IN-13 Lifespan

[TheJBW]

About 1.5years ago, I built this really nice stereo vu-meter using IN-13 bargraph tubes. About two weeks ago, one of the tubes made a "tink" sound and the current control went away -- the tube stayed ignited, but the "bar" was stuck at the bottom. I noticed that the spacer at the far end of the tube was also suddenly free. I figured this was an unlikely failure, so I replaced the tube the spare stock. Unfortunately, one week later, the tube on the other channel did the exact same thing! I've attached pictures of my drive and power circuitry -- the signal conditioning circuitry is omitted, but I think it unlikely to be relevant. I figure something must be wrong with my circuit that is causing early death, but I'm not sure as to what. Ideas anyone?

[Instrument Resources of America]

As far as I know, one can NOT build a bar-graph, or neon lamp style, LED, or any other indicator type, of VU meter. (Perhaps you built a 'level indicator' of sorts.) The very definition of a VU meter (at least the professional type) involves the mechanical ballistic damping, and mechanical response of the needle and the meter movement.   Ira.

On 1/23/2017 11:41 AM, TheJBW wrote:

About 1.5years ago, I built this really nice stereo vu-meter using IN-13 bargraph tubes. About two weeks ago, one of the tubes made a "tink" sound and the current control went away -- the tube stayed ignited, but the "bar" was stuck at the bottom. I noticed that the spacer at the far end of the tube was also suddenly free. I figured this was an unlikely failure, so I replaced the tube the spare stock. Unfortunately, one week later, the tube on the other channel did the exact same thing! I've attached pictures of my drive and power circuitry -- the signal conditioning circuitry is omitted, but I think it unlikely to be relevant. I figure something must be wrong with my circuit that is causing early death, but I'm not sure as to what. Ideas anyone?

--
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.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/ffefa631-9ae9-4ee1-a55c-b648cf5c7012%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[gregebert]

If the op-amp is running on a +5V supply, the maximum possible current in the IN-13 is about 20mA per the schematic.

But if the op-amp runs off the +12V supply, you could get up to 50mA .

I think IN-13's are rated for 4.6mA full-scale, so either way you could be overdriving it.

[TheJBW]

The amp does indeed have a 12V supply, but the input to the opamp follower is driven by PWM output of a 5V micro. I suspect I am overdriving it, but not due to the opamp supply -- in my experimental experience the current through the tubes has to be about 15mA to get full scale. If there's a way to get full scale at 4.6mA, I'm unaware of it, but I could be doing something stupid.

[Jon]

Sounds like a good excuse for a tube autopsy...

If you are indeed over-driving the current then you'll be sputtering material off the cathode which after a while will have two effects. Firstly there will be some gradual darkening of the tube as the material deposits elsewhere. And secondly you'll see gross physical erosion of the cathode - you can see a similar effect on dekatron cathodes where the glow has been sitting stationary for a very long time (usually the zero cathode of the tubes making up the higher order decades of a multi-tube counter). Given that you describe a 'tink' noise, my guess is that the cathode has eroded to the point where it physically broke, perhaps under the thermal stress of being activated. Time for new tubes.

The Soviet datasheets only promise 1000 hour life for IN-9 - I can't find a similar overall guaranteed life specification for IN-13. My experience is that they will last much longer than that if not overdriven.

With regard to the current required to achieve full scale, yes there's normally some cathode poisoning which needs to be burned off when you first fire up the tubes (worse on IN-9 than IN-13), but then they should settle down and operate at the specified sensitivity. The last centimetre or so of the cathode is difficult to illuminate even in a well 'oiled' tube - they lose linearity at the upper extreme. Best thing to do is not to worry about trying to light up this bit.

Cheers,

Jon.

 

[JohnK]

But, if you get the maths right you can actually mimic the response.

[I have audio and broadcast radio station background btw]

Been here before with this argument of course. So, it is possible to get the same indication but that leaves the semantics to deal with.

