First nixie clock

[newxito]

Hi, I’m new in the forum. I’ve finally managed to build my first nixie clock. It’s a 6-digit clock with 4 x Z5660M and 2 x IN-8 nixies.

I’ve decided to build the clock with direct drive so I’ve used 8 x 74HC595 and 64 x MPSA42 transistors.

The controller is a Atmega 328P, the 170V psu was made with a MC34063, coil, transistor and mosfet. The psu seems to work fine, no audible noise and the components remain cool.

I had no idea how to shut down the 170V so I’ve used a relay to cut the 12V supply to the 170V psu. I’m sure there must be a better way to do this.

Finally, I’ve added a GPS time sync, a motion sensor and a mp3 module, the clock plays a sound every hour.

I was surprised how easy it was to write the software. In the latest release I’ve implemented a simple cathode poisoning prevention function (just heard about this problem). For the next release I have planned to use the pps signal for a more accurate GPS time synchronization.

I know, the whole thing looks like a nightmare… but it works :-)

Regards

[gregebert]

Congratulations, you've avoided a lot of problems with direct-drive and the MPSA42 drivers. I have to admire your bravery with the point-to-point wiring. Very impressive, especially for a first clock.

Nothing wrong with the relay to control the HVDC supply; just make sure you have a diode across the coil to clamp the inductive spike when the coil is de-energized. It's far better to shut the HV down when the tubes are off, rather than letting it run with no load.

[Roddy Scott]

I second Gregbert's remarks of the bravery on the wiring! Epic to say the least.

When people see a finished clock they have no idea about the amount of work that is contained in the casing as all they see is the case and Nixies.

Well done!

As to bits for clocks, the attached is part of a clock I am building just now. 

[GastonP]

It looks great!
If I could make a suggestion, that would be to put some kind of grill for the speaker in the front. Speakers are known to be kid and cat magnets and that one looks very effective on that regard. :)
Going the direct drive way frees you of a whole set of problems and complexities.
Keep them coming!

Gaston

[alb.001 alb.001]

I think we should crown you as the new king of point-to-point wiring.  The clock looks great

regards Phil

I know, the whole thing looks like a nightmare... but it works :-)

Regards

 

--
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/7ed97fca-9de1-4a50-8403-f3f69cbdf33f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

 

[David Forbes]

That's quite a construction! Your wire salesperson must be very happy.

You might want to turn your GPS receiver so that the top of the chip
faces the sky. It will be able to see more satellites that way.

> --
> 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

> <mailto:neonixie-l+...@googlegroups.com>.

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

> <mailto:neoni...@googlegroups.com>.

> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/7ed97fca-9de1-4a50-8403-f3f69cbdf33f%40googlegroups.com

> <https://groups.google.com/d/msgid/neonixie-l/7ed97fca-9de1-4a50-8403-f3f69cbdf33f%40googlegroups.com?utm_medium=email&utm_source=footer>.

> For more options, visit https://groups.google.com/d/optout.

--
David Forbes, Tucson AZ

[Dave Brown]

Well done!
I'd be interested in the schematic for the high voltage psu- have been
toying with using the 34063 for that myself.
DaveB, NZ

--
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 an email to neoni...@googlegroups.com.

To view this discussion on the web, visit

https://groups.google.com/d/msgid/neonixie-l/7ed97fca-9de1-4a50-8403-f3f69cbdf33f%40googlegroups.com.

[newxito]

Sorry, I don’t have any schematics for this thing, but I remember that the HV psu is based on this great video with some modifications. I have changed the BAV21 diode to BYV26C,  connected pin 1,7, 8 directly to +12V without the 3 resistors and changed the value of the 2.2nF cap to another value (unfortunately I can’t see the value I have used). That worked for me.

https://www.youtube.com/watch?v=I2AdWwZpPzY

9:53

[newxito]

Thank you. The wiring wasn’t so hard but for the next clock I will definitely try to design a PCB. I hope that the autorouter will do the hard work for me.

