unpoison a IN-1

[Benoit Tourret]

Hello, I have some IN-1 that seems to be poisoned. poisoned or rather oxided, as they seems to  be really new. it is fine mesh from September 1971.

two of them are fine after one or two day, tree other doesn't show any improvement.

I read that I must overload them a while but how many and how long should I keep one digit on before light the next digit ? is it 1 second or one day...

by the way, as I read that this tube doesn't have mercury inside, can it be unpoisonned ?

[gregebert]

I never had an IN-1 run long enough to get poisoned. Mine developed internal shorts, resulting in 2 numerals glowing at the same time.

From what I've heard, nixie degradation increases exponentially with current, so you want to stay as close as possible to normal operating current. Try running the worst digit continuously for a few hours at 20% higher current, and see if that helps. I was (briefly) drawn to the IN-1 over 10 years ago because of the large size for the low cost; I bought a case of 25 for around 40USD including shipping. Only a few survive today that weren't dissected or tortured in a microwave oven (yeah, pretty hilarious but kinda toxic because of the bakelite).

A few of the IN-18's in my clock have shown signs of poisoning even though I run a cleansing routine for 1 hour every night. This is most-visible on the first day of the month for the months digit, and it can take a few days to recover. This clock has 14 tubes; kinda ridiculous but it looks impressive. It was inspired by the movie "Tomorrowland", which has a brief scene with two 6-digit IN-18 clocks. I'm certain one of our neonixie members is the creator of that clock (that was an invitation for you to take a well-deserved bow...).

Tubes that run 0-9 during the day need no cleansing (unit seconds, unit minutes, unit hours). Tubes the run 0-5 during the day are cleansed by running 6-9 (tens seconds, tens minutes). Tubes that are static, or near static (tens hours, month, day, year) run 0-9 for cleansing. 

The static tubes, notably the 4-digit year, show very little signs of poisoning because I have swapped them around to put the most-vulnerable-to-poisoning tubes where they run 0-9 during the day.

I havn't seen any signs of poisoning on Burroughs tubes.

[gregebert]

I couldn't resist asking google Bard this question: Are IN-1 nixie tubes reliable ?

Well, here's the answer. Feel free to laugh.....love the part about the filaments. Geez, maybe that's why I had so much trouble with mine ?

IN-1 nixie tubes are generally reliable, but they can fail over time. The most common failure mode is a leak in the vacuum tube. This can happen due to a number of factors, including manufacturing defects, physical damage, or exposure to extreme temperatures.

If a leak develops, the tube will slowly fill with air and the filaments will eventually burn out. This can happen over a period of months or even years. To prevent leaks, it is important to store IN-1 nixie tubes in a cool, dry place and to avoid dropping or banging them.

Another potential failure mode is a failure of the heater filament. This can happen due to a number of factors, including overheating, overvoltage, or physical damage. If the heater filament fails, the tube will not light up.

To prevent heater filament failures, it is important to use a regulated power supply and to avoid exceeding the maximum operating voltage.

[Benoit Tourret]

I ran each tube with 180V 2mA / two hours, turning each plate for 2 minutes.

on 10 tubes:

7 are OK

3 are not perfect.

I will wait to be able to lit them all together.

[gregebert]

The IN-1 datasheet says 2.5mA typical operating current, so you should try cleaning the other 3 tubes at just under 3mA. The other 7 tubes might clean-up on their own at normal current if the usage is balanced.

So did you see some improvement ? 

[Nick Andrews]

So I shouldn't be storing totes of nixies on a high shelf in my shed in Texas?

" Only a few survive today that weren't dissected or tortured in a microwave oven (yeah, pretty hilarious but kinda toxic because of the bakelite)."  This is why I used to have a shop microwave, just for experiments and such...  They are cheap enough!  But I gave it away while moving from NM to TX (it only took 5.5 years to complete).

--
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/61137a91-eefa-4a14-bd84-e173bf4af97bn%40googlegroups.com.

[Benoit Tourret]

I don't know how to rise the intensity...

I wish I had a laboratory DC power supply, but mine stops at 30V.

so I put a  NCH6300HV to raise the voltage. If I can adjust the voltage, I cannot adjust anything else. 

however I added a voltmeter and ammeter to check what I'm sending to the tube. Do you know if there is an adjustable power supply that can be used for our beloved tubes?

So, I regulate the voltage and I check that I do not exceed the desired values.

should I change the 32K resistor between the PS and the tube? could a potentiometer be used?

So I did see an improvement. but some need to be "heated". I of course doesn't speak of temperature, but they need 10 to 20 seconds to get their full glow, a bit like a neon tube with an old starter.

