120Vdc to 120vAC

"lerameur" schreef in bericht news: snipped-for-privacy@f14g2000yqm.googlegroups.com...

Some questions. Do you need a common ground for the AC and the DC? What type of load will the AC get? Do you need a real sinewave or will modified sinewave be good enough?

petrus bitbyter

That depends on lots of things

120 DC to 120 square wave is doable and efficient. "H" bridge with a multivibrator driving it. Some PWM "modified" "sine" wave might also be possible if you load could tolerate it. - downside is you really need about 170-180 in for a direct conversion without transformers if you need a modified sine wave.

How good does the output need to be with respect to harmonic content and what frequency were you going for?

Perfect sine waves are inefficient to generate from DC - A motor-generator might be called for 5KW - out of the range of most audio amps.

You'll probably have to design it yourself. There isn't much call for

120 DC to AC. Some research into electric car technology might turn up something - they use high voltage batteries and drive brushless motors (usually three or multi phase motors but the ideas could work for single phase if you need that).

o

The Idea here is to power a house. I have 11 batteries being charge by solar and wind power. I thought it would be simpler just to convert

120VDC to 120VAC then 12v to 120cAC. A 5Kw inverter is expensive, but the way I want to do it, I am saving myself the boost of voltage.

Ken

lso

r

Then you probably want something closer to a true sine wave. It's not particularly simpler and it probably will not end up being any cheaper. Buy the inverter. Check other voltages like 24v. You night find something there too. If you could get your hands on a used computer room UPS some of those had big ac inverters in them.

to

a

also

for

n

ght

d text -

It also occurrs to me to point out that a pretty large load. A dozen or so auto style wet cell batteries are going to do that for maybe a few hours at best I'd guess. If you really need 5kw then you need more batteries. And if this is a regular house keep in mind that standard line service is about 40kw. You are not likely to ever need that but do not plan on running any big loads like central ac off batteries. If you're going self reliant power you need to build the whole house to be far more economical about power usage.

to

d

r to

th a

ht also

if

t

l for

urn

k e t t

night

ted text -

So if I aim for a 40kW version, then I will need 8 5KW inverter, this is why I would like to build the circuit myself. I am good with electricity but less in electronics, I know I have to put safety first, but I think this is doable.

k

C to

uld

tor to

with a

ight also

ly

s if

ent

t

all for

turn

s

ork

rge

ert

ut

not

u night

uoted text -

You would need a room full of batteries and a 40kw inverter is going to be about the size of a clothes washing machine. I really don't think you could build it for less than you could buy it.

VDC to

hould

lator to

e with a

might also

ally

ers if

ntent

ost

call for

ht turn

ess

work

harge

nvert

but

s not

You night

quoted text -

en

ed

e

t -

A 120Vdc bus is a standard for "industrial" UPS's, called a "low bus" design. There are companies that make them, such as SCI and Gutor.

To get a clean sinewave will require a lot of complex control in the inverter algorithm AND require an expensive set of filter inductors and capacitors. This is by no means a trivial task. Shooting from the hip, I would guess than buying a unit for this application would cost at least $20K USD. On top of that, you would need some serious battery capability. A string of 'car' batteries will not give you much up time.

Only if you're planning on using square wave output. 120V AC means 120V RMS, which requires 170V peak for a sine wave or 147V peak for a "modified sine" wave.

Parallel your batteries & run 4 1500w 12v inverters off them ($100 ea):

The Idea here is to power a house. I have 11 batteries being charge by solar and wind power.

Ken

Can your charger keep up with your demand?

Tom

120VAC is an RMS reading, it's about 170-180V peak-to-peak. If you got yourself five more batteries you could do this without transformers.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

What are you most comfortable with? You could do this either with a low-level sine wave oscillator and an analog amplifier, or you could do it with a processor that has on-board PWM generation. I'd go with the microprocessor because that's what I'm comfortable with, but I know some excellent designers who would blanch at the thought, and would have a wien-bridge oscillator or some such over at the top left of their schematic.

