EEVblog #948 - Nixie Tube Display Project - Part 1

[Macbeth]

I agree, a more general solution for a wider variety of Nixies would suit viewers wanting to use the circuit for their own use.

Yes... now tell me wilfred, where does it end. What particular parameters would you be happy with?

Now pray tell how you come up with a universal "Nixie" PCB tube layout that will suit IN-12's but also every other possible "Nixie" on the planet?

[rs20]

I agree, a more general solution for a wider variety of Nixies would suit viewers wanting to use the circuit for their own use.

Yes... now tell me wilfred, where does it end. What particular parameters would you be happy with?

Now pray tell how you come up with a universal "Nixie" PCB tube layout that will suit IN-12's but also every other possible "Nixie" on the planet?

+1, but going even further I'd say the goal for a video like this should be about teaching the process; the specific circuit that comes at the end is more or less irrelevant. After all, if viewers want a completed, ready-to-go schematic, they can just find a schematic on the internet. All of the discussion and detail in the videos would be totally wasted on such viewers if they ended up just using the exact same gerber files as Dave.

[niekvs]

I agree, a more general solution for a wider variety of Nixies would suit viewers wanting to use the circuit for their own use.

Yes... now tell me wilfred, where does it end. What particular parameters would you be happy with?

Now pray tell how you come up with a universal "Nixie" PCB tube layout that will suit IN-12's but also every other possible "Nixie" on the planet?

I agree that obviously you can always find improvements, but this particular one doesn't have any meaningful drawbacks and has quite a benefit, and it teaches the viewer to engineer with good margins rather than 'hack' your way around an issue by (in this case) lowering the strike voltage and relying on a specific type of component. A more generally applicable solution, with no additional cost, would seem to make sense in this case and is good practice in general.

You can also turn your argument around, and ask yourself: which particular parameters or margins would you NOT be happy with? When you are designing something, do you also prefer components that, should at one point they are no longer available, cannot easily be replaced because the margins don't allow for this?

Of course, Dave can do what he wants: it's his project Also, it's clearly a one-off, so these arguments may not be very meaningful.

[rs20]

niekvs, you seem to be forgetting that your proposed solution requires separate darlington arrays and shift registers. This requires double the number of components, which is a far more concerning downside that the non-existent downsides of reducing the supply from 170 to 160V.

[niekvs]

niekvs, you seem to be forgetting that your proposed solution requires separate darlington arrays and shift registers. This requires double the number of components, which is a far more concerning downside that the non-existent downsides of reducing the supply from 170 to 160V.

This is true, but a couple of 595's can be had for pennies. Personally, I would multiplex it to limit the amount of components (e.g. in pairs of 2 or 3). But I agree it is a small downside. But to say that reducing the strike voltage to 160V is not a downside, I disagree: it's outside of the datasheet spec for the IN-12. I'd much rather operate my components within spec and have a more general solution that can be used for different types of nixies. If you prefer to operate your components out of spec, well I guess we disagree on that one. Just don't be surprised if it fails e.g. even when using a different batch of IN-12s.

[b_force]

There are many solutions to a certain problem.
As I posted a little bit back, if it's about just board size and total surface area, the current solution is also not optimal.
For the same reason I personally don't like a SN75468 (or similar) buffer. It's an all in one package, yes. But you do need an extra shift register.

What I personally would like to see, is some answers to questions that are still open. There were some ideas and suggestions with totem pole outputs for example.
We still don't really know if they are usable with maybe some tricks. Maybe they are not going to be used in this specific project, but we are already that far that it doesn't take much to at least check out some possible solutions, just because it is interesting or helpfull for future projects.

[splin]

A simple way to reduce the pre-bias voltage would be to connect another output from the microcontroller to the power supply feedback resistor network to change the supply voltage. It could normally run at say 140V to ensure that off elements don't glow even with a relatively low pre-bias voltage, but periodically raised to 170V+ for 50us or so to strike/ionize any digits which have changed or failed to strike earlier. Raising the anode voltage isn't strictly necessary, but it speeds up the strike time which depends on some random event such as a charged particle initiating ionization. In the dark it could take a second or more (including never) for a digit to strike but that would be unusual.

The anode resistor values would probably also need to be reduced in value as the anode voltage will rise considerably when no digits, or only the DP is lit, requiring a higher pre-bias voltage to prevent unlit elements glowing. Eg. Dave's 22k will cause a 22V rise in anode voltage when the 1.6mA digit current reduces to the .6mA or less of the DP.

You could probably get away with 33V drivers if all the Nixies are reasonably matched, but in general at least 50V would be required, even if you modulate the supply voltage, to allow for the relatively wide range of operating voltages of different Nixie types or changes over their lifetime.