And there are subtle aspects of the visual persistence of different displays to deal with too.

 

The argument usually revolves around the substitution of standard moving coil meters for the correct type. The original spec document is on the web. The attenuator that 'must' be used and the standard levels topic gets quite interesting.

[Talking about the way the indicated value on the meter when used in its most sensitive setting got used as the way to refer to the actual level, rather than the fact that the actual level was higher - I didn't tell that story very well; have a look at the specs and the circuit and how the common useage developed.]

 

John Kaesehagen

Australia

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/c917d5ac-89ab-61f0-3a06-c1c85b523176%40HUGHES.NET.

[JohnK]

I too want to see the results of the post mortem. What IS the mechanical reason for the spacer being free? Please tell.

 

John K

--

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.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/d11e0f79-e6f3-4657-83f8-1aed7fd14ede%40googlegroups.com.

[threeneurons]

A couple of data points, but by no means statistically significant. I've had an IN-13 running in my nixie thermometer (pre-kit) for close to 2 years, now, 24/7, and it runs and looks like new. One of my TRW buddies has a simple "winker", that uses the IN-13. It still glows and winks, but its no longer anchored. The glow, grow and shrinks, in the middle, and drifts a bit. That tube has been running constantly, for at least 5 years.

These tube tubes may have come from "good" batches, that exceed the average. They could just be average. Without taking a survey of a lot of tubes, who knows.

If you had been over driving them, the graphs would have been "pegged", visually, most of that time. If they were bouncing to the music, then you were not over driving them. Maybe you just got some from the wrong end of the statistical life curve.

   

[TheJBW]

Sadly, both tubes got tossed with the trash before I even made this post, so no autopsy is possible. That said, I did inspect them visually. There was definitely no visible darkening of the tube, but I couldn't get a good look at the base of the cathode due to shielding. I would have noticed sputtering (which I looked for) after the first failure when a new tube was side by side with a tube destined to fail. The older tube was not noticeably dimmer than the new one.

Signs indicate that erosion is the cause however, because of said "tink" -- The question is, could the sputtering have been confined to the area very near the cathode? Plus, it would make sense that the "low scale" portion of the cathode experienced much more erosion than the high scale, as the control portion of the system has an auto high voltage power off after 30 seconds but it would turn on when there were random transients (ground bounce, etc) during the day, and then go back to sleep. During that time, the tubes would be on for 30s with 0A control current. Perhaps that was bad for them.

I'll run a test / video for everyone with controlled current over the weekend -- to demonstrate bar length at 0-15mA control currents.

[TheJBW]

Alright,
I did a little investigation. Hopefully, someone can point out the "oh god, you're overdriving them" problem...

Investigation steps:
1. Scope HV line (HV supply is TES 1364 module)
2. Multimeter on 220Ω shunt resistor to get I_control
3. Input AC Signal -- steady 10kHz, varous levels
4. Input AC Modulation signal -- 10kHz AM'd 100% depth, 1Hz
5. DC Signal Input -- remove Microcontroller and input DC to 100kΩ resistor LPF before the voltage follower.

Results:
1. HV Line, no input signal: Mean = 143V, V_pp = 8V
2. HV Line, full scale input signal (1.8V PP 10kHz sine wave -- to front end): Mean = 143V, Vpp = 9V
3. Measure control current full system, suing AC signal input:
    I_shunt, no signal = 1mA (micro is still here, did not scope micro, but I suspect it was putting out a small nonzero PWM value)
    I_shunt, full scale = 11.5mA
4. Measure control current, DC injection to IN-13 driver:
Vin (V)   I_shunt (mA)    Col Len. (cm)
---------  ------------------    ------------------
0            0.045                0.5
1            2.41                  2.5
2            4.9                    5
4            9.4                    10
4.5         10.9                  11
5            11.4                  11
6            13.9                  11

I also measured the current through the cathode resistor (220k in series with the non-control cathode to ground):

