Pure Sine Wave Power Inverter

Status
Not open for further replies.
Mosfets without a 100 ohm resistor results in a simulation error. (Wouldn't they burn out due to excessive current without the resistor ?) How to determine how much power will the mosfets deliver ?

Your Mosfets without the 100 ohm resistors produce a peak-to-peak output of about 26V then the 20:1 transformer produces an RMS output voltage of only 92V instead of 120V and the voltage will drop as the battery voltage runs down.
Click to expand...

In India, we have 230V mains. If the mosfets are producing 26V, then how come the output be 92V ?
 

DasPreetam said:
Mosfets without a 100 ohm resistor results in a simulation error. (Wouldn't they burn out due to excessive current without the resistor ?) How to determine how much power will the mosfets deliver ?
Click to expand...
Maybe your simulation software did not learn that a transformer has inductive reluctance that is a high impedance to the AC signal. The load on the output of a transformer causes current to flow in the Mosfets.

In India, we have 230V mains. If the mosfets are producing 26V, then how come the output be 92V ?
Click to expand...
My calculations before were wrong.
The 12V battery is about 13V when charged and is loaded. Each Mosfet switches about 13V into half the primary of the transformer. The transformer is shown to have a turns or voltage ratio of 20:1 so its secondary produces 13V x 20= 260V peak which is 184V RMS. To produce 230V RMS then the transformer must produce 230V x 1.414= 325V peak. Then the turns or voltage ratio of the transformer must be 325/13= 25:1.

Transformers are not spec'd with their turns ratio. Instead they show a voltage ratio. For 230V RMS then the low voltage winding must be 9.2V RMS center-tapped, not 12V and not 24V.
 
Transformers are not spec'd with their turns ratio. Instead they show a voltage ratio. For 230V RMS then the low voltage winding must be 9.2V RMS center-tapped, not 12V and not 24V.
Click to expand...

9.2V RMS center tapped. Means 4.6V RMS on each halves ? That gives about 4.6*1.414=6.5 Vpeak. But the Mosfets are switching 13Vpeak. Can you please explain a little more ?

Also, having a hard time with the LC filter. All I am getting is a DC output at 230V
 

I have never needed an inverter. My electricity is very reliable and I never go away from civilization.
Use a 18.4V center-tapped to 230V transformer.

An LC lowpass filter with an PWM input produces low frequencies, not DC.
 

Okay solved the filter problem. Now I'm getting a nice sine wave at the output. However, the output voltage has dropped tremendously (It is only few Volts). I'm using 50 ohms resistor as load. Why is this happening ?
 

230V/50 ohms is a current of 4.6A RMS or 6.5A peak.
Before you had resistors in series with the Mosfets that reduced the current a lot. Are the resistors removed?
 
Sorry, I forgot to remove the resistors ! Now it's working fine... One question. Commercial Inverters are rated in watts (eg: 200 watts, Pure SIne Wave). How to determine how much power can this inverter deliver ?
 

Your Mosfets are very good and are turning on properly. Then the limit of the power output is the quality of your heatsinks.
Oh, I forgot. Your simple circuit is missing circuitry to provide voltage regulation, overload limiting or shutdown and shutdown when the battery is discharged too low.
 
for sinusoidal waveforms you need to look into Sine-Triangle Modulation theory. Google it and see what you can come up with. also, most application will benefit from some sort of feedback control so that a system parameter changes (input voltage, output load etc.) your controller will compensate for it and keep a continuous desired output.
 

@Audioguru : That means if I use those large, fancy heatsinks with fans... Can I get output of 1000 Watts ? But if a use a transformer of 600 VA, Will it restrict the output to 600 Watt ?

@Audioguru and Mr. Cool : Yes, I looked for some schematics in the internet and found they all had some sort of feedback system from the output. However, I'm completely in the dark in this matter.

also, most application will benefit from some sort of feedback control so that a system parameter changes (input voltage, output load etc.) your controller will compensate for it and keep a continuous desired output.
Click to expand...

This controller part is a bit intimidating. How will the controller work ?
 

DasPreetam said:
@Audioguru : That means if I use those large, fancy heatsinks with fans... Can I get output of 1000 Watts ? But if a use a transformer of 600 VA, Will it restrict the output to 600 Watt ?
Click to expand...
Simply look on the datasheet of your mosfets to see their maximum allowed current and heat.
If you overload a transformer then it might smoke or catch on fire.

@Audioguru and Mr. Cool : Yes, I looked for some schematics in the internet and found they all had some sort of feedback system from the output. However, I'm completely in the dark in this matter.
Click to expand...
When the load current is low then the circuit has almost no losses so the output voltage will be too high.
When the load current is high then the losses are also high so the output voltage will be too low.
You need an error amplifier with negative feedback that keeps the output voltage at one level.
 

Simply look on the datasheet of your mosfets to see their maximum allowed current and heat.
Click to expand...

The datasheet excerpts (from https://www.vishay.com/docs/91291/91291.pdf)



Here, there are two types of currents, Pulsed and Continuous. Since I am using a 10KHz Carrier wave, should I consider the PULSED drain current ? It is given 200A. That means maximum power possible (before it burns out) is 12*200 = 2400 W ?
 

The datasheet note for pulsed current lists figure 11 that shows the thermal heating for momentary pulses that have durations as shown.
The duration of your pulses vary with the sinewave voltage. If your heatsink is very good then you should limit the maximum current to 50A to avoid destroying the Mosfets with thermal fatigue caused by heating then cooling over and over. The heatsink will keep the temperature of the Mosfet case fairly constant but the chip inside can heat much higher than the case temperature.

If the peak current is 50A then the RMS current is 35.4A then the maximum power is 12V x 35.4A= 424W RMS.
 
Okay so I need a better MOSFET with more current handling capacity in order to produce about 1000-2000 Watts. Will look into the website of International Rectifier.

As goes for the load regulation part... Where should I connect the feedback output ?
 

I guess you have never seen the schematic of a sinewave inverter that works. They use Mosfets in parallel for higher power.
They also use an error amplifier that has negative feedback from the output (maybe opto-isolated) to keep the output voltage swing at one level.
 

I've seen schematic of sine wave inverters, but never understood them because they vary greatly in the type and number of components.

I can connect the mosfets in parallel, but will the PWM signal from the comparator be able to drive all those gates ?

Also, what is the power amplification factor ? Means if I connect two mosfets in parallel, will that double the power ?
 

Obviously, if you double the number of Mosfets by connecting them in parallel then you double the output current and double the output power.
I can't remember which comparator you are using, most comparators have a small amount of output low current and use an external resistor to make the output high. You can add complementary (NPN and PNP) emitter-follower transistors to boost the comparator output current.
 
Actually The mosfet gates are driven by the CMOS NOT gate.
 

DasPreetam said:
Actually The mosfet gates are driven by the CMOS NOT gate.
Click to expand...
Oh no. You are using an obsolete CD4009 that was replaced by the CD4049 many years ago. These are ordinary old Cmos that have output current much too low for driving Mosfets. At least you should be using complementary transistor emitter-followers.
 

Attachments

  • Mosfet driver circuit.PNG
    9.3 KB · Views: 105

So can CD 4049 drive 3 mosfets ? And if I use transistors ? What should be their parameters ?
 

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…