Question about my pwm sine power inverter schematic

[gaetan8888]

Hello

I never succed with PIC programing, so I've assembled a schematic of a simple version pwm sine inverter.

Anyones can look at my schematic and tell me if it would work ok ?

Thank you

Bye

Gaetan

 

[gaetan8888]

Hello

I've put a 1k resistor (from gate to ground), to remove gate charge of power mosfet´s.

Here's my new schematic.

Is my new schematic are ok ?

Thank

Bye

Gaetan

 

[ronsimpson]

The SG2524 was never intended to drive MOSFETs. It can but there are other parts that do drive MOSFETs.
In your first post the PWM will turn on the MOSFETs but there is nothing to turn them off.
In post 2 ....1k turn off resistor will work at low frequencies but not well at high frequencies.

Your PWM will oscillate at 4khz. .033uf 10k I never drop down to a audio frequency. People can hear it. I usually go for 100khz or more.

Normally a error amplifier compares a reference and the output of the supply. In this case it looks like you are comparing the reference and the sign wave. I need to look at the Rs and C around the error amp to see what it is doing at 60hz. I don't think this is going to work.

The SG2524 is not the part you want. Assuming the error amp and sign wave is working right; The duty cycle will go from 0% to 100% and back in a sign wave. Driving all MOSFETs will not work! You want to take the top MOSFETs through 1/2 the sign wave with out the bottom FETs. Then reverse and drive the bottom FETs with out the top FETs being on.

 

[gaetan8888]

Hello

I have few SG3524 but no SG3525, that's why I use a SG3524 in my schematic.

Btw, there is a typo error in my schematic, the Ic should be SG3524.

Thank

Bye

Gaetan

 

[Mr RB]

I found driving accurate sine PWM into a transformer did not give a good sine output fromt he transformer. I needed to change the sine table quite radically to get the transformer output to replicate a good sine, and this was true for various load conditions.

I worry that you seem to be going to a lot of work to use a proper sine to drive the PWM but the end result will likely be a very poor sinewave out of the transformer. If you used a PIC it helps as you can modify the sine table, but you said no PICs.

Maybe you could make an analogue wave generator that is quite ajustable and then tune it to give a good sine from the transformer output?

 

[gaetan8888]

Hello

I agree that I won't get a very good sine but my goal are to get a better signal than a modified sine(a kind of modified square wave), so even a poor sine will be better than a modified sine or modified square wave, I can ad a filter at the transformer output.

Maby adding a negative feedback from the transformer output to the SG3524 input will do a better sine, wich side of the transformer output I use for a nfb ?

I have too much difficulties with Pic programing, so I will stick to analog solutions.

I can't find any SG3525 here around, so I've redone my SG3524 schematic for a better turning off of the mosfets.

Is it better ?

Thank you

Bye

Gaetan

 

[gaetan8888]

Hello

I think it's better to do biasing with a 10k resistor
at the base of those bjt transistors.

Here's the new circuit.

Thank you

Bye

Gaetan

 

[ronsimpson]

You don't think you know the logic of the SG3524/25. How it works.

What is the waveform coming out of your transformer? (it is a mix of 4khz and 60hz) Please draw that.

Then I will show what that looks line on the primary of the transformer, what the FETs have to do to make that.

 

[ronsimpson]

The green is your sign wave. (two signs ) 60hz
The red is the voltage on the transformer secondary. The width of the red is the duty cycle. 4khz modulated at 60hz.
Normally the 4khz is filtered out by a filter we have not talked about yet.
The output after filtering is nothing.

 

[summitville]

gaetan8888,
I think, I see what you are trying to do.
If the "input" signal, at IN-, is 0V then the ouput is at 1/2 Max Volts because IN+ is biased at 1/2 Vref.
Then you apply the AC signal to modulate the output from Max Volts to 0 Volts at 60hz.

The problem is the output of the Transformer Secondary is still oscillating at 4Khz.
If you rectify the output to DC then YES the DC output will vary from 0V to Vmax at 60Hz.
Almost 60Hz AC but not exactly.
AC oscillates between +Vmax and -Vmax at 60Hz

Options ...
1) Slow down the existing circuit to 60 hz
2) Use the existing circuit to create steady 165V DC which then feeds an H-Bridge to create 60Hz AC.

 

[ronsimpson]

Options ...
1) Slow down the existing circuit to 60 hz

This is square wave output.

2) Use the existing circuit to create steady 165V DC which then feeds an H-Bridge to create 60Hz AC.

Yes! If you use the existing...remove the 60hz osc and use feedback.

 

[Mr RB]

4kHz seems way too low? Why choose 4kHz for the PWM frequency?