Vin(V)  I_cath (mA)
--------  ----------------
0          0.04
0.5       0.26
2          0.38
4.5       0.5


So what we see is that under normal operation, the tubes should never see more than ~11mA, which is substantially higher than the expected max current to get the full scale, and these tubes have seen a couple of hours of burn in.

tubetester.com's translated datasheet (http://www.tube-tester.com/sites/nixie/dat_arch/IN-13.pdf -- scroll down) agrees with everyone here, that the max working current *should* be ~5mA. I'll grant, my striking voltage is above the max. Taking a closer look at voltages,

I see that the voltage between the anode and the indicating cathode when the tube is at 0 scale is 111V. When I drive it to full scale, it rises to 139V. If I'm reading the aforementioned datashee correctly, the 0 scale drop across the anode to indicating cathode should be at max 99V, so I'm out of spec but not grossly so.

The only thing I can think to do is adjust the HVPS to be driving them at a lower voltage (120V), but my recollection from when I breadboarded this is that they do not strike reliably below about 140V.

Okay, so this has been a huge ramble, but if any of you have any insight into this -- even if it only addresses a small part, it would be greatly appreciated.

[gregebert]

I'd rule-out voltage as the problem; consider neon devices to be current-controlled (more current --> brighter, etc). When you drive a neon device from a voltage-source, there usually a current-limiting resistor (or a transistor) that determines how the neon device operates.

If you are only pumping 11mA intermittently, and most of the time it's low/zero current, it doesn't seem reasonable that excess current is causing your tube to fail. The data you provided doesn't reveal anything suspicious to me.

I'm wondering if you have a bad batch of tubes ?

I think I will setup my bench supply with an IN-9 (that's all I have) and see what happens after several days at 6mA. Stay tuned.

[TheJBW]

Looking forward to it. Also, reading back, I realized that it's not intuitively clear what I mean by "full scale" or "half scale", so here are some pictures. 5mA for the half scale one and 11mA for the full scale. Note that the tubes look *much* dimmer than this as it was taken in a dark room with a long exposure.

[Joe Croft]

When I had a tube that took 15ma to drive it, I discoved that I had the two 'anodes' wired backwards. The tube had a short life, in the order of weeks running full time.

-joe

On Jan 23, 2017 1:41 PM, "TheJBW" <the...@gmail.com> wrote:

About 1.5years ago, I built this really nice stereo vu-meter using IN-13 bargraph tubes. About two weeks ago, one of the tubes made a "tink" sound and the current control went away -- the tube stayed ignited, but the "bar" was stuck at the bottom. I noticed that the spacer at the far end of the tube was also suddenly free. I figured this was an unlikely failure, so I replaced the tube the spare stock. Unfortunately, one week later, the tube on the other channel did the exact same thing! I've attached pictures of my drive and power circuitry -- the signal conditioning circuitry is omitted, but I think it unlikely to be relevant. I figure something must be wrong with my circuit that is causing early death, but I'm not sure as to what. Ideas anyone?

--

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+unsubscribe@googlegroups.com.

To post to this group, send email to neoni...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/ffefa631-9ae9-4ee1-a55c-b648cf5c7012%40googlegroups.com.

[TheJBW]

Joe,
I actually discovered exactly this late last night. Hadn't gotten around to replying yet. Turns out that the simple footprint got flipped at some point. I didn't think that the tubes would operate correctly at all with the anode swapped for the auxiliary cathode, but they seem to have lasted for 18months!

[Joe Croft]

I am surprised they lasted so long!

-joe

On Jan 30, 2017 12:20 PM, "TheJBW" <the...@gmail.com> wrote:

Joe,
I actually discovered exactly this late last night. Hadn't gotten around to replying yet. Turns out that the simple footprint got flipped at some point. I didn't think that the tubes would operate correctly at all with the anode swapped for the auxiliary cathode, but they seem to have lasted for 18months!

--
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+unsubscribe@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/cac3256a-ab8a-427c-b839-45d05009dcd2%40googlegroups.com.