Boost 12Vdc to 100Vdc ?

Get a good electrical physics book and do some reading then get back to me.

TCM, if you're gonna say something like that it's best to explain rather than sending a user on a wild goose chase with no clue what they're looking for, this is one of those VERY confusing portions of electronis that ticks me off.. This is right up there on my current list of things that irritate me about electronics.

Currently number 1 is the "split phase" system used in the US for 240V lines (AC/electric stoves etc..) The two lines are 180 degrees out of phase with netrual so if you refrence either one to ground you'll get 120 volts but if you refrence them against each other you'll get 240 volts.
Now the raw words that make up polyphase are poly and phase. Phase is obviously well defined, but poly means 'more than one' So you'd think that American 240V lines are polyphase systems but they're not because someone decided that to be called polyphase there had to be an obvious progressing phase change, IE one phase had to be leading or lagging. In a split phase system this doesn't happen so an AC motor that required a polyphase supply could (even if it would start) would start in a random direction depending on the exact moment it was turned on.

2 would be this one, AC means actuating (changing) current. Even if a true AC wave is DC biased above 0 volts it still actuates. But because the engineers decided it was this way to be called AC the current actually has to change direction, meaning it has to have current flow the opposite direction during the lower polarity shift or it's just a pulsating DC waveform, regardless of how much it changes. That's just the way it is.

The third would be "Ohm's law" The equations that are used in non-linear circuits even though they have equivalents are not 'technically' ohms law because Ohm's law were actually based off experiments he did on very heavily control thermally and physically conductors. The equations that were derived from it aren't actually 'the law'

Fourth would be the direction of current flow. Current is almost always referenced to be flowing from the positive to negative terminals on a voltage source, however the electrons (the bits that actually provide the force) flow from the negative to positive terminals.


There are ALL technicalities, but important ones to learn otherwise you'll get confused fast.

VERY well said Sceadwian.

As wonderful as electronics/electrical engineering is, there is always slight ambiguity with terms and phrases. For the average or professional engineer, they can deal with them just fine, but for education, it can be a minefield, especially if some teachers teach it one way, and other teachers teach it the other.

I apologise for the super long post above. It was not my intention to patronise, or put off the OP, nor to start at the very basics ands lowly working up...thats liek saying 'yeah, you can build your converter, after 6 years of experience' - doesn't really help. I try to be practical with these problems. As ambitious as this project may be, picking a relatively complex task is a wonderful way to learn, it forces one to 'go look up' unknown area's as and when one comes to them. Its how I learned!

Mind you blueteeth, TCM stepped on all those landmines, that's why he knows.

electrookie you may find this tutorial of use in understanding how a simple boost converter works:
NerdKits - LCD Backlight - Electroluminescent Inverter

Did it make you think?
Did it piss you off just enough to want to go and read up on things until you found enough information to prove me wrong?
If you did do you now have even more conflicting answers and definitions as well?


Welcome to the land mines I have to deal with here every day!

Some times one sentence can teach a person more than writing a whole book! Today I used my sentence. Dont make me use it again.

P.S. checkmate.

Yes TCM sometimes a sentence can teach more than a book.
But "Go read the book" is not one of them.
Your go read a book reference wasn't in this particular case useful to anyone.

You can scream checkmate all you want, doesn't mean you've won =) Doesn't mean bluetooth saying 'well said' actually means I said anything well either. But no one here should be slamming down simple facts as a single sentence as a sign of superior knowledge. We are all here to learn. Blueteeth, Eletrorookie, me, you, everyone should remember that.

Don't matter to me. How you interpret what I say and how any one else interprets it is always different.

One of the best teachers I had used that physics book sentence on me and I spent days trying to find out how to prove him wrong. Turned out I was the one who did not understand what he was referring to and was wrong.
I guess maybe it just works better in person or with the right person. So does the checkmate reference.

There seems to be more misinterpretation of intentions between people than usual going on around here which concerns me being I am not necessarily trying to provoke arguments as of lately either.

tcmtech, did you provide a link or even a title to a book that might have been helpful?

I'm sorry, but in this particular case there is no misinterpretation. You replied with a vagary to a serious question. Are we going to be bad teachers and fault him for not knowing something we haven't actually even TRIED to teach him?

What did I miss here? There is no intention to be confrontational and you shouldn't feel imposed upon to answer in the forum if you don't want to, my pm box is always open =)

Well, bottom line is this- a charge pump is not possible to use here. It'll take too long to explain why if you don't know. It's not gonna work on any practical level.

Pulsing an inductor as a "boost" converter to make non-isolated 100v from 12v is only good for very low power levels. The high vout/vin ratio boost converters don't work really well with significant current, the transistor, inductor, and capacitor sizes become impractically huge.

A 20 amp/100v output requires exceptional performance. What standard car power inverters do is use transformers. Usually with a center winding at 12v and two poles alternately pulled down with NMOS. It's possible to use an h-bridge but this one seems to be preferred, since it saves on expensive/complicated high power/high freq transistor stages.

The output is FWB rectified to a DC rail somewhat over 120v (there is a dead time between positive and negative phases, and RMS needs to stay at 120v).

In fact, lemme just say it. Building a 20A output inverter is WAY beyond your apparent skill level. There's like a dozen very "hard parts" here.

