Sinewave inverter

I had (naively, perhaps) thought that the main reason for having a transformer in a sinewave inverter circuit was to provide isolation, between the mains voltage output and the low voltage input, so that the latter would not be live.

My thinking was then that if that's not a consideration (because the input is also at a dangerous voltage) then one could dispense with the transformer.

Yet the standard PWM sinewave inverter circuit seems to rely on the very high transformer primary inductance for its function. Certainly, I haven't managed to conceive a transformerless PWM circuit that works, even in SPICE.

So I tried conceiving of it as a buck converter, where the regulated output voltage tracks the required sinewave. That doesn't work because there's not enough output current at lowish points in the output cycle to discharge the smoothing capacitor fast enough for the output voltage to track properly.

My generator's original inverter clearly did not have a 2.5kW 50Hz transformer, just two chokes (perhaps 300uH) on the mains output lines, and two electrolytic capacitors (220uF, if memory serves - certainly about that).

Equally, my 300W pure-sinewave inverter does not contain a 300W 50Hz transformer. It contains what looks like a transformer, but nothing like that big.

It appears I'm missing something, and multiple Google searches have not been informative. Anyone have knowledge of this?

Sylvia.

Reply to
Sylvia Else
Loading thread data ...

You don't necessarily need a transformer on the output of the inverter to provide isolation from the mains voltage to the low voltage side. AFAIK the way most "pure sine" inverters work is by using some kind of boost topology to create a high voltage rail, and then PWMing the high voltage using a modulator connected to an H bridge or something and filtering, like a class D amplifier. If the boost topology is isolated, like a flyback or half bridge or something, then you can opto isolate the MOSFET driver for the output devices , and then your power stage is isolated from the input.

--
----Android NewsGroup Reader---- 
http://usenet.sinaapp.com/
 Click to see the full signature
Reply to
bitrex

** Converting a DC supply to a 50/60Hz sine wave using PWM is not in any w ay magical - but since the PWM signal is not purely high frequency it canno t be passed through a small transformer.

When the DC supply is from a 12V or 24V battery, it must first be stepped u p to a high enough DC voltage so the PWM converter can directly output 120/

240 AC. Often, there are two PWM stages operating in phase opposition to ma ke this easier - enabling a single 340VDC supply to be converted to 240VAC.

The DC to DC step up circuit will normally be a high frequency, square wave inverter relying on the step up ratio of a small ( battery isolating ) tra nsformer. This type of inverter is very efficient but non regulating and th e PWM converter regulates against load and battery variations.

The PWM converter is then followed by a simple LC filter to remove unwanted high frequency components from the PWM wave - basically as an EMI measure.

When a petrol driven alternator provides the DC supply, the output voltage level is normally designed to be within the range needed by the following P WM converter. A regulating DC to DC converter is often included to extend t he usable RPM range of the alternator and improve fuel efficiency.

.... Phil

.... Phil

Reply to
Phil Allison

If you replace the diode in the buck converter with a transistor, and drive the two transistors alternately, you get a synchronous buck, which does track PWM% at all times.

All you need to do is remove the DC offset, so you get a clean sinewave output with no DC (an obvious drawback if it were an AC inverter..). This is usually achieved by running another inverter at opposite phase, and taking the output between them. So the DC levels subtract, and the AC adds.

Tim

--
Seven Transistor Labs 
Electrical Engineering Consultation 
 Click to see the full signature
Reply to
Tim Williams

Hi Sylvia

How about this one?:

Here is a one [active] switch inverter - but besides the one switch, it also has two parallel coupled reverse-blocking-switches closer to the output, that functions as active rectification - see the schematic at page 2:

A Synchronous Single Switch Inverter - Purdue School of:

formatting link
Quote: "... Four modes of operation were detected in creation of negative and positive polarity voltages. ... Not only did the bench test work, it lead to the discovery of several other circuits and controllers for high-power inverters with lower switching loss, higher voltage performance and lighter reconfigured circuits.

