My Nixie Clock

[Jean-Pierre G]

Hi all,

Here is my nixie clock :

• 6x IN14 russian tubes
  
• 6x Neopixels Leds
  
• 2x HV5530 high Voltage drivers
  
• 1x 210V High Voltage step up power supply
  
• 1x 5V step down power supply
  
• 1x ESP32 microcontroller
  
• 1x radar sensor
  
• 1x light sensor
• 1x (option) temperature + pression sensor
• 2x home made pcbs
  

 

And one Android application speaking to the ESP32 with Bluetooth Low Energy

Just add a piece of wood (from Vietnam) a CNC mill, a lot of hours

 

See it in action :

 

JP

[gregebert]

The home-made PCBs look very good; I tried making my own more than 30 years ago and they were difficult to solder because the holes were not plated-thru and the tinning solution left the pads over-oxidized.

Nice to see a simplified design that can be built with a single-sided PCB.

[Jean-Pierre G]

Thank you !

I got this idea to stack two pcb because I wanted the neo pixels not
to be "jailed" into the nixies pins.
So due to this I needed two pcbs...
Finally it offers a lot a compacity in the design as the ESP32 and the
power supply have their own 2 sides pcbs .

2018-06-18 7:12 UTC+02:00, gregebert <greg...@hotmail.com>:

> --
> You received this message because you are subscribed to a topic in the
> Google Groups "neonixie-l" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/neonixie-l/fMrqiBukFl8/unsubscribe.
> To unsubscribe from this group and all its topics, 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/48eaf9d5-5b03-44f2-a001-935a28eb8552%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>

[Flynn Harrison]

im currently trying to do somthing simmiler with only 1 pcb, but ive been sort of stuck on the best method for my logic level conversion (3.3 to 12v) for the esp32 to HV shift register. so im wondering what method did you use, i.e MOSFET or BJT ect.. ?

[Jean-Pierre G]

Simple mosfet + 2 resistors !

--

You received this message because you are subscribed to a topic in the Google Groups "neonixie-l" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/neonixie-l/fMrqiBukFl8/unsubscribe.
To unsubscribe from this group and all its topics, 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/1cd09ac7-2368-4c47-9b2d-a7f79d6dd857%40googlegroups.com.

[Flynn Harrison]

thanks, did you use somthing like a BSS138?

On Monday, 18 June 2018 22:46:14 UTC+10, Jean-Pierre G wrote:

Simple mosfet + 2 resistors !

Le lun. 18 juin 2018 à 14:44, Flynn Harrison <fl...@belmontharrison.net> a écrit :

im currently trying to do somthing simmiler with only 1 pcb, but ive been sort of stuck on the best method for my logic level conversion (3.3 to 12v) for the esp32 to HV shift register. so im wondering what method did you use, i.e MOSFET or BJT ect.. ?

[Jean-Pierre G]

Well I don't remember exactly... It's the smd version of 2N7000 n channel mosfet.

Or any other tiny mosfet !

To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/66a3bfa1-19f9-4d42-8544-226bf1ee93c3%40googlegroups.com.

[Jean-Pierre G]

The exact reference is 2N7002  datasheet here : https://assets.nexperia.com/documents/data-sheet/2N7002.pdf (any manufacturer will be ok)

[Jean-Pierre G]

and schematics (with another reference !)

[Nick]

Rather than doing it all with discrete components, have a look at the TI TXB010x series of bidirectional level shifters.

They are cheap, come in a range of packaging including soic, and in various widths, i.e 2, 4, 6 or 8 channels.

Nice chip.

Nick

[gregebert]

Be careful when cascading HV5532's. The datasheet does not specify a minimum propagation delay, but it DOES require a minimum hold-time of 10nsec. Even though the datasheet says you can cascade the data-out/data-in pins of multiple devices, there are no spec numbers to support this so the next-best thing is to route the clock signal in the opposite direction of the data. I suspect the minimum prop-delay of the data-out pin is at least 10nsec, but I cant prove it.

I tend to overdesign with paranoia and pessimism, so I actually use a separate clock pin for my HV5532's when I cascaded 2 of them, so there is no possibility of a timing issue as long as I sequence the clocks correctly. Also, my clock has a total of six HV5532's (I have 14 tubes) , but I dont cascade them into a single chain. Instead, I break them into 3 parallel chains and use a level-translator IC (MC14504) to shift from the 3.3V FPGA to 12V for the HV5532. It's been running flawlessly for 2-3 years now.

[John Rehwinkel]

> Rather than doing it all with discrete components, have a look at the TI TXB010x series of bidirectional level shifters.
>
> They are cheap, come in a range of packaging including soic, and in various widths, i.e 2, 4, 6 or 8 channels.

Those are handy chips (I've used 'em with a Raspberry Pi to drive LED matrices), but they're not suitable for 12V signals like these HV chips use.

- John

[Nick]

True. What about the CD4504B? 20V tolerance.

Nick

[John Smout]

I have just read about the 25% tax that Mr Trump is going to put on Chinese imports to the USA from July 6th. Most electronic components will be taxable.

What effect will this have on US manufactured electronic goods or the hobbyist? Will it delay the import of Chinese goods and how will the consumer pay the tax?

John S

[gregebert]

A lot of the items I purchase via Ebay are labeled as "gifts" by the sender in China, so I suppose those might slip under the radar.

[Jean-Pierre G]

Hi
Why do you take care of such a small propagation delay ?
I simply generate the clock with 10ms period. And it works like a charm with two hv5530 for 6 tubes.

JP

[gregebert]

For setup time, I agree the 10ns is insignificant for a 10msec period. But for hold-time, the timing analysis starts and ends on the same clock edge. A positive hold time, in this case +10nsec, means that you must hold the data_in signal stable for 10nsec after the sampling-edge of clock, which is the falling-edge for the HV5530. If the data_in signal comes from a part that propagates in, say 7nsec, you would violate the hold-time spec by (10 - 7) = 3nsec. It may not sound like a lot of time, but hold-time violations are not forgiving. You should see what we do at my day job to fix a 25psec hold-time violation.

Another less-obvious source of timing difficulty is clock-skew; if the clock-delay to the receiving part is *longer* than the clock-delay to the sending part, the resulting difference in delay (clock-skew) worsens your hold-time. A common practice is to route clock and data signals in opposite directions.

As I said, I overdesign things. I've seen too many timing problems on PC boards and inside ICs causing failures that are very difficult to find and fix, so I always take the worst-possible operation for granted, and design around them.