Best idea is to get a power inverter off the shelf, FWB the output to make the high voltage DC rail again, and build a stepdown converter. Although, gotta tell you, a 20A high voltage stepdown converter is already beyond your skill level too (and this will cost a LOT in parts and development).

Might be better to feed the power inverter output into a big-ass variac, put it on a FWB rectifier and filter, and adjust the variac to get 100v. That's the cheapest and most likely to be successful option. It'll still cost you a bundle, a 2KW continuous inverter is not cheap, nor is a 2KVA variac, nor HV filter caps with 20A of ripple rating.

All this is beside the point if you don't have a DC source capable of ~200A. A car battery will only supply that for a short time. Even a running car would need a HUGE alternator to sustain such high current.

WTF do you actually need this for? 20A of 100V DC??

Oznog, there is no bottom line here, he's already been given everything he needs to learn, and he's likely chewing hard on what information he was given. Lets not try to beat the knowledge into him with more numbers at this point.

Unless it's a common thought that information and knowledge instantly transfer with every post in this forum... in which case, I will precede to hide and cower forever.

The OP hasn't logged on since the "real AC" bickering began.

A lot of food for thought

that is a lot of information and I feel overwhelmed with this project now. Nothing simple about it. I think I will go look into just buying something that comes close to what I want. I don't "need" any of this, just had an idea for a project and now that project is way down on the list. I do appreciate the information. I don't know that the definition or distinction between DC and AC is any clearer, but it seems that what I need to do is a lot more studying, which I will do.

I will get back to this forum on this subject when I am a bit better versed on the subject, or when I come across another question in some other project. I am finding electronics in general to be exciting as well as challenging and there is always more to learn. It is a kick to put something together and throw the power switch and observe the thing NOT blowing up, and actually working. The more I learn, the more I want to learn.

Happy holidays to you all...

electrookie said:

that is a lot of information and I feel overwhelmed with this project now. Nothing simple about it. I think I will go look into just buying something that comes close to what I want. I don't "need" any of this, just had an idea for a project and now that project is way down on the list. I do appreciate the information. I don't know that the definition or distinction between DC and AC is any clearer, but it seems that what I need to do is a lot more studying, which I will do.

Click to expand...

It wasn't my intention to 'put you off' bnut you are correct, it is an ambitious project (something I would love to do, but would be beyond my skills). Apologies for the awkward defintion of AC - aswith many things in life, this is more to do with 'wording' rather than a hardcore theory.

electrookie said:

I will get back to this forum on this subject when I am a bit better versed on the subject, or when I come across another question in some other project. I am finding electronics in general to be exciting as well as challenging and there is always more to learn. It is a kick to put something together and throw the power switch and observe the thing NOT blowing up, and actually working. The more I learn, the more I want to learn.

Happy holidays to you all...

Click to expand...

Exactly! It's why I have studdied it most of my life....such a huge subject, theres no chance of me getting bored easily or feeling I fully understand every principle. I too get the 'kick' of building something from scratch. Even if it does blow up....for geeks like me, that is just an excuse to 'tinker', probe, and find out why it isn't behaving as expected.

As always, this forum is a place of learning, sharing idea's, discussion etc.. I'm sure I can speak for every hear when I say that we are happy to field any questions that google doesn't appear to answer (or even if it does, themore definitons of a theroy you have, the better the understanding).

As far as I'm concerned there is no such thing as overly ambitious, just lack of space restrictions and time. If you have the space to put whatever it is you end up making and don't mind the long period of time it will take, dive right in, start with a smaller unit of the same type get that working, decide if it's worth your effort to do something on a large scale and go from there. You could start with a 100V 100ma supply, that's much more reasonable, and the same theory applies for a larger supply, the parts are just bigger and more expensive.

It's pretty much the same with any hobby, think big, start small.

If you are experimenting, I'll stay out of the conversation. If you have a solid goal, perhaps you should look into a GenSet. Inverters output a square wave at worst and a pseudo sine wave at best. My fiber optic fusion splicer cost more than my house, I'm not going to trust it to a pseudo anything. The bucket truck has a small 60 Hz alternator powered by a shunt wound DC motor. 12 volts DC at beau coup amps in and 120 volts, 10 amps, true sine wave out. The whole rig is about 5 inches diameter, foot and a half long and has 3/0 welding cable to the battery. But, the true sine wave output is what I need.

You could use an oscillator and voltage doublers cascaded until you get the voltage you need. Problem is, to get any current, the trunk and back seat will be full of capacitors if you need more than 10-15 mA.

Ever heard of a true sine wave inverter Artificer? They do exist for fussy loads =)

Your fusion splicer is a glorified precisions heater, you could probably feed it DC at near it's required voltage and get 100% perfect results.

Depends on whats his house is worth.

I filled my 1000 gallon propane tank and now thats worth more than my house!

I can think of only one theoretical way you can do it without a transformer. Charge a bank of 9 capacitors in parallel and discharge them in series. That will give you a little more voltage than you want, but it will probably load down anyway unless you have a good output filter. Given the currents you want, you will have to build each of the 9 capacitors out of a bank of capacitors. I saw GE do it once to generate a large current pulse. The bank they needed filled a small room and the charge cycle was minutes long. That is the problem with theoretical solutions; they are not always practical. Use an inverter and transformer; it will be both smaller and cheaper. There is a good chance that it will also be more efficient.

Sorry, accidental double post.