[HELP] Wristwatch with VFD tubes

[trixtan]

Hello,
I am designing a two digit wristwatch similar to those cool nixie watches you can find on cathodecorner.com. Since I am quite an electronics newbie (I am more a programmer dude) I wanted to try something different and maybe easier. My idea was to use two LD8025e VFD display instead of the two nixies.
These tubes are very different, they don't need 180V to work, but only 12V. Id like to design a power supply for my watch, and since It's the first time I do something this complex, I would like to share with you my findings and maybe you can guide me a bit through the path.
My objective is to learn as much as possible.

Talking about power absorbed current, the LD8035e have the following specifications:

• Filament: 26mA @ 0.8V
• Segments: 0.85mA @ 12V
• Grid: 70uA @ 12V

Furthermore, I need 3V for the microcontroller.

Now, In this project I have a big restriction, which is the available space. I'd like to use a single battery, and since the microprocessor will be always on and the VFDs are on only on demand, I was thinking that maybe the better idea was to use a 3V battery with a DC/DC step-up controller.
Being the first time I have to design such a "complicate" (for me) power supply, I was thinking to follow this flow:

• Estimate power requirements
• Decide which "topology" of power supply to use
• Search the market for integrated circuits

Is this a good approach? I am now at the step one, and this is my thinking:
Filaments use 26mA@0.8V. I have two filaments, if I put them in series I can say filaments use 26mA@1.6V, which is 0,042W.
Segments use 0.85mA@12V. I have max 7 segments which can be on at the same time (I will use multiplexing). This makes additional 0.00085*12*7 = 0.0714W
Grids use 0.00070A@12V. I have only one grid on at a time. 0,0084 additional watts

I will use a TD62783 8Ch Source driver to drive the segments. The datasheet specifies 0.52W max power consumption in recommended operation conditions.
So for the two VFDs and their driver, I have a total required power of circa 0,642W.
I still have to consider the power needed for:
 

• Power supply conversion
• microcontroller
• +20% safety factor

Am I thinking more or less right?
Thank you for your support

Enrico

[madires]

The TD62783's power consumption will be much lower since each segment needs just 0.85mA. It's the input for driving the first transistor  (3V * input current) plus the losses of driving the output stage and passing 0.85mA, multiplied by the segments driven. My guess is around 10-20mW in total. I haven't worked with VFDs yet, but at 12V a CMOS logic chip might also work. Maybe someone else has more insight.

[trixtan]

I took some decisions about the components I'll use for this project:

• Microprocessor: Microchip PIC16LF722. Reasons: enough pins, timer/counter present, works with 3V, small power consumption, I already have all tools to program it.
• Segments and grid driver: TD62783
• Power supply: CR2 3V 750 mAh Lithium battery
• VFD grid and segments power supply: DC/DC Boost converter TPS61040
• VFD filament power supply: DC/DC Buck converter TLV62565

For the moment I'm oriented to the 0805 SMD format. Next step: first layout of the circuit, I will post it here when ready.

[Howardlong]

It's up to you, but you may want to look at the more modern "Enhanced Mid-Range" PIC16(L)F1xxx series, there are usually equivalents to the older devices. They have a number of improvements that make life easier in the 8 bit world.

[mikeselectricstuff]

Why would you use a non-rechargeable battery like a CR2 for this, rather than a liPo?

You may find that the filaments draw quite a bit more when cold.

[timb]

Why would you use a non-rechargeable battery like a CR2 for this, rather than a liPo?

You may find that the filaments draw quite a bit more when cold.

Actually, he could do both! Illinois Capacitor has a new line of rechargeable coin cells (RJD series) that puts out *a lot* more current *and* has a much higher capacity than a normal LiR type coin cell! (In fact, capacity is actually close to or better than a non-rechargeable cell of the same size.)

http://products.illinoiscapacitor.com/seriesDocuments/RJD_series.pdf

Sizes start at 2032 (85mAH, discharge currents up to 170mA) and go up to 3555 (500mAH, discharge currents up to 1A(!)).

They're pretty great. I'm using two 3032's in a solar calculator project.

Apparently, most typical LiR style coin cells actually contain a square thin film battery element inside the round shell which wastes space. Whereas IC devised a method to fill and seal the entire battery element.