A 5kW DC-to-AC converter is getting pretty close to rocket science (in the sense that you're more likely to find it in a rocket than in a dishwasher, at least). If you have to ask the above question, then you're going to go through a _lot_ of parts before you get something that works right.

I kinda like the massively multiple Harbor Freight converters from +12V, unless you're determined to come out of this a better circuit designer than you are going in.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

Maybe you missed this part from one of the OP's earlier replies:

The thermal issues of an analogue design would be bad enough at 5kW. At

40kW, he's going to need 5kW just to power the cooling fans.

And:

If the objective is energy efficiency, a switching design is the only way to go. But at that kind of power level, this isn't a beginner's project. Unless you have a good understanding of EMI issues, 40kW of DIY SMPS is going to screw up TV reception a hundred miles away.

There are some serious issues here. You'll be drawing in the neighborhood of 50 amps from each of the batteries to get ~5kw. That means you want some heavy conductors to put the batteries in series - think automobile battery cables for size. They will need to be switched out of series if they are charged individually from your sola5r and wind sources. If they are switched out of circuit for charging and back into series when powering the load, you'll need switching circuits that can carry that relatively high current with low loss. None of that is a "show stopper" but it *must* be addressed in the design. And that needs to be addressed _before_ worrying about how to build the inverter - i.e. there's no point in having the inverter you want if you cannot provide it with the needed input power.

Safety is a huge issue. If you do not know how to accomplish what you want, then you *cannot* know how to address the safety issues.

Cost. The inverter/batteries/solar&wind charger setup will cost you WAY more than a 5 kw gasoline powered generator. You need to figure the run costs to see if your plan has merit. If it is just for backup when the grid is down, the gas generator is the way to go. If it is a house that is far from the grid and far from a supply of gas, solar/wind may be viable (or possibly the only real cost effective solution). You need to work out a detailed energy budget and compare. Space. You need to consider how much space is available to dedicate to whatever equipment you use to provide the 120VAC, and you need to consider the reguirements of that equipment - airflow, ambient temperature, service clearance etc.

Location. Part of the engineering is determining whether there is enough solar/wind energy available at your location to make the effort worthwhile.

As a general rule, solar/wind is not a viable alternative to grid connection to power a house.

You indicated that you already have batteries being charged by solar/wind. That *might* give you a leg up on some of the issues mentioned above - but it might not.

Ed

I think you're confusing "analog" with "linear". I was, indeed, suggesting that a _switcher_ built out of analog circuitry is the way to go if you break out in hives at the thought of a microprocessor squatting in the middle of your board.

Well, that was my point. He's clearly a beginner now; if he manages to finish it he won't be a beginner any more, but he needs to think about what he wants to spend his time on.

--
http://www.wescottdesign.com

text -

Maybe some experimenting is in order to determine if you need a sine wave.

I'm running some 120 V compact fluorescent lamps off a little kludged together 12VDC/120VAC pair of cross connected transistors driving a 12 radio shack transformer. It only puts out 95 volts AC but the lamps work and a switching power supply I use for 12V projects works on the

95 V square wave input.

Providing the switching supply has fast enough input rectifiers, I can't see a problem with the idea.

With 120 VDC you might just get by with DC into a switcher - but each load would have to be looked at separately. A switcher usually has a FWB input to charge some caps to 150 VDC then circuitry and a transformer to lower and isolate the output. More complex stuff that incorporates power factor correction may just smoke . . .

You could buy a commercial unit for the sensitive stuff and build your own device for lighting circuits and none sensitive loads.

DC to

ould

ator to

with a

might also

lly

rs if

tent

st

call for

t turn

ss

work

arge

vert

but

not

ou night

quoted text -

n d

I might put this project on hold or possibly go a different route. thanks for all the feedback.

K