They are 52 years old now and never been used.

[Adrian Godwin]

When I wanted a higher voltage supply to refresh some tubes, I found an electrophoresis supply on ebay (Shandon Vokam 400-100). It provides up to 400V at 100mA with voltage/current metering and did a good job for me. I think it cost me about £40.

--

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/b554ff13-3806-4c16-af85-8bb7bb10e277n%40googlegroups.com.

[gregebert]

Ben - You can just reduce the resistor value slightly to increase the current. At 180V, you had 2mA with a 32K resistor; using ohms law, the voltage-drop across the resistor is 32K * 2mA = 64V. Since the power supply is 180V, the voltage across the nixie tube is about 116V.

To calculate the value for 2.9mA, assume the voltage on the nixie tube is still 116V. In reality it will be a bit higher due to more current, but for simplicity assume its the same. So, the new resistor value is 64V/2.9mA = 22K (a common value).

[Moses]

I've always just "unpoisoned" tubes with slightly higher current then normal, not higher voltage. A couple of hours to a few days usually fixes them for me. I don't think a 150-200% overdrive on the current will hurt it too much for a short duration. I've seen some tubes unpoisoned right in front of my eyes after just a few minutes at around double the current.

I've used brand new tubes where some digits didn't light fully.. usually after a few days of normal usage, at normal current levels, they came back. I've always viewed this as normal. I figure they have been sitting unlit for decades, it may take a little while to get everything lit up.

Regards,

-Moses

[Benoit Tourret]

In order to be able to talk about the same thing, I have a few questions regarding my DC power supply.
I have in order: a 0-30V / 0-10A adjustable DC power supply, a  NCH6100HV (12-24V to 85-235V HV) module, a 0-300V DC voltmeter, a 0-30mA ammeter and a 32.4 kΩ  limiting resistor  mounted on the wire that goes to the tube.
in my opinion, the voltmeter should be in place of the resistance and vice versa. so far, am I okay?

If I replace the adjustable resistor of the HT module with a potentiometer on the front, I can modify the voltage.
My current limiting resistor (a USSR model salvaged from a IN-12 nixie board) measures 10mm x 4mm. I have no idea of its power or the power needed, as I want to change the intensity, I have to replace my resistor with a second potentiometer.
I thought of a potentiometer of 50 or 100k. But what power... a small Chinese knob or a large rehostat of 200 kilograms?
I don't want to power much more than 5 or 6 tubes.
Until now my DC power supply never showed anything other than 0.00 A

Am I completely off the mark or is it doable?
I also have a DC-DC yh11068a module (https://dalmura.com.au/static/YH11068A.pdf

Here in France, ebay is not used by companies to resell their used equipment. Removing an item from an inventory requires tons of paperwork. So, finding a power supply for an EDM machine is rather complicated.

[Moses]

Benoit,

I'll try and clarify what I can for you.

"the voltmeter should be in place of the resistance and vice versa"... I think you mean ammeter here? If so, the ammeter is in series, so it shouldn't matter where it is really. If you really did mean voltmeter.. it depends on if you want to read the power supply voltage or the voltage drop on your nixie tube.

For the power of your resistor: Once the nixie tube lights.. the sustaining voltage is going to be around 140v or so on the nixie tube. Leaving about 40v drop on your resistor. Power dissipation on your resistor will then be.. P=VI, P=40*0.0025ma, P=0.1watts. I normal 1/4 watt resistor should be fine. Same calculation applies if you want to put a rheostat on your tube, a smaller one should be fine. No need for 200kg one lol

Do remember, one dropping resistor per nixie tube. Pay attention to the voltage rating on the resistor too, most small ones are a maximum of 200v or 250v, for the rheostat see if you can find the datasheet for this max voltage. Yes, the sustaining voltage is a lot lower on the resistor, but I would try and use one that is rated for your supply voltage.

If you are using a rheostat for adjusting the tube current, put a resistor in series as well to set your lower value. This is so you don't drive the tube with 0 ohm resistance when you turn it all the way down. A 20k or so should be ok.

That yh11068a module is rated for 40 watts! That is like 10 times what you would need for most nixie clocks! Be careful at higher voltages, a 200mA capable supply at a few hundred volts will hurt you good.

Regards,

-Moses

[Mac Doktor]

Reading this thread has begun to make my eyes glaze over. I find the following diagram to be helpful for understanding some of what's going on. Perhaps it will add some clarity to the discussion. This in from Marcin Saj's github and specifically concerns his PRO Nixie Tester:

It took a while for me to completely understand how all of this works and this diagram was a huge help. I doubt that I would have worked it clearly otherwise. For the record I haven't gotten around to reading the JB Dance book—yet.

