I've been searching for an ideal voltage step-up method for a circuit I'm designing. Ideal in this case meaning compact (aprox. <=7mm), in at least one dimension.
I've been experimenting with (and previously posted about) fancy CCFL transformers such as piezo transformers. While small, they appear to be too difficult/costly to operate efficiently enough for my application.

My output requirement is circa 500-1kv at 5-10 ma.
To be operated by a thin power source (such as 2032 3V coin cells), using small components.

Before moving on to a complicated boost converter/multiplier design I found, I was wondering if there is a simple solution to this problem?
Are there cheep tiny transformers out there on the internet, or any other simple means of achieving this which my novice experience level has no knowledge of?
-Thanks for any advice

 

You want almost 2 amps out of a coin cell designed to supply less than 0.0005 amp and convert that to 1000 volts in 1/4 of an inch?
I think you should run the math again.

 

You want almost 2 amps out of a coin cell designed to supply less than 0.0005 amp and convert that to 1000 volts in 1/4 of an inch?
I think you should run the math again.

No...
Where did you get 2 amps?
I wouldn't be using one cell, I would be using several

 

What's the duty cycle? How much energy do you think X amount of 2032 coin cells can deliver?
http://www.embedded.com/electronics...h-energy-can-you-really-get-from-a-coin-cell-

 

No...
Where did you get 2 amps?
I wouldn't be using one cell, I would be using several

Duh!
Power OUT must be less than Power IN. Sometimes it will be much less.
Power Out = 1,000 V x 10 mA = 10 Watts
Let us assume for grins that your process is 80% efficient
Input power is 10 Watts / 0.8 = 12.5 Watts.
Input current required is 12.5 Watts / 3 Volts = 4.166 Amps.
In a boost converter you will need an inductor that can handle twice that current or 8.32 amperes.
To do that in a small space in any number of dimensions is just plain impossible.

It is like making rigid toys out of cream cheese.

 

Where did you get 2 amps?

My output requirement is circa 500-1kv at 5-10 ma.

500V x 0.005A = 2.5 watts
2.5 watts/3volts = 1.666amps @ 100% conversion efficiency

Service capacity: 200 MaHr
Radio Shack Enercell Battery Guidebook, 1985

0.2 amp hours / (1&2/3 amps) at 100% energy conversion to 500 or 1000 volts = 7.2 minutes lifespan
If you can even get that much current out of a battery designed for 0.0005 amps

I gave you every possible advantage in the math and your requirements are still impossible.

 

500V x 0.005A = 2.5 watts
2.5 watts/3volts = 1.666amps @ 100% conversion efficiency

Service capacity: 200 MaHr
Radio Shack Enercell Battery Guidebook, 1985

0.2 amp hours / (1&2/3 amps) at 100% energy conversion to 500 or 1000 volts = 7.2 minutes lifespan
If you can even get that much current out of a battery designed for 0.0005 amps

I gave you every possible advantage in the math and your requirements are still impossible.

I wasn't as generous in the math, but the result is the same.

 

I was able to achieve slightly over 250vrms 2.5 ma in a 100k load with the piezotransformer attempt, using 4 3V 2032(3 in series, one in parralel).
That's just a bit less than what I need.
Take bare minimum expectations ~300-400V 3-4ma, longevity is of no importance to me. I don't care if it runs out of juice in 5 seconds.
Clearly these batteries can't release that much power at any given moment time, and it wouldn't have to in this application.

The closest example I found of a circuit that does something similar successfully is one of those shocking prank pens/bubble gum packs. I do not want to shock anybody, but it seems as if they solve a similar problem.

They do it with four tiny coin cells, a small step up transformer, and a little white component (capacitor?)
They reach around 300v, and being that you can feel a shock it's probably in the couple of ma.

I'm failing to figure out how I can can get something similar done...