I burnt almost dozen mosfet while trying to test the switching of secondary mosfet of my flyback inverter. It was in fact something trivial because i thought to switch one mosfet at a time to get a half wave waveform at a time. I discussed it with my supervisor and he pointed me towards the solution. "Operate both the secondary mosfet instead of one at a time" he said. "Because when you operate only one switch there comes a problem, that is during its turn off time there is no alternate path for the energy stored. Therefore, high voltage spikes emerge that are resulting in the failure of mosfet" he added.
I operated both the switches simultaneously, such that for positive half cycle Sac1 i turned on and for negative half cycle Sac2 is turned on. The experiment of getting one half cycle only by trying to switch the Sac1 alone should be avoided.
![]() |
Flyback inverter (Click to zoom) |
For worldwide viewers
For viewers in Pakistan
The flyback inverter secondary winding acts as an unfolding stage in a grid connected scnario. The switches associated with them needs gate driver circuits. The question here is that can we consider them as a half bridge circuit so that it is possible to use a half bridge driver ic with it. To answer this consider the Fig. 1 that depicts two different types of unfolding configurations. The working although is same lets consider the one with no center tapping.
Problem : While experimenting the flyback inverter i observed a visible spark on the output of the flyback transformer.
Reason: One of the secondary winding was open possibly because of a dry solder.
Solution: Hunt for the open circuit and solder it properly.
Reason: One of the secondary winding was open possibly because of a dry solder.
Solution: Hunt for the open circuit and solder it properly.
The switching frequency is 100kHz. The mosfet is connected on the primary side of flyback transformer. It is a low side mosfet and hence the gating signal should be provided with respect to the ground (PV panel negative). The gating signal however, appears to be a ramp with a sudden drop to zero , just like a triangular waveform. The possible issues as pointed out by experts contacted on various forums are as follows
Consider Fig. 1 below to understand the topic. When the primary switch Sm is turned on, no current flows to the grid because of the direction of the diodes on secondary side.
A well explained answer is given in the reference below. Quoted as it is from the text.
Discontinuous Conduction Mode (DCM) is widely used in flyback inverter. The advantages and disadvantages of using DCM are given below.
Microinverters for single phase grid connections have an inherited issue which is commonly known as double line frequency problem. The problem is about the power pulsations of the AC grid which is double the line frequency. This power is time varying in nature while the power output from the PV module is constant (MPPT is used ). This is shown in Fig.1. This means that there must be some storage device for balance of power. This storage device is often called as decoupling capacitor.
Flyback inverter is a device that converters the DC into AC with the use of flyback transformer. For this inverter topology no dc-dc converter is required as the dc voltage is not related to the operation. It is simple, has less component count hence reduced cost and has advantage of isolation through transformer[4]. Let's see how it works.
The designing of the flyback transformer requires to list the following quantities ahead of design.
- Input values
- Output parameters
- Switching frequency
The question is what scheme of conduction to be used for PV grid tied flyback inverter. Either the Discontinous Conduction Mode (DCM) or the Boundary Conduction Mode (BCM). So here is the literature review for this question.
The difference between DCM and BCM is shown in the figure above. Now that it is clear that DCM has dead bands in it while BCM operates at the boundary of CCM lets have evaluation of these conduction modes.
![]() |
Difference between BCM and DