Learning how to design power supplies

Yes, You have to study electrical/electronic course in order to understand each of the electronic component and it's function, calculation, formula, and e.t.c. You also need to understand the electrical theory and the electricity fundamental

What do you mean by power supplies?
Are you talking about 12v at 1 amp? or about high voltage distribution/ generators etc?
Del

Wow! Even ME students here have to take EE courses (and EE students have to take ME courses too). How did you avoid that?

Start with basic electricity books for youngsters to see the basic theory of circuits. Then look for ones that use "transformers," and learn what they do.

After you've mastered that part, learn what diodes (aka rectifiers) do, and apply that to the secondary of the transformer. You now have DC voltage/current, but it's not good DC voltage/current. You'll need a capacitor to stabilize the output.

Next, if you want something other than what you have, you'll need a voltage regulator. You may design and build your own, but it takes a lot of learning to do that, so I suggest buying any one of several on the market. Some data sheets for regulators may also show circuits which allow adjustment of the output voltage, so if you want that, you'll need all the components shown in the circuit.

I hope you can tell that I'm being facetious ( or smart-ass, whichever you prefer ).

If you want to learn how to build state-of-the-art DC power supplies, then you'll really need to start with EE undergraduate courses, and then take some graduate courses.

If you're talking about power plants, then the same applies, except you may need more graduate school.

I want a power supply to generate an Alternating signal, 1-4 KV, less than 1 AMP, high frequency. Want it to hopefully fit in a small shoebox. For the wise guys & comedians, I'm not powering shoes!

I do realize power supply design involves the study of electronics and electrical principles. My question is: Is there a definitive resource on power supply design given the parameters I mentioned above? If I can't do the final design myself, I want an understanding of what an electrical engineer is faced with if I have to sub this out.

When you say "alternating signal, 1 - 4 kV, high frequency," how high is the frequency?

The higher one gets, the more problems one runs into, and one of those is inductive reactance. Other problems can be more difficult.

You might check out some of Nikola Tesla's work in HV, at fairly high frequencies.

My answer to the question "how high a frequency" unfortunately is as high as possible. Ideally, I'd like 40-50 kHz. My on duty time only needs to be 10% of the cycle. Initially I'd like to vary the frequency to dial in the optimum for my process. This involves digital imaging and particle charging. It's new technology and at this point I don't know all the factors. I believe I can make this work at lower frequencies, but the optimum will quite possibly be beyond what is typical in power supply manufacturing today because of the frequency issues. My approach is to get the signal shape I want. Then increase the frequency to the edge of the envelope and back off to make it feasible for the marketplace. I'm assuming there will be issues that can be solved with added cost, but I'd guess I would hit a break point that would keep me from being able to market this device.

Many PWM servo amplifiers and VFD's (Variable Frequency Drives) are manufactured now which operate at 40 - 50 kHz and beyond, so the technology exists to handle the power you need. Not to mention that much higher frequencies are generated at high power for such things as digital TV, radar, etc.

4 kV is not so high that it can't be done directly switching at 50 kHz, but you may want to talk to an electrical engineer to devise a way to switch lower voltages to the primary of a transformer that is designed to handle the frequency you want. Ordinary power transformers which are designed to operate at 50 - 60 Hz don't 'like' frequencies as high as 40 kHz.

I can't tell you anything specific, 'cause I don't know anything specific about what you desire. I could propose something specific, but it may not fit your application, in which case, I would be wasting my time and yours.

Hey Cincy10,

There is a family of electronic ballasts (for fluorescent lighting) that provides a very similar sounding signal. The frequency is dependent on the load which could be some high wattage resistors. The maximum frequency occurs at no-load and decreases to between 30KHz and 50KHz depending on the brand (mfgr). They are fairly fault tolerant and may be worthwhile simply because of the cost. You would probably want to focus your search on an "instant start" type that is sufficient for up to three or four fluorescent lamps.

BTW, you mentioned particle charging which sounds like a DC charge rather than an AC phenomenon. For that you would simply use a voltage tripler or quadrupler and some sort of spark gap device to generate the high frequency burst at a given level. Perhaps you would describe the requirements a little more specifically. You can look me up in the "Show Members" section to the right, if you like.

Any good audio amplifier has a bandwidth of at least 50 KHz. A 200 W power audio amplifier has an output of some 55 Vpeak (110 Vpeak-to-peak). That is not enough for the voltage and power that you want as an output, but a high voltage, high power stage can be driven by this 200 W amplifier. There are discrete devices (transistors) that can withstand high voltages. It is not easy, the voltages are high, a lot of electro-shock possibilities but with good advice you can do it.

As I write this, I recall some op amps for high voltage made by AMP.

You might want to have a look to this:

http://www.edn.com/contents/images/101404di.pdf

Good luck, boy scout.