sine wave inverter

A

aamir shabbir

Member
hi guyz howz u all..............guyz i want to make a pure sine wave inverter using the chopper technique i know some basic things about this project:

1- i will use a avr microcontroller to generate the pwm reference signal of about 30 khz
2- i know that i have to use ir2110 for mosfet switching ill use full bridge topology

but the problem is i dont know how to make 312 DC out of 12V dc how can i boost that much ?? i said 312 Vdc as the peak of 220Vac is 312 Vdc....
another question is that what will be the connections for the chopper ???

please guide someone in detail and also provide the schematic if anyone can
 
There are hundreds of sinewave inverter projects in Google. Some of them work.
 
Hi,

This sounds like a transformer design. It's very hard to boost 12v up to 312v because that takes very low series resistance for all the parts including the inductor, and there is a rule that relates series resistance to min duty cycle where the boost reaches a maximum, so the best bet is probably to use a transformer to boost the voltage not a conventional inductor based boost circuit. You can either do this for the DC or for the AC which means only having to handle the full power once. The size of the transformer however will depend on how much power output you need.
It wont be easy to do either way though.
 
Last edited:
Using a transformer seems like a good way to go but you will need a somewhat uncommon transformer voltage of 220V to 8V such as this. You drive the 8V windings with a bridge powered from the 12V supply. This 24Vpp PWM drive signal (adjust PWM to give 8Vrms) from the bridge applied to the two 8V windings in parallel will give a 220VAC signal at 24VA maximum from the transformer 220V winding. If you need more power you would need a larger transformer.
 
Hi there Carl,

Yes that sounds good, but wow i am surprised at what they are charging for that transformer and the shipping too, wow that's high.
Every time i look at transformers these days it seems that the price went up again.
 
I think something may be changed here, the 30 KHz is not for the PWM of the 50 Hz generation, it is simply to generate that 312 VDC from the 12 V then you shape that DC at lower frequency with PWM too to form that sine wave. the transformer may be something like the one used in PC power supplies except in reverse. Note that the PC power for 220 v have transformer of 110 ac as input because the half bridge configuration used.
12 VDC will be ok for up to 1 kw but higher power will need 24V for reasonable battery current
 
There are two obvious ways to do this:

1. Use a switching inverter with a relatively small, high frequency transformer to convert the 12Vdc to 312Vdc and then use the 312Vdc to drive a high voltage PWM bridge circuit to generate the AC output. This has the problem of driving a very high voltage bridge circuit.

2. Use a 12V PWM bridge to drive mains transformer to convert the PWM 12VAC to 220VAC. This has the problem of required a large, expensive mains transformer, depending upon the required power.

So it's rather a tradeoff as to which is the better solution.
 

Hi,

Yes but generating DC of 312vdc at any significant power would be hard to do without a transformer, and if you are going to use a transformer anyway then might as well use it as the output transformer where you generate the AC signal at the primary.
If we use the transformer to generate DC, then we need a switcher just for that, then if we convert that to AC we need another switcher. That requires handling (switching) the power twice rather than once with the direct AC output approach.

For lower powers though it may be possible to use an inductor based boost circuit followed by a switching bridge but it does get a little tricky.

So we also need to know what the output power has to be.
 
Well, the switching to generate a DC will have a constant frequency and relatively stable pulse width, this makes the ferrite transformer design easier, if you use it for the output, the pulse width will change from near zero to near 100% to form the 50/60Hz which will make it difficult with different loads, also you will have to deal with chopping in one direction to form the +ve half cycle of 50/60 Hz then reverse for the -ve cycle, which may cause a saturation problem
also if you use inductive loads like the transformer of any electronic device or a motor etc, or neon lamp, that should be considered in the design of the output transformer too but if the output is an H MOSFET Bridge , it can handle that easily, just increase the MOSFET current capability.
Anyway that design is what I found in a lot of commercially available units
 
Hi again,

Either way represents a challenge. The direct drive AC output method does work however as that appears in many higher powered inverters. Some big output transformers though for 30KVA 3 phase systems.

The cheap commercial units are the ones that use DC to DC converters followed by a bridge. If you have seen models that go from 12v up to 220vac output then that means there is hope for this project after all using a DC to DC front end. We could take a look at the main requirements i guess, but i am waiting for a spec on the output power before i do anything else because that really tells us what work we have to do.
 
Hi,

Well 500 watts is a significant amount of power although it's just getting up there not too high.

For a 500 watt DC to DC converter i dont think a boost circuit would be recommended, although it may be possible.
A transformer based DC to DC converter would be better, and if you want to try a direct DC to AC via transformer that's up to you too.

The transformer based DC to DC would be a switching bridge driving the transformer and the output would be rectified and filtered creating a higher voltage DC. That would then be switched by another switching bridge to get the AC output.

A direct DC to AC would be driving the transformer the same way but with a different switching pattern that contains the line frequency.

The difference is that with the DC to DC a smaller transformer can be used because you can switch at a higher frequency without any low frequency component. The DC to AC will require a transformer that can handle 500 watts at the line frequency (probably 50Hz in your case). So that's the difference.

From this it sounds like the DC to DC would be better. The DC output voltage could then be controlled to regulate the AC output of the second bridge.

What have you done or built in the past, anything like this or any other electronic project? Just so you know this isnt really a beginner type project.
 
I came upon this thread.
Pls sir could you explain better what you mean by direct DC to AC via transformer method.i know half bridge are used so pardon me for asking this
 
Hello,

By direct DC to AC i mean you create a switching bridge and that switches the primary input voltage (like 12vdc) into pulses, but the pulses are of varying width so that they create a sine wave of frequency 50 or 60Hz. That applied to the primary of the transformer then gets stepped up and the secondary has the line voltage like 120vac or 220vac or whatever. So it means the power only gets switched one time only and that creates the AC directly from the DC. The switching pattern looks like a pulse width modulator with a sine wave input. So the pulses are wider near the sine peak and more narrow near the zero crossing for example.
 
how efficient will is this technique if at all employed in terms of percentage and overall performance?
 
Hello again,

It can be up near 90 percent or slightly better, but it really depends on what components are used and their series resistance and how well the transistors switch. But really it is like any other converter...the better the components the better the efficiency.

In contrast it is harder to make a DC to DC followed by a DC to AC as efficient, but with todays MOSFET's it may be possible to get to a reasonable efficiency that way too although i dont see 90 percent happening there.
 
You must log in or register to reply here.

Similar threads

Output transformer of homebrew inverter.
Replies
6
Views
2K
shokjok
S
D
FET or Electronic Relay for Wave Runner
Replies
12
Views
4K
ZipZapOuch
P
Square Wave Pulse Correction
Replies
9
Views
5K
ZipZapOuch
  • Solved
Square wave generation for automotive dashboard
Replies
10
Views
4K
Menticol
B
  • Question Question
Sine to square wave conversion
2
Replies
27
Views
15K
bob4bvm
B
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…