Prevention of core saturation in push-pull topology

[Tahmid]

Hi all,

I want to use voltage mode PWM controller like SG3525 for making push-pull dc-dc converter from 12vdc to 350vdc for making sine wave inverter.

I want to be assured of different ways of preventing core saturation (due to flux imbalance) in push-pull circuit by using EI type core and voltage mode PWM controller (SG3525) at 50kHz. This dc voltage will be used in full bridge converter to convert this 350vdc to 220v sine wave. For conversion of sine wave, from 350vdc to 220vac I will use Bubba Oscillator.

But I am very much concerned regarding the staircase voltage-second core saturation phenomenon in push-pull topology.
I request suggestions to be put forward for preventing this saturation problem.

I donot want to use current mode pwm controller.

Thanks.

[Bootstrapper]

If the current-mode control is not allowed, there are basically no ways to prevent the possible (evident) saturation. If you are designing an inverter of mediocre power level, you must use current-mode control!

If you have a voltage-mode controller with 2 error amplifiers, you can use one error amp as a current sense input. That means you need a current sense resistor, whose voltage you insert into error amplifier (some scaling and filtering may be required).

You cannot design a reliable push-pull converter with pure voltage-mode control. As an example, if your design delivers 350Vdc @ 1 amps, you need more than 30 amps at 12V input!!

[Eva]

Unregulated push-pull converters are self-balancing if enough dead time and a gentle Vds rise ramp at turn-off is provided.

To get a very reliable regulated converter you may try peak current mode.

Another approach that seems to work fine is to sense the output voltage indirectly at the primary side, through the primary winding not being currently used, thus sensing any flux imbalance too. Steg audio amplifiers (an Italian brand) do it that way.

[TechGuy]

You can add some switching deadtime to will help reduce the chance of saturation but it doesn't eliminate it.

Personally I would avoid using push-pull and go with a full bridge. with a low input voltage of 12V finding Mosfets with low Rdson is easy. In addition it makes a lot of sense to use a PWM drive on the step-up transformer to form a sine wave output, unless your planning to support the output from multiple input sources ie (12V Battery + 220 VAC mains). Search for Microcontroller\DSP white papers for PWM sine output drives (typically used for AC variable speed AC Motor control). Or look for AC variable speed motor controllers:
http://www.datasheetcatalog.org/data...on/mXruuwq.pdf

You should also consider using a higher input voltage than 12 VDC. At the mininum I would go with 27.6 (two 12V (13.8 V) batteries in series) or higher. Double the input voltage and you cut the input current in half. P = I^2R losses at low input voltages is a killer.

You also need to consider some current managment. Lets suppose you have a short (do to a a bad load, or a fault in your circuit). This could result in a very bad problem as large battery with a short can cause a fire. Consider the size of the fuse you will need if your input current can be 50 to 100 Amps under normal operating conditions. You need a current sensor so that you can shutoff the power when something goes wrong.

[Tahmid]

Hi,

a gentle Vds rise ramp at turn-off is provided.

You mean an RC snubber, right?

Thanks.

[Tahmid]

Originally posted by Eva

Another approach that seems to work fine is to sense the output voltage indirectly at the primary side, through the primary winding not being currently used, thus sensing any flux imbalance too. Steg audio amplifiers (an Italian brand) do it that way.

Could you please elaborate on this?

Thanks.

[Tahmid]

Hi Eva,

I'm still waiting for your answers.

Thanks.