inverters

[bob g]

has anyone here either built one or contempated doing so?

the reason i ask is basically from an economic vs comprimise viewpoint

where for every 100 dollars of inverter retail price there is probably less than 20 bucks in parts involved in its manufacture

whats got me thinking is for instance an inverter that retails for 2500bucks and has all sorts of bells and whistles, fancy artdeco cases or other stuff i don't need or want.

i am thinking what can one build for the 2500bucks spent on parts?

i don't need pure sine wave,, but would like a better stepped waveform than is typical of the cheaper modified square wave units(aka modified sine wave). i don't need a fancy 4 line readout, puter link, comm link, 3 step battery charger or a fancy case,,,

i don't need to make a profit, or pay for sales, advertising or all the other stuff that a company has to pay for.

what i would like is a fairly decent waveform from a 24vdc front end, and maybe 5-7 kwatt output.

it seems like most of the offerings in this price range include all the other stuff i don't need, and they also have alot of overhead that has to be paid for as well..

so i am thinking it is likely that a 2500dollar inverter retail has probably 3 or 400bucks in parts in it...

what can one do if he was to spend 2500 on parts?

we can use a microcontroller to produce the timeing and step pulse to drive a bank of high power mosfets,,  so makeing a fairly decent waveform should be doable?

if you strip all the charging, com link, led screen readouts and the other stuff not directly related to makeing AC power,, there really isn't much left in a typical inverter.

any thoughts?

bob g

Moved to a more appropriate section.

[bob golding]

hi bob,

i have tried, but after talking to some very clever people who do this sort of thing for a living i came to the conclusion that it was beyond my capabilities given the test gear i have available. if i was going to attempt this i woud start with something like a UPS and try and redesign it. that way all the complicated stuff has been done for you and you just have to tap in to the right bit. dont ask me where the right bit is, but that is the way i would go about it. what you are paying for is the R and D not the cost of the components. they are the easy part,its all those hours of bench testing and expensive test gear you are paying for, plus the wages of some very talented people. of course if you did design your own you coud make it bullet proof by using more rugged components, but there s trade off. for instance higher voltage fets come with higher on current so the effiency drops. there was a bullet proof design on the "last page of the internet" that was a while ago dont know if it is still there. good luck.

cheers

bob golding

[finnsawyer]

Perhaps you should investigate Buck Converters.  You can Google same to find out how they work, but basically you channel the output dc current into an inductor for half a clock cycle and then during the second half as the inductor current collapses it is channeled to the battery.  Another nice thing made possible by the modern diode.  There has been talk about these things here, but I haven't seen any schematics (here).

[boB]

If you have a lot more time than money the building your own inverter

can be very educational and rewarding.  Do you have the necessary lab

equipment like, scope and a decent meter and soldering iron ??

Another thing you could do is to procure a 12V car amplifier and step

up transformer and then build a 60 Hz sine wave oscillator to drive

the amp with.  Then you have a sine wave inverter. After that, you could add

feedback to adjust the sinewave amplitude to get voltage regulation.

Building your own inverter WILL take time and some money and you

will most likely let the smoke out a few times. Certainly a modified square

wave inverter would be easier. Either way you have to make sure it will sink

as well as source to/from the AC load, or "clamp" to the zero crossing

for the modified square wave version.  I don't know WHY anybody would want to modify

a perfectly good sine-wave !

boB

[bob g]

boB:

yes i have a scope, meters and a good iron

and i would agree,, "why would anyone modify a perfectly good sinewave"?

lol

i also have a collection of high power parts, and also

have an industrial exide ups that is 15kwatt, 3 phase that i could canabilize for parts as well,, it has a 240volt dc front end which i have no intention of trying to feed. it has some hercan big parts in it that ought to be useful,, and maybe it could be modified for my intended use,, i dunno.

what i am wondering is

there are numerous schematics for square wave units, that seem to do the job they are intended for,, and they have internal triggering to get them somewhere close to 60 hz.  usually they use sort of a push pull arrangement to feed a transformer

what if one was to build a multistep push pull and instead of triggering the steps internally with descrete components one used a micro controller with various output pins to trigger each step.

anyone ever tried it,, or have a schematic for a microcontroller triggered inverter?

bob g

[vawtman]

Inverters are like drugs,but with no generic options.

 Go get em guys.

[scottsAI]

Bob g,

Older inverters used a milti tapped transformer to do what your asking for. (early switech?)

Problem the transformers require an activation energy, only a few percent and an efficiency hit.

My APC 3kw, it's sin out, uses a transformer for AVR and inverting. No load it draws 47watts. The transformer gets warm, room at 73'F, transformer is 115'F. Quick idea is most of the 47 watts is in the transformer.

Your piece price for parts is 4x higher than a huge OEM. Even with their markups you end up paying about the same for the parts as a complete working unit. Engineering is free!-) OK, not always.

I would like to remove the transformer from the circuit unless its needed, would save most of the 47 w.

To make the unit more efficient use several transformers, 100w, 500w, 1500w, 3kw...

Use one or more transformers as the conditions require.

New units do not use a 60hz transformer. High frequency simulated sin wave. Efficiency is much higher. Very complex to design.

I bought four 3kw UPS to back up my house. I only installed one unit to keep the batteries and back up my 15 SETI computers. (Off line for the summer due to AC cost) Not liking the 200watts running current the UPS uses. Bought several broken UPS: 5kw, 2kw, 1kw to modify to switch the transformer in/out as needed. Not got that far. Don't expect I ever will.

I plan to build a zero energy house in the next year. The inverter will be one of the newer grid tied inverters and pay the $2000.

Have fun,

Scott.

[wdyasq]

http://www.energychallenge.org/2001Reports/TAMU.pdf

The editors hate folks just pasting a link so here is some text.

Ron

[healerenergy]

I haven't bought the plans yet for this inverter but it won't be long now.  I thought the people in this thread might like it.

http://fuellesspower.com/Paypal/invert2.htm

George

[bob g]

the only problem i have with the ad,, is the "free energy" claim... that i have an issue with.. but..

if you buy the info,,let me know what you think

bob g

[healerenergy]

After I do get the info I will post a report on it.

George