Dummies guide to building a 120watt 140 volt power supply to drive IN-9 Nixie tubes.

[fireant007]

I've built a few 'nixie' clocks, one has an IV13 linear neon bargraph, they take more power than the average nixie.

There are 2 approaches, 1 is power them direct from the mains as vintage equipment often did, and the other is an smps, 100 tubes is going to use at least 100watts, so I'd be looking at a push pull converter, probably a sg3525 smps ic, and a etd39 or etd44 ceramic transformer, a couple of mosfets and the required dicretes, push pull mode without feedback would be adequate and simple.
Do you have much experience with switching supplies?

Hi Doc,

I do understand them ... yes... laying them out is a different matter.
See post 13 for the ideal SMPS for this project. UNLESS you are trying to feed it 240Vac...... because as you may or may not know ... you can't reduce a step up or boost supply to less than the supply voltage.

I have not found any buck boost chip that will handle any more than about 60 volts... otherwise my problem would be solved !

Hence why I think Spec is on the right track ... I just have to convince him that you can't put the anode resistor on the cathode side ....
The cathode is the indicator !

 

[fireant007]

Hi Doc,

I do understand them ... (mostly) yes... laying them out is a different matter.
See post 13 for the ideal SMPS for this project. UNLESS you are trying to feed it 240Vac...... because as you may or may not know ... you can't reduce a setup supply to less than the supply voltage.
Hence why I think Spec is on the right track ... I just have to convince him that you can't put the anode resistor on the cathode side ....
The cathode is the indicator !

Please delete.

 

[fireant007]

Well... my theroy might be errr.... off.

This pdf shows the resistor in the cathode ?

 

[spec]

I am off to build....
Read this article .... might give you some insight.
https://www.dos4ever.com/Z550M/Z550M.html
Look at figure 12.... this is a similar concept to the IN-13 - the cousin of the IN-9

Well... my theroy might be errr.... off.

This pdf shows the resistor in the cathode ?

Gday fire,

Thanks for the above links- very interesting articles. The second link gives a particularly comprehensive, but succinct, description of cold cathode display tubes (CCDTs).

In any situation I'm not the slightest bit interested in being right or wrong- only the facts. But it does seem, after reading up on CCDTs, that all the 1N-9 knows about is the voltage between its cathode and anode, From this it follows that the position of any current controlling device, resistor or transistor, should have no impact on the operation of the 1N-9.

With the 1N-9, the mechanism for ensuring that the ionization starts at one end of the cathode is different to normal:

'The third and final electrode is termed the auxiliary cathode. It is much shorter than the main cathode, approximately 2mm long and also made
out of Molybdenum. This electrode is positioned at the bottom of the tube between the anode and cathode. The role of this electrode is to ensure
the discharge between the anode and main cathode begins at the start of the cathode and hence the bottom of the tube. A discharge between the
auxiliary cathode and anode provides positive and negative ions in its vicinity. The auxiliary cathode discharge thus effectively lowers the
breakdown potential locally, forcing the cathode glow to start from the bottom of the tube. In the ИН-9 tube (similar to the IN-13) no auxiliary
cathode was used. Instead, a Zirconium spike (even lower striking voltage than Molybdenum) was attached at the bottom of the cathode having
the same effect.'

Also, from the second link, here is the graph of voltage versus current for a CCDT which I have posted for easy reference:

 

[fireant007]

No worries ... this will be the definitive thread on IN-9 and IN-13's by the time we have finished with it.
Your method did not work by the way

Busy uploading the videos to Utub

Post links when they are done.

 

[spec]

No worries ... this will be the definitive thread on IN-9 and IN-13's by the time we have finished with it.
Your method did not work by the way

Busy uploading the videos to Utub

Post links when they are done.

