IN-9 problem

[Mateusz Dziuba]

Hi! I started working on another project involving IN-9 nixies, so I bought 10 of them on ebay and I am in the middle of the testing. Unfortunately the tests didn't go as far as I would like to.

I built a current mirror on 2xMPSA42 and 1k potentiometer as reference. As most of the lamps light up only to the half of cathode I left them burning on max current for hour or two and achieve 8-8.5 mm glow only. Now I am stuck. When I measure the current flowing through Nixie, I get 10.1-10.4 mA max on each lamp and cannot get further.

Is this a problem with gas in my IN-9 or maybe is this a configuration problem?

[threeneurons]

Full "deflection" happens with 12mA, and the tube gets a little non-linear past 8mA. Monitor the current thru the tube, and the voltage across the tube, at the same time. Make sure your supply is up to the task before finding fault in the tube.

[Jon]

Nice pictures Mike!

I'd agree with what you're representing here (esp the non-linearity at the top end), with one additional comment - this won't necessarily apply to tubes that have been sat unused in storage for a few decades. Some IN-9 seem to suffer from a sleeping sickness / cathode poisoning / whatever that can mean that even at 12mA you initially only get about 50% glow coverage as the OP describes. Reconditioning these so they operate smoothly over the full length under the intended operating conditions is possible, but it's a bit hit and miss. Back when I made this I was having to chuck away a reasonable proportion of tubes that never made the grade even after extensive work.

As a starting point Mateusz, I suggest you tweak your circuit to allow you to put up to 25mA through the tubes. Even on heavily poisoned IN-9, that should clear out the crud and allow the glow to reach the top of the tube. But don't leave them unsupervised or keep them for long in this state - the tubes get hot and prolonged over-current is not going to help a long operating life! Interestingly, IN-13 don't seem to suffer the problem anywhere near so much.

Jon.

[Mateusz Dziuba]

Hi again guys. 

As I finally found time to continue with my IN-9 project I did small checks according to your suggestions.

I connected 2 separate tubes from 2 separate current mirrors connected to my HV PSU, and both of them light up same amount like before.

Current draw through HV PSU get doubled so I assume there is no problem with PSU capacity. Both of tubes draw current not exceeding ~10mA and again I cannot increase it to 12mA not to mention to higher values to overcome cathode poisoning. 

I am stuck again.

[Mateusz Dziuba]

Forget to mention, that voltage od tubes stays around 100V all over the time. In one point, when i am turning reference potentiometer tubes are switching off and starting to glow from around middle of bar with random length.

[Jon]

Are you sure your current regulator circuit is capable of being set to sink >10mA? You should be able to test it at a lower voltage without the IN9 in place to experimentally verify that.

Jon.

[Joe Croft]

Give them 6 to 12 hours. This is what I have learned over time. In-13s are more reliable at this than in-9s. On my kits I fing I get a 25% loss.

-joe

On Sun, Nov 25, 2018, 11:23 AM Jon <deka...@nomotron.com wrote:

Are you sure your current regulator circuit is capable of being set to sink >10mA? You should be able to test it at a lower voltage without the IN9 in place to experimentally verify that.

Jon.

--
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/e2f4678e-fad4-4008-9edf-6ed6998a3a2d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Mateusz Dziuba]

Hi again!

After abandoning the project for a long time I came back to my IN-9 tubes. 

I replaced the PSU (NE555 based -> MC34063A based), replaced transistors for MJE340 (which will be used in final design) and did a bit of burning of each lamp separately. Finally I get to the top of the tubes (well 9,5 mm). 

It seems lamps required to be burned with higher current which my previous PSU, current mirror and potentiometer arrangement was not able to provide. I am blaming poorly designed PSU. 

I would like to thank all of You for your time and advices.

Now I am facing another problem.

I built set of band-pass filters using LM324 with single supply 0-5VDC. As the 2.5VDC is feed to non-inverting input of op-amps, and audio signal is feed on inverting input, the value is oscillating around 2.5V (middle of nixie). I was planning to use microcontroller and map the output voltage to 0-5V by software and then feed on MJE340. in the end I decided to not use microcontroller, and now I am stuck at point "How to get a 0-max swing of the bar on the nixie?"

[Bill Notfaded]

My advice... Switch to IN-13's unless you have a case of IN-9's.

Bill

[Mateusz Dziuba]

I have few spare. Two with broken lead unfortunately :( I am working on repairing them. Angway I do not have any IN-13 right now, but probably will get some if succeed with this project.

[gregebert]

I did limited experimenting with these tubes. What I found is that a current-controlled supply is the only way to get repeatable behavior out of those tubes.

If I just used a traditional variable-DC supply with a series resistor, I had erratic behavior such as glowing from the opposite end or from the middle.

For the amount of energy consumed, they produce substantially less light than NE-2 bulbs. I think I had to pump almost 20mA into mine to get 90% glow, which is twice the spec limit.

[GastonP]

A 5 Volts power supply is a big limitation regarding to a current source compliance. I'd do the following:

1 - Decouple the outputs of the filters from the rectifiers, so as to have an output voltage referenced to 0 and up to 5 volt. The rectifiers act as peak detectors and your signal moves from 0 to 5V before clipping.

2 - Build a current mirror sink to ground, knowing beforehand that close tracking of discrete devices is an utopia, per output tube (I support the IN-13 advice).

3 - Inject the minimum bias current and sum up with a variable current to swing the total needed by the tube. This, done with simple resistors to +5V and the output to unity gain buffers (LM324 will work fine).

I'd try to make a schematic when I get time, but the circuits are very simple.

Gastón

[Joe Croft]

To solve the glow starting from the center or top of the tube, i never turn the tube fully off. I will turn it on full then ramp the current down until a small segment is still lit. This helped make them much more reliable.

-joe

> --
> 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 view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/0e6ab50a-f576-4418-a018-9717d071866f%40googlegroups.com.
>