my Z5660M nixie clock

[Jonathan F.]

I want to share this project with you. Its about 90% finished.

I made a nixie clock with Z5660M tubes.

My basic idea was to develop clock pcb that i can use for future projects, and a second pcb for tube sockets. This allows me to take use the main pcb for different projects (eg. IN-18 Clock).

I'm currently design a website for all my nixie,vdf,plasma stuff, so don't wonder about the web address on the pcb's, that's my future site-address.
I have choosen the name SWISSNIXIE, because i live in Switzerland, and i like nixies and do projects with them =)


Here are some details:

"Nixie Clock Controller Board" aka SN-NCB-01

• Atmega328 with Arduino-Bootloader
• HV5122 Chips for controlling the clock, gives 6x10 Outputs for nixie tubes, and 4 outputs for controlling neon lamps and stuff
• DS3132 RTC, its very accurate and has built in quarz. http://datasheets.maximintegrated.com/en/ds/DS3231.pdf
• Terminal block for usage of a GPS or DCF77 receiver and a "enable" input. The GPS/DCF is not code yet. For the enable input i use a motion-detector. If something moves, the clock will switch on for about 20min.
  
• 4 Buttons for Settings and a Potentiometer for Backlight
• Power supply based on MAX1771. Its based on the fantastic supply i found here : http://www.desmith.net/NMdS/Electronics/NixiePSU.html
  

"Tube Board B13L"

• Can be used with all large B13B Sockets
• IN-3 neon bulbs or smiliar can be used as decimal spaces. all 4 individually controllable.
  
• WS2812 LED's for backlight.  (Mounted on a small board under this one)

Case & Assembly

I had a few pieces of wood at home, so i choose a wooden case, because it was the easiest and cheapest material. The case is made of one solid block, no glueing/screwing. I drilled out 6x 31mm holes and 2x 13mm holes. Then i milled out a space to fit the pcb in. Case finish was made with sand paper and natural wax.

For the column separators i used lab probe glasses and cut them to the right size. They are mounted with silicone directly on the pcb.

What do you think of the project? If someone has tips/advises please tell me. If you have any questions, just ask :))

[gregebert]

I'd suggest using the DS3232; it has the same pinout , accuracy, etc as the DS3231 and it also has some general purpose registers that are battery-backed-up in the RTC. They could be very useful for your microcontroller for non-volatile storage of setup info, etc. 

[John Rehwinkel]

> I want to share this project with you. Its about 90% finished.
>
> I made a nixie clock with Z5660M tubes.
>
> My basic idea was to develop clock pcb that i can use for future projects, and a second pcb for tube sockets. This allows me to take use the main pcb for different projects (eg. IN-18 Clock).

Makes sense to me, I've done designs with similar thinking.

That's a good-looking board layout, I like it.

- John

[Mich...@aol.com]

Is there going to be an order form?

 

Michail

 

In a message dated 3/5/2016 12:10:23 P.M. Pacific Standard Time, jfre...@gmail.com writes:

90% finished

[gregebert]

Regarding the display board, be sure to add mounting holes (I didn't notice any in the photo).

I'd recommend placing the driver device on the display board, and use a constant-current regulator on each anode.

I also used socket pins (instead of actual sockets) like your board, and they worked great.

If there's interest, I'll post pics of my latest clock. It has fourteen IN-18 tubes for month/date/year & hours/minutes/seconds.

[Allen]

Color me interested!

Allen Wisbey, W1SBY

--
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/41f3f932-6dd8-46da-9619-87d2318c5c31%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jonathan F.]

Hi

I choose to place the drivers on the main pcb, because of space and changeability.
Yes true, no mounting holes here because of my wooden construction. I also have a layout with matching holes to the main pcb (make the tube board the exact same size.)

What would you suggest for constant current regulator? I used a 10k resistor at 165V input.

[gregebert]

I attached the schematic for the adjustable current regulator I used on my latest project. Each anode has a separate regulator. The PDRV signal sets the driver current and requires no current because it controls the gate on PMOS devices. Current is sensed by the source resistor (R7) which provides negative feedback. The zener diode protects the PMOS gate-source voltage from ESD, transients, noise, etc. The 10meg bleeder resistors protect the PMOS devices from ESD.

[NIXIE : TIMES]

Wonderful design! I love it! How did you managed to achieve a clean PCB? By ultrasonic cleaner?

[NIXIE : TIMES]

I'm really curious about the way you generate PDRV. Is it from a voltage divider with potentiometer or automatic control circuit?

在 2016年3月7日星期一 UTC+1上午1:23:08，gregebert写道：

[John Rehwinkel]

> I attached the schematic for the adjustable current regulator I used on my latest project. Each anode has a separate regulator.

What's the purpose of R8 (the 1k series resistor)?

- John

[gregebert]

VXX200 is the +200V supply; the PDRV signal varies from roughly +200V to +188V. I used an isolated 12V DC-DC converter to generate "negative 12 volts with respect to VXX200". There's a potentiometer across the 12V DC-DC converter's output, and the wiper connects to PDRV. That way, I can adjust PDRV between 0v and -12v with respect to VXX200. Changing PDRV changes the anode current.

R8 limits the potential for over-current in the nixie tube in case something goes wrong with the current limiter, such as a power-up transient. Also, I can infer the anode current by measuring R8's voltage drop. The way the PCB is laid-out, I can't access R7.

C3 provides filtering of any noise that might get coupled onto PDRV.

The 'unlabeled' anode resistors (2 per IN-18 tube) were put on the PCB in case only one of the anode pins should be used. At the tme I sent my PCB to fab, I wasn't sure. As far as my research went, I could find no problems using both anode pins.

[Jonathan F.]

I cleaned the pcb with a pcb cleaner "Kontakt LR " from Kontakt-Chemie. The english name is "Kontakt PCC" just sprau it over and brush lightly with a fine brush. Then use hot water to remove the cleaner and use a air-duster to blow the water away. Let it dry for about a day.

The R8 is acutally a 10k Resistor (SMDcode 1002 =10k) this is used because the poti is a 5k, so you can set the voltage more exactly. So 10k is the lowest setting and 15k the highest. The voltage divider has to output 1.25V feedback to the MAX1771. So the first resistor is a 1.5Meg, then the poti and 10k in series. The feedback output is between 1.5Meg and Poti. I will post a schematic later this week

[Mitch]

I can't seem to get my boards as clean as they could be. I spray with a pc board cleaner, then wipe and let it dry, and deal with the residue. I never thought about washing with water. I'll give it a try. Thanks.

[David Forbes]

Mitch,

PC board assembly factories don't use rosin flux these days, because it
can't be cleaned without dangerous chemicals. My city is still
remediating the ground water after the solvent trichloroethylene (TCE)
got into it from the local missile factory and poisoned a bunch of
people in the eighties.

I've been using water-soluble flux for 30 years. I use Kester 2331
organic liquid flux with Kester 331 organic core solder. It's rather
active, so you have to wash it off within a day or the board gets
corroded. I wash the boards under hot running water while scrubbing with
a toothbrush, then remove the water with compressed air.

Another benefit of this flux is that it has super wetting action, which
makes it easy to solder 0.5mm pitch TQFP packages with a standard
soldering iron. I can even solder 0.4mm pitch parts.

There is also 'no-clean' flux, which leaves less residue than standard
rosin core solder. That's what is used in many factories where low cost
is king, such as consumer audio electronics. It's not show quality,
though. The Kester organic flux makes clean, shiny solder joints.

On 3/8/16 5:55 AM, Mitch wrote:
> I can't seem to get my boards as clean as they could be. I spray with a pc
> board cleaner, then wipe and let it dry, and deal with the residue. I never
> thought about washing with water. I'll give it a try. Thanks.

--
David Forbes, Tucson AZ

[Nicholas Stock]

I've had success with washing boards with isopropyl alcohol then washing with water and drying. I use regular solder, nothing out of the ordinary...takes a few scrubs with IPA and the boards look clean afterwards.

--
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/56DEF85A.3030406%40dakotacom.net.

[Mitch]

Thanks for the advice. I'll try both Kester 331, and alcohol and water with regular solder.