Full info here:

https://github.com/marcinsaj/Universal-PRO-Nixie-Tester-and-Healer

Right now I'm using the PRO Nixie Tester with three DMMs but I've purchased some DPMs for a compact, dedicated meter bridge. Someday I'll get around to making a panel and wiring them up. And rigging up isolated power supplies for them.

On Apr 22, 2023, at 1:50 PM, Moses <mo...@neonixie.com> wrote:

That yh11068a module is rated for 40 watts! That is like 10 times what you would need for most nixie clocks! Be careful at higher voltages, a 200mA capable supply at a few hundred volts will hurt you good.

Zaparooney.   x_x

Terry Bowman, KA4HJH
"The Mac Doctor"

https://www.astarcloseup.com

"If only you could see what I've seen with your eyes."—Roy Batty, Blade Runner

[Benoit Tourret]

Thanks Moses.

I put a drawing to explain the "resistor and the Ammeter" cause one is serial and the other in //

I found the specs of a rheostat: https://docs.rs-online.com/736e/0900766b8142087d.pdf but I don't see the max voltage.

[Benoit Tourret]

Thanks Terry.

this also will help me a lot.

My electronic courses are too far to be clear in my mind...

When I read gregebert's thread, I wanted to write the complete equation to calculate the resistor for a wanted current. but the result was the opposite of the reality... 

so I decided to use a potentiometer and a galva...

but you gave me the result !!!

[Benoit Tourret]

Here is  the first try for my testing devices.

My HT power device, with the voltage regulated with the rheostat. next step will be the rheostat for the output power and placing the voltmeter after the limitation resistors and an on/off switch.

The box is just to have an idea of the size.  may be I will print something or find a more accurate box... ;-)

I will add a 5V USB power supply

on the left an arduino uno with a hat and a zif for testing K155id1 with a 12 pin SIL connector for a nixie tube, here a support for IN-1.

I made one for IN-12, I am making one for IN-2 and one other for IN-4.

on top of the box, a 12 position manual switch. I will use a Protoboard for better contacts.

[gregebert]

Looks good. I've gotten very fond of the inexpensive digital meters on Ebay and started putting them in many of my projects.

[Benoit Tourret]

I am a bit afraid of all those inexpensive displays made in PRC, the seller often doesn't know what he is selling, the specs are wrong or trunked, and I don't know their behavior with 300V DC or more...

Well, I ran a few of my IN-1, I had to run a couple of them at 4mA 180V and the black marks disappears completely after 4h. I can reduce to 150V 2mA and still have a good glow.

those tubes are very nice with the "normal" 5, much better than the IN-4. 

should be nice on a 6 digit clock with a decatron for tenths of a second.

[gregebert]

I've used a few different flavors of these cheap digital meters (0-500VDC, 120VAC, 300VDC) and have not had any problems....yet.  Most of them are self-powered, which is handy because it doesn't require a separate power supply, though risky because they do create a potential source for a short. Everything I build has fuses, so even if one of these things shorted-out, it would just blow a fuse. All but one of them are in metal cases, so if they caught fire, it would be contained.

The most dangerous line-powered items I know of are LED bulbs; even worse than CFLs due to their construction. I've replaced several on the verge of failing, or already failed, that have obvious signs of overheating or will actually still hot in the socket. Hmmmm....now I gotta go burn one to see if the plastic is self-extinguishing. I'm pretty sure those cheap meters will burn on their own once ignited, but I'm not sure I want to destroy one in the name of science.....unless I can find a really cheap one.

[Benoit Tourret]

Thanks for your help, 

I finished my HV power supply...

From right to left, top to bottom:

Voltage Output of the DC-DC converter

Ajustment of the DC-DC output

Input 8-32V (Usually Set on 12V)

AMeter Output

50K Adj. + 4.7k Resistor (2w each)

Voltage Output to the tubes

output (up to 300V DC)

So far, all is working just fine, tested up to 250V.

I may add a shunt or a change the AMeter if needed.

Regards,

Benoit.

[gregebert]

NICE!  You will find this very useful for testing and characterizing nixie tubes.

For my more-valuable tubes (RZ568m and b7971), I collect current-voltage data for each numeral (or segment) and plot on a graph. From there, I can see if there is consistency across tubes, which is an indicator of quality. Anything inconsistent is an indication of a quality issue, or possibly end-of-life.