Hi fire,
I am surprised that the 1N-9 tube did not illuminate in the test circuit because that means that the tube does not comply with the data sheet.
If you put the resistor in the anode side does the tube illuminate?
Can you gradually increase the voltage and see what voltage the tube strikes?
Sorry to keep on, but are you sure that the anode and cathode were connected the right way around. In any event can you try reversing them and see what happens.
Is the tube you have been testing new? Is it a proven tube that is known to work?
spec

 

[fireant007]

Hey spec,
I am just as interested in getting to the bottom of this as you. Even more so as I want a working circuit !
So please feel free to nag me as much as you like.

There are pictures identifying the anode around on the net .. but I can't find one to hand at the moment.

How ever ..... there are 2 leads. one is in the middle ... that looking through my jewelers loop .. just seems to go right up the middle.
That wire is the cathode.

The outer wire connects to to a grid that goes up the outside that is .. the anode .. the central wire is what gives off the electrons .. to ionise the neon.

BUT ..... If you put a high enough voltage on it .. I have seen the whole tube "reverse" ionise .. ie from anode to cathode. My video only partially shows this ... but when it does it you can clearly see that it is the outer "Grid" that is ionising.

Tell you what....
I will "dismantle" one.... (read smash it up) and post pictures.

Youtube video is at %50

 

[spec]

Hi Doc,
I have not found any buck boost chip that will handle any more than about 60 volts... otherwise my problem would be solved !

Hy again fire,
I'm no switch mode PSU designer but I have messed with them and do know some of the theory. Just to let you know that the design of a PSU with the following characteristics would not only be possible but would be fairly straight forward. It would not be as easy or cheap as a mains transformer approach, or possibly non isolated approach though:

(1) Input supply voltage 110V to 240V RMS or 156V to 340V DC
(2) Input frequency: 49Hz to 500Hz or DC
(3) Output voltage: 170V DC
(4) Output current 0A to 1.2A
(5) Input/output isolation: yes

spec

 

[fireant007]

Sorry spec,
Been down this road with both of them.

Texas instruments.
And On Semiconductor disagree.

From On Semi.

"Your question:
I am looking at the NCP1602 chip. Can you please tell me if the output voltage on this is adjustable down to 155v ? I will still need to have a "universal AC input from 90 to 265VAC ? Can it be done by adjusting the capacitor on Vctrl ?

has been answered:
The 6-pin PFC controller NCP1602 is designed to drive PFC boost stages, therefore 155VDC is too low for the output.



If you have more questions please visit: AskAnExpert application

Kind regards, ON Semiconductor"


Same reply from TI

 

[spec]

Hey spec,
I am just as interested in getting to the bottom of this as you. Even more so as I want a working circuit !
So please feel free to nag me as much as you like.

There are pictures identifying the anode around on the net .. but I can't find one to hand at the moment.

How ever ..... there are 2 leads. one is in the middle ... that looking through my jewelers loop .. just seems to go right up the middle.
That wire is the cathode.

The outer wire connects to to a grid that goes up the outside that is .. the anode .. the central wire is what gives off the electrons .. to ionise the neon.

BUT ..... If you put a high enough voltage on it .. I have seen the whole tube "reverse" ionise .. ie from anode to cathode. My video only partially shows this ... but when it does it you can clearly see that it is the outer "Grid" that is ionising.

Tell you what....
I will "dismantle" one.... (read smash it up) and post pictures.

Youtube video is at %50

UPDATE- wrong sense originally

From what you say, you have definitely got the anode and cathode the right way around, so no need to smash one of your tubes.
Can you try heating the tube in the test circuit and if that fails can you shine a bright LED type torch into the tube.
spec

 

[spec]

Sorry spec,
Been down this road with both of them.

Texas instruments.
And On Semiconductor disagree.

From On Semi.

"Your question:
I am looking at the NCP1602 chip. Can you please tell me if the output voltage on this is adjustable down to 155v ? I will still need to have a "universal AC input from 90 to 265VAC ? Can it be done by adjusting the capacitor on Vctrl ?

has been answered:
The 6-pin PFC controller NCP1602 is designed to drive PFC boost stages, therefore 155VDC is too low for the output.

If you have more questions please visit: AskAnExpert application

Kind regards, ON Semiconductor"

Same reply from TI

Hi fire

Those answers do not surprise me one bit but that does not alter my assertion.

spec

 

[fireant007]

Try TI's webench app .. it's amazing for developing SMPS ... spat the dummy for me and my requirements.
Could make 4 0.25A SMPS's at 24vdc input .... that was it.

On Semi have a spreadsheet.
Poo's it's pants when you put in a lower voltage out than is coming in.

Have a crack. https://www.onsemi.com/PowerSolutions/supportDoc.do?type=tools
look for the NCP 1602 tool.

 

[fireant007]

UPDATE- wrong sense originally

From what you say, you have definitely got the anode and cathode the right way around, so no need to smash one of your tubes.
Can you try heating the tube in the test circuit and if that fails can you shine a bright LED type torch into the tube.
spec

You will see all when the video is up and running ... all mystery of the tube with be reveled !
I have made it VERY clear. (apart from my fat hand in the way sometimes

 

[fireant007]

Hi fire

Those answers do not surprise me one bit but that does not alter my assertion.

spec

I agree .. you could make a buck boost converter fer sure.
It's just that I can't find a chip that will to it.

How would you do it ?
Without a chip ?

Edit: this conversation only applies to input voltages greater than output voltages.
Therefore .... you need to buck if the input voltage is higher.

I have been looking ... HARD ... for such a circuit.

 

[spec]

This must be a day for links: https://www.electro-tech-online.com/threads/how-to-diy-170vdc-power-supply.21569/

 

[spec]

I agree .. you could make a buck boost converter fer sure.
It's just that I can't find a chip that will to it.

How would you do it ?
Without a chip ?

Edit: this conversation only applies to input voltages greater than output voltages.
Therefore .... you need to buck if the input voltage is higher.

I have been looking ... HARD ... for such a circuit.

It is all pretty straight forward. I will see if I can find a standard circuit if not I will post a circuit to illustrate the principle.
spec

 

[dr pepper]

At first I tried to control my bar with a npn tranny to ground essentially a resistor in the cathode, it didnt work too well the bar tried to illuminate part way up the tube, theres some strange thing with these tubes like that, the IV19 I think it is has a seperate cathode at the base of the tube, to 'anchor' the glow.

For a one off I'd reccomend the torroid with split primarys, esp if your not really into switchers.

Anyway I think I'll bow out of the conversation now, sounds like you've allready had plenty of reponses.

 

[fireant007]

It is all pretty straight forward. I will see if I can find a standard circuit if not I will post a circuit to illustrate the principle.
spec

Spec...
Please don't bother.
I have seen all those circuits before ...
None of them supply more than 80mA
And all of them rely on a DC input !
I have haunted google mate .. I really have.

https://threeneurons.wordpress.com/nixie-power-supply/hv-supply-kit/
Not a bad one....

https://www.tayloredge.com/storefront/SmartNixie/PSU/index.html
Better one (I have 2 on order for another project).

Also have this one... bought for 400Vdc .. decatron project.
Very low amps.
https://www.ebay.com/itm/141837951981?_trksid=p2057872.m2749.l2649&ssPageName=STRK:MEBIDX:IT

 

[fireant007]

At first I tried to control my bar with a npn tranny to ground essentially a resistor in the cathode, it didnt work too well the bar tried to illuminate part way up the tube, theres some strange thing with these tubes like that, the IV19 I think it is has a seperate cathode at the base of the tube, to 'anchor' the glow.

For a one off I'd reccomend the torroid with split primarys, esp if your not really into switchers.

Anyway I think I'll bow out of the conversation now, sounds like you've allready had plenty of reponses.

No worries Doc
I will probably go down the torroid road myself.

Spec has been a very big help to me ! .. hats off to the man !

 

[fireant007]

Insert banana rama song here.
This is what I want .. what I really really want !

https://www.learnabout-electronics.org/PSU/psu33.php