Therefore, as the number of high frequency switching devices is decreased, the efficiency is increased. For instance, a 90% efficient [H-bridge] converter becomes 97.2% efficient. ... CONCLUSION A new power inverter circuit was introduced that required only one high frequency switching transistor. The inverter used a synchronizing structure to change the voltage polarity on demand. Therefore, real time generation of infinite voltage levels was realized. The state space equations demonstrated a forth order system. ..."

The inverter could be used for solar micro-inverters:

formatting link

The inverter could also be used for:

  • brushless motor
  • step motor

The circuit is a "all-in-one". It could be used for (T1 and T2 refer to the iupui.edu article):

  • Positive DC, DV. T1 is used for active rectification. T2 is not used.

  • Negative DC, DV. T2 is used for active rectification. T1 is not used.

  • Any curve shape can be amplified with a signal from a suitable signal generator. (retangular, saw tooth, triangular...) T1 is used for positive curve parts - and T2 is used for negative curve parts.

Could it be used for a Class D audio amplifier with the right control circuit? Low enough distortion?

The circuit bear resemblance with a reversed SEPIC:

formatting link

-

Indiana University. (2012, October 17). New class of power inverter could mean cheaper, faster hybrid vehicles. ScienceDaily:

formatting link
Citat: "... Izadian's invention, the result of a creative reconfiguration of an electrical circuit during a laboratory experiment, would make inverters cheaper, lighter and therefore more efficient than current models. ... For example, unwanted harmonics are greatly reduced with Izadian's invention. ..."

-

Additional reading:

formatting link
formatting link

Active rectification (synchronous rectification):

formatting link

-

They must be but besides the one switch, it also has two parallel coupled e.g. Reverse Blocking IGBT (RB-IGBT). (Or two serially connected Power MOSFETs. The Power MOSFETs must be connected so that their substrate diodes have opposite directions):

formatting link

Application Characteristics of an Experimental RB-IGBT (Reverse Blocking IGBT) Module:

formatting link

A High Efficiency Indirect Matrix Converter Utilizing RB-IGBTs:

formatting link

-

Definition:

formatting link
Quote: "...

  • Active switch: Switch state is controlled exclusively by a third terminal (control terminal).

  • Passive switch: Switch state is controlled by the applied current and/or voltage at terminals 1 and 2. ... Single-quadrant switch: on-state i(t) and off-state v(t) are unipolar. [e.g. diode-like, reverse blocking] ..."

Better name: One active switch inverter.

The two active rectification switches are passive switches.

.

A half bridge uses two active switches.

A full bridge uses four active switches.

-

Alternative:

Design:

T1 serially connected to T2.

T1 is N-MOSFET with drain "up" (drain connected to L2 and C1).

T2 is N-MOSFET with drain "down" (drain connected to zero/commen).

T1 and T2 sources connected together.

.

Active rectification:

When positive output is needed T2 is on. T1 do active rectification.

When negative output is needed T1 is on. T2 do active rectification.

Glenn

Reply to
Glenn

It'll take me a while to work through that.

In the mean time, I've managed to get something in Spice to give more sensible results, though the output filter still requires a large inductor (~1mH) capable of operating at 50 - 100kHz, and 14A peak. These are not out of the question, but they're not so easy to source either.

The original inverter had a couple of large inductors wound on 45mm (outside diameter) toroidal cores. I'm a little sceptical as to whether these really avoided saturation, and thus worked as an effective filter, at high loads. Certainly if they're just an ordinary powered-iron core, rather than something more exotic, the math suggests that they'd saturate. I never looked at the output before the inverter failed. Maybe doing so would have been enlightening.

Sylvia.

Reply to
Sylvia Else

Wow! this is quite a post! Thanks. I know little about SMPS. The paper has only simulation data. By bench test I assume you (or someone) has built it and tested it?

The fast switch is the RB-IGBT you refer to later? What about T1 T2? (I haven't chased down all your links. that would take a while....)

George H.

Reply to
George Herold

...

Hi George

For T1 and T2 (passive switching/active rectification):

  • you either use two parallel RB-IGBT (as mentioned in the article)
  • or use two series connected MOSFETs

I have unfortunately not tested it yet.

Glenn

Reply to
Glenn

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.