 

[ronsimpson]

4kHz seems way too low? Why choose 4kHz for the PWM frequency?

This was mentioned in post #3.
There are so many things wrong. He needs to find a working inverter and see how it works.

 

[Nigel Goodwin]

4kHz seems way too low? Why choose 4kHz for the PWM frequency?

It's 67 times the target frequency, which seems plenty?.

Although choosing higher would probably be more normal.

He seems rather lacking on output filtering as well, so not removing the 4KHz carrier - which a higher carrier would simplify.

I'me presuming he's using a high frequency transformer?, and not just a 60Hz one?.

 

[ronsimpson]

The 4khz carrier will come through, TV sets, radios, electric drill motor, etc. To hear your appliance sing at 4khz is not pretty. Some clocks will go nuts.

also:

A 100khz transformer is smaller than a 4khz transformer. (and filter)

He started off with a 60hz square wave design and added 4khz PWM to it. Because neither if us know how it works, my guess is that he needs to stay with a large 60hz transformer.

 

[summitville]

This is square wave output.

Yes, that is a Square Wave. But that is what a PWM creates. The wave can be "shaped" with an Iron Transformer, Inductors and Capacitors into something resembling an AC Sine wave. Sort of like turning a lump of coal into a diamond. The thread title does not say "PURE" sine wave. I would not go this route but it is an option with the circuit presented.

Yes! If you use the existing...remove the 60hz osc and use feedback.

It may be possible to use the 60HZ to modulate the intermediate 165 DC Voltage. Then use an H-Bridge to create the 60 Hz AC. It is an option. One issue with modulating the intermediate DC votage is it will not drain to 0 Volts when there is no load. Maybe an issue maybe not.


If you go with the Regulated Intermediate 165 volts DC then remove the AC input signal, add (-) feedback, increase the frequency and do whatever else is needed to get the DC-DC converter functional.

 

[summitville]

gaetan888,

Google Mosfet H-Bridge and IR2110 driver chip then
"OR" the outputs from the SG3524 into a single signal
which drives the proper IR2110 pins
via some glue logic where ...

On the Positive Half of your Sine Wave the SG3524 OR'd ouput drives ONLY
Leg #1 High Mosfet and Leg #2 Low Mosfet simultaneously.

On the Negative Half of your Sine Wave the SG3524 OR'd output drives ONLY
Leg #1 Low Mosfet and Leg #2 High Mosfet simultaneously.

At / near the Zero Cross of your AC Signal no Mosfets in either Leg of the H-Bridge are ever on
and the glue logic switches the SG3524 OR'd output from one Mosfet pair to the other Mosfet pair.

Increase the frequency to 40KHz?
Add a little ouput filtering.

Voila - a DC PWM to "Pure" AC Sine Wave Inverter!

You still need 170V DC to power the H-Bridge.

OK ?

 

[ronsimpson]

Yes, that is a Square Wave. But that is what a PWM creates. The wave can be "shaped" with an Iron Transformer, Inductors and Capacitors into something resembling an AC Sine wave. Sort of like turning a lump of coal into a diamond. The thread title does not say "PURE" sine wave. I would not go this route but it is an option with the circuit presented.

The FETs and transformer will make 4khz square wave and will never have 60hz. and Using inductors and capacitors (not in the schematic) will require huge inductors and capacitors. Low pass filter with a roll off at 70hz.

It may be possible to use the 60HZ to modulate the intermediate 165 DC Voltage. Then use an H-Bridge to create the 60 Hz AC. It is an option. One issue with modulating the intermediate DC votage is it will not drain to 0 Volts when there is no load. Maybe an issue maybe not.

I agree. I have used a H-bridge that is PWM to create 60hz sign wave, from the 200 volt supply. Now the filtering will be to remove the 100khz noise. (small L and small C)

If you go with the Regulated Intermediate 165 volts DC then remove the AC input signal, add (-) feedback, increase the frequency and do whatever else is needed to get the DC-DC converter functional.

See H-bridge above.

 

[summitville]

The FETs and transformer will make 4khz square wave and will never have 60hz. and Using inductors and capacitors (not in the schematic) will require huge inductors and capacitors. Low pass filter with a roll off at 70hz.

See H-bridge above.

You forget, my Option #1 solution did state to reduce the PWM frequency to 60Hz. Therefore, a 60 Hz wave will be created. As the PWM frequency gets lower the transformer will be larger and the inductor and capacitors in the filter section (not shown) are larger, too. All are just typical trade-offs for different PWM designs.

 

[ronsimpson]

If the PWM is running 60hz then why not use a LM555. I don't see the need for a PWM.