here is a close-up showing just this (apologies, it looked in focus on the camera). The thin wire is the end of the long cathode wire connected to the getter and stretched by bending that U-bracket.

If you want to activate the getter without the tube on...place the tubes in an oven. There are plastic parts inside so don't overdo it. Bake for 1h until crispy but not brown.
 
It is clear that your tubes have been left in storage for decades and therefore require cleaning of the impurities inside. The term for that is cathode cleaning by sputtering, or cathodic sputtering. Have a look at my post here for a method I use on IN-13s, same principles apply: https://groups.google.com/forum/?fromgroups=#!topic/neonixie-l/GAREG0tdVsI There are also a few comments on full-wave rectified or smoothed power supply.

You can easily clean the tubes with your setup. Just change the resistor to 220 Ohms (to reduce the current to 200% of rated) and add an electrolytic tank capacitor after the rectifier to have a smooth DC supply. 47uF-470uF at 200-500V should do it. For a capacitor you can use the ones from an old PC power supply. Please use an appropriate voltage and be careful. Dont worry about the higher average voltage, the transistor will take care of that. Set your input to 5V and power on. Leave it for a couple of minutes. A 'dirty tube' will glow from some point on the bar and slowly fully illuminate. Others will be very close to the top and others at the top and very bright. Reduce the time for tubes that start at the top to say half. Once done safely discharge the capacitor.

Good luck!

Regards,
Alex
Reply all
Reply to author
Forward
0 new messages