[gregebert]

It's hard to justify in so few words, but you're better off hand-routing a PC board. Overall, it will take more time to layout a PC board than to breadboard it once. However, a PC board will give you a much neater and reliable product. If you end up making 2 or more, the PCB will save you time and money vs breadboarding.

[Mark Moulding]

For almost all of my new projects, I use a PCB now.  Even in the prototype stage, and knowing that I'll have to do some "blue wire" modifications, I find that the reliability is much higher, and trouble-shooting is easier.  By using the design rules check, I can guarantee that the PCB matches the schematic, which saves time tracking down wiring errors (because there aren't any), and as Greg pointed out it's a much neater product when finished.  And having done it enough times now, I find that I can usually go from the schematic to the PCB layout in less time than it takes to use point-to-point wiring; counting board stuffing, the first article is probably a wash time-wise.

I also hand-route almost exclusively.  Occasionally, when I have a bunch of small discretes, don't care too much what it looks like, have plenty of room, and am feeling lazy, I'll let the auto-router loose, but that's a lot of conditions to be met, and it doesn't happen often.

~~
Mark Moulding

[David Forbes]

Mark,

I agree with this. I am a big fan of PC boards.

However, I also built my first Nixie clock, and my first oscilloscope
clock, on perfboard. The scope clock logic board started out life on a
solderless breadboard, as did my first Nixie watch.

There's something to be said for doing it the hard way the first time.

On 9/29/2016 9:30 AM, Mark Moulding wrote:
> For almost all of my new projects, I use a PCB now. Even in the
> prototype stage, and knowing that I'll have to do some "blue wire"
> modifications, I find that the reliability is much higher, and
> trouble-shooting is easier.

[newxito]

For the next PCB based projects I’ m not sure if I should stay with the 74HC595 and the MPSA42 transistors or if I should start using modern components like the HV5622.  Is there a recommended way to implement direct drive?

[David Forbes]

I recommend the Texas Instruments SN75468 Darlington driver IC instead of all
those transistors. It has seven 100V transistors. The clamp diode pin 9 wants to
be connected to 1/2 the Nixie supply voltage, about 90V. A resistor-based
voltage divider will work.

I use a lower voltage part in my Nixie watch, the TD62083. It only handles 50V,
though, so you cannot blank a digit cleanly. The SN75468 allows clean blanking,
with the clamp diodes connected as I described above.

> --
> 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

> <mailto:neonixie-l+...@googlegroups.com>.
> To post to this group, send email to neoni...@googlegroups.com
> <mailto:neoni...@googlegroups.com>.

> To view this discussion on the web, visit

> https://groups.google.com/d/msgid/neonixie-l/6d376eb3-0e31-49d4-a68b-7d34ef185c47%40googlegroups.com
> <https://groups.google.com/d/msgid/neonixie-l/6d376eb3-0e31-49d4-a68b-7d34ef185c47%40googlegroups.com?utm_medium=email&utm_source=footer>.

> For more options, visit https://groups.google.com/d/optout.

--
David Forbes, Tucson, AZ

[gregebert]

For direct-drive of true nixies (non-segmented), I'm really happy with the design of my last clock. The cathodes are driven by a HV5532, and the anode has an adjustable current-regulator. Having the HV driver and the deserializer+latch in a single device is really handy. The only drawback to this device is that it's spec'd to run at 12V, so a level shifter is needed.

I have 2 different clocks using cathode-current regulation and I'm not seeing any problems with the circuit.

My first nixie clocks use unregulated drivers, but the anode supply is rather high (+340V) so variations in the tube's voltage-drop over it's lifetime have minimal impact on the current. That approach wastes a lot of energy (about 330mW per tube * 6 tubes/clock * 4 clocks running 24/7 =70KwHr per year....), which is why I switched to current-regulators and lowered the anode supply to around 190V.

I hate to think how many kilowatt hours each year are wasted running my 6 nixie clocks; that adds up to 53 tubes.....