Nixies are actually current controlled devices so ideally constant current drivers would be used. Using voltage drive + resistors is a proxy which is cheaper/more convenient but actually a bit tricky to design to accommodate all variances in tube characteristics, power supply and driver voltage and component tolerances etc. without having a high voltage supply of 180V+ and high pre-bias of 70V+.

The Burroughs application notes are helpful - see N102: http://worldpowersystems.com/archives/Burroughs/.

[EDIT] Here is another useful Burrough's datasheet/applications note: http://www.decadecounter.com/vta/pdf2/burroughs_616.pdf

[HugoPT]

I'm following this nixie tube awesome project but I can't understand how this the clamping circuit works
What is the theory involved on this clamping circuit to  drop the voltage?Maybe it would be a great topic to "fundamentals friday"

[Spuddevans]

What is the theory involved on this clamping circuit to  drop the voltage?Maybe it would be a great topic to "fundamentals friday"

This is how I understand it,

The problem is that many control IC's are limited on the voltage they can withstand, so switching close to 200V is out of range for most generic ic's. The way we are driving the Nixies is with "Low-side" switching, ie switching the "Ground", so when the nixie element is OFF the switching device (either a IC or a HV transistor) will "see" 170-190v across it's collector-emitter.

That's no problem if your output device is rated for that voltage, but most IC's are not. The way some IC's have got round it is to put a diode in each driver transistor and bring that out onto it's own terminal, you can then attach a Zener diode on that pin that is within the voltage rating of the output transistor(along with a resistor connecting the zener to HV+ to bias it). What this does is provide a means to "Clamp" the max voltage "seen" by the output transistor in the driving chip to the value of the Zener diode. This voltage is still below the voltage needed to turn on the Nixie.

Hope this explains it, I'm sure others can explain it better.

Tim

[rs20]

What is the theory involved on this clamping circuit to  drop the voltage?Maybe it would be a great topic to "fundamentals friday"

The problem is that many control IC's are limited on the voltage they can withstand, so switching close to 200V is out of range for most generic ic's. The way we are driving the Nixies is with "Low-side" switching, ie switching the "Ground", so when the nixie element is OFF the switching device (either a IC or a HV transistor) will "see" 170-190v across it's collector-emitter.

That's no problem if your output device is rated for that voltage, but most IC's are not. The way some IC's have got round it is to put a diode in each driver transistor and bring that out onto it's own terminal, you can then attach a Zener diode on that pin that is within the voltage rating of the output transistor(along with a resistor connecting the zener to HV+ to bias it). What this does is provide a means to "Clamp" the max voltage "seen" by the output transistor in the driving chip to the value of the Zener diode. This voltage is still below the voltage needed to turn on the Nixie.

That's a fine explanation, but I'd add a bit more numeric detail to clarify a few things. Specifically,

• If the Zener clamps the IC output voltage to, say, 50V, and the power supply is set to 160V, then the Nixie tube will have a voltage of 110V (160V - 50V) across it. The voltage across a device is all that ever matters.
• 110V is below the voltage required to turn on the Nixie tube, so it will appear to be off. Also important to consider if the Nixie tube is off, very little current will flow through it, so our Zener clamp will not have to dissipate very much power.

The point is that you can put 110V across a Nixie tube, and it'll appear to be totally off and only a tiny amount of leakage current will come through. We can use this to our advantage by switching from 110V (off) to 160V (on), and we can achieve these relative voltages by hooking an IC that outputs 0 - 50V on one side of the nixie tube, and a fixed 170V supply on the other.

I'm actually not sure how this would be turned into a FF, it's just a combination of knowing the non-linear VI curve of a Nixie tube and a super-basic application of Kirchoff's Voltage Law, which already has an FF video:

[NivagSwerdna]

A trivial question to some....

What physical size anode resistors would you use? SMD small-ish resistors seem to be pretty low wattages... through hole for the anode resistors?

Looks like my 0805 pad layout isn't going to cut it.    2512?

[Macbeth]

All this talk of 160 vs 170V etc, it appears that even as low as 130V is well enough for these IN-12 nixies, e.g. http://mightydevices.com/?p=379

Very nice using a micro to produce the switching and feedback to maintain that 130V HV Also working with multiplex too.

Just because a CCCP datasheet states 170V (but with no min/max to go with it) doesn't mean it is a fact. Frankly I believe the 170V was stated because it happens to be the half wave rectification of typical Soviet 240 VAC mains. No more, no less. The cheapest dirtiest possible HV DC at the time in Soviet Union.

[Macbeth]

Looks like my 0805 pad layout isn't going to cut it.    2512?

According to  tomasz.watorowski of MightyDevices he used 0805 and 0603. 

[FxDev]

Hello,

I post here and said finished to design PCB and yesterday I finally got it!

STM32F072 48pin MCU control everything; RTC, 5V to 170V converter, USB connection ext. It has buzzer and also RGB led. I used 300V sot23 NPN. Its dimension is only 4.5x4.5cm. I can be used with bigger Nixies too.

I will share all details after I finished my job.

[vladco]

i started playing with nixies around the same time as the first video appeared - i already had some IN-4 nixies and a power supply based on this https://threeneurons.wordpress.com/nixie-power-supply/ following the suggestions from Threeneuron i went with an Darlington array (w/ 47v zener) because i didn't feel like soldering 10 SOT transistors for each nixie tube (went with direct drive instead of multiplexing just for fun). I already knew i was going to use a 595 since i wanted to make something that could be daisy chained.

After watching the third video i saw the TPIC6B595 and i thought i could replace the 595 + uln2008 with it but because the layout was done already i decided to send it to OSHpark. Couple of weeks i received the boards the 595 + uln2003 works almost great exept the clamp voltage (47V) at least for my in-4 nixies is too low and when the nixie is not driven by the uln2003 some of the digits are partially lit (see link https://goo.gl/photos/dRfTq2JFU1gpqUnq6). Dropping the supply voltage from 180V to 165V made the effect less visible but still present i guess going with SN75468 and higher clamp voltage will fix the problem, anyway im glad now that i didn't go with the TPIC since it has a clamp voltage is only 50V.

[FxDev]

While waiting Dave, I will finish my second nixie clock with using dummy 300V sot23 NPN transistors   

Here is the first demo video.

[b_force]

While waiting Dave, I will finish my second nixie clock with using dummy 300V sot23 NPN transistors   

Here is the first demo video.

Nice with the little SOT NPN's. Which ones did you use?
Where can I find the project on your website btw?
(little difficult to search if people don't speak/read that language )

[FxDev]

Nice with the little SOT NPN's. Which ones did you use?
Where can I find the project on your website btw?
(little difficult to search if people don't speak/read that language )

I didn't share this tiny project yet because it is not done yet. 
I will share all details when i finished. You can look my other nixie clock here: http://www.firatdeveci.com/gecmise-donus-nixie-clock/
I don't know Google Translate is good or not for my website.

I use "MMBTA42-HT" with 1k base resistor.

Nobody asked me but I think nobody cares about how I can convert 5V to 170V that small area

[b_force]

There are multiple ways to convert 5V to 170V.
I have seen people using all sorts of things.
Easiest way is using a little step-up transformer.
You can also just use a boost converter with just an inductor.
Or just use a ucontroller as a boost converter (which i think is clever, because you're using one anyway).

[FxDev]

There are multiple ways to convert 5V to 170V.
I have seen people using all sorts of things.
Easiest way is using a little step-up transformer.
You can also just use a boost converter with just an inductor.
Or just use a ucontroller as a boost converter (which i think is clever, because you're using one anyway).

Yes, we generally use uCs in our projects so why can't we use them like a simple boost converter.

[vladco]

just a small update regarding my other post regarding the clamp voltage, after changing the clamp voltage to 75V (replacing also the ULN2003 with an SN75468) tubes are completely off when they are not driven.

@FxDev can you share your design for HV power supply driven from the MCU ?

[FxDev]

@FxDev can you share your design for HV power supply driven from the MCU ?

Of course!
I use TC4427 mosfet driver because my MCU is working with 3.3V. Of course we can use other methods but this is the  easiest way for me.
On the other hand, input caps are 1206 package 100nF and 1uF ceramic. Output cap is 2.2uF/250V. L1 is 330uH, I also tried 100uH to 680uH with no problem. I use PID algorithm for control this converter. Duty is around %70-80 when nixies are working. R24 is 150k/1206 and R25 is 2k2/0805. D1 is SK120 200V schotket diode.

I finished programming, now I'm writing small PC program to configure my nixie clocks. Few days later I will publish all!

[vladco]

Of course!
I use TC4427 mosfet driver because my MCU is working with 3.3V. Of course we can use other methods but this is the  easiest way for me.
On the other hand, input caps are 1206 package 100nF and 1uF ceramic. Output cap is 2.2uF/250V. L1 is 330uH, I also tried 100uH to 680uH with no problem. I use PID algorithm for control this converter. Duty is around %70-80 when nixies are working. R24 is 150k/1206 and R25 is 2k2/0805. D1 is SK120 200V schotket diode.

I finished programming, now I'm writing small PC program to configure my nixie clocks. Few days later I will publish all!

thanks for sharing, few more questions, switching frequency, saturation current of the inductor, max output power 

[boffin]

I'm just using some of Taylor Edge's smart nixies (need to get some more), and made a backplane using some perfboard.  (the analog discovery is monitoring the i2c feed in this shot)

[NivagSwerdna]

I managed to get my Christmas Nixie clock project mostly completed now.  I use a 62v zenner and HV5812s and have no glow on any element apart from the lit one.  Just need to laser cut an acrylic case now.