Quick question for you electrical guys

[Badot]

I'm working on a little project of mine and need to turn a battery to a very large voltage. Is this circuit feasible? Simulation runs decently, comes out a tad over 200 volts with a 3v input, just not sure if the components are 'practical' for simulation or building. Power efficiency isn't critical, but I want 200-300 volts as simple as possible.

That'd be a 455khz ceramic resonator, NPN transistor, 5 microhenry inductor (not 6 as in the picture). Cap is very small to 'charge' fast in simulation, else my laptop would be there for hours.

[john]

Look at this page.  Very high voltage from low input.  At least it will give you maybe some reference.  http://www.syntiac.com/tens.html

[Kiwingenuity]

In theory - yes, that circuit will work.. There are some big limitations with using Workbench / multisim using ideal components however...

A single transistor is not going to probably switch fast enough (or attain saturation at least) at that sort of frequency.  It is possible that a Darlington or even using a FET could work, but lowering the frequency would lead to less problems... not least of which will be inductance in your circuit board which will have very undesirable effects.

You will also probably have trouble getting an inductor *exactly* at the value - any variation with inductance will considerably affect the output voltage (and even when the temperature changes you will see some variation).

The output diode may also suffer from the high frequency - a Schottky diode may be able to deal with switching that fast, not too sure.

Last time I built a little flyback DC-DC boost converter like this I had about a 60kHz switching speed which meant I didn't have too many issues.

Try this page - it shows the limits you run into quite well and saves a lot of tinkering time.
http://www.dos4ever.com/flyback/flyback.html

You can also cascade boost circuits - that can sometimes be more reliable than trying to get a single stage boost.

Let us know how you get on 

[Badot]

And so it becomes a not-so-quick question 

First, I'm not sure how to tell how fast a particular transistor can switch and google doesn't seem to be helping much. I turned up a few of these things: http://www.onsemi.com/PowerSolutions/product.do?id=MJE181 which has some f-sub-T thing listed as 50 mhz in this instance. Is that the switching speed?

As far as diodes go... 455000 hz would be a cycle every 2.198 microseconds (10^-6) so a reversal every 1.1 microseconds. On the wikipedia page on Schottky diodes it states standard diodes "reverse recovery time can be in the order of hundreds of nanoseconds and less than 100 ns for fast diodes" where nanosecond is 10^-9... so that'd be looking at something along the lines of a reversal every 1.1(E-6) seconds where the diode takes around .1(E-6) seconds to switch... anyone have any idea if that would be a problem?

EDIT: Another question... do I have to worry about getting 'counterfeit'/misrepresented electrical components off ebay? I assume yes.

[Kiwingenuity]

It gets rather confusing rather quickly with these tricky little circuits..

Without getting too complicated - you end up with other problems with the oscillator / duty with driving a transistor - especially if you want to regulate the output. Add to that if you don't regulate things you may get output oscillations which can get some very large spikes and fry things..

I know from recent experience that the f-sub-T thing (Bandwidth gain product) of most general purpose NPN transistors = not enough gain to turn on (saturate the junction and de-saturate) as the frequency goes up above about 100kHz.  The circuit I designed recently used an IR810 MOSFET as I found it was favourable as it was easier to control the gate voltage rather than a base of a transistor which especially had problems turning off. The circuit ended up using an LM3900 for oscillator and PWM regulator and got all very complicated..

After that little exercise I found something which would have made my life a whole lot easier...
http://www.semtech.com/images/datasheet/sc2604.pdf

Other type may be a Linear Tech LTC1156.