step up converter.

Can be less than $1.

You would need 40mA to 50mA current from 3V.

Somewhere between the 7V @20mA:

and the 30V @ 1mA:

Try playing with R1 (Load resister) to get the right output power.

Cool! How did you program that interactive user interface?

That one times out but could be a traffic jam on the web.

Cool! How did you program that interactive user interface?

That one times out but could be a traffic jam on the web.

Do you have a pwm output available, or can you afford to make a software pwm waveform? Figure 10-25% duty cycle at 20-500 KHz. Then a very simple inductive boost circuit will work.

John

Yes I can get a pwm output,but even a simple inductive boost cct is beyond my knowledge. The larkin cct is ok but there's lots of components and I need to turn it on and off. Easy with software pwm O/P... Easy with a boost chip as well but I can only find expensive ones....

Apache + Php + Spice3f5 + gnuplot

Yes, even with 3GHz quad xenon server with 8G memory. Every time I post, people are hitting the server all at once. Try a little later, it should be better.

Fet is a small logic-level n-channel mosfet.

L could be, say, a small 200-500 uH inductor.

Diode is almost anything, 1N914, whatever.

Cap is, say, 0.1 uF

Just futz with the pwm duty cycle and period until the output voltage is right. Try maybe 25% on, 100 KHz to start.

The output can't go below 3, so it's possible the piezo may peep a little down there. If so, you can use another fet or just a port pin to open its low side.

Or you could do this, which does go to zero when you quit driving it...

or this, with a small transformer,

John

And never, ever, let L go into saturation by keeping the FET on for too long. Else, phssstt ... POOF.

Might want to add here that the transformer needs to have an airgap when used as a flyback like this.

An RS232 level translator such as a MAX3232 (or equivalent from another vendor) can be an easy solution to this. It has a charge pump step up converter built in (just needs a few small caps) and you can either use the enable signal where available or put a diode in series with the output to drive a self-oscillating sounder by controlling the input. You could even drive a piezo sounder (non-self oscillating) directly and give it 40v p-p.

The MAX3232 is guaranteed to provide 10V between the outputs when driving 5mA total. (2.5mA each).

kevin

It will drag down the 3.3 rail lithium battery, the low-voltage reset will kick in, and everything will turn off!

A resistor from gate to ground might be prudent, I admit.

Hey, this is Digital Power, the wave of the future!

John

Thanks for all the answers and help. I have enough to get a solution I think. Not going to 0 is ok as the piezo is cpu controlled and puts out a hi(3V3) for off and the pwm will be off. neat idea on the 40Vp-p with a max chip too!!

One last thought: Zetex (and others) make tiny LED driver ICs that are meant to do exactly what you want. They come in regulated and unregulated versions and make up a boost supply as others have mentioned. ZXSC310 is just one example.

... after which the LV reset lets go, the uC wakes up again and then the whole thing starts pumping. Hey, a new oscillator!

Here's hoping that it's not an industrial grade cell that decides to go "exotherm".

A SW-engineer once proclaimed that since Max Planck everything can be reduced to digital. And this is a guy who plays the guitar and likes tube amps.

on

Who makes 56 Henry inductors in 0603?

John

Who makes 56 henry inductors these days in anything smaller than laser printer?

Jim

-- "If you think you can, or think you can't, you're right." --Henry Ford

Hammond manufacturing:

message

on

Should be possible some day soon, with nano-technology. Of course, then it'll saturate at a few micro amps so the market would be a bit smallish.

Le Sun, 12 Aug 2007 17:28:13 -0700, Joerg a écrit:

Hmmm, what can you do with 56H nano-inductors and nano-tubes?

Unlimited dreams. Nano motors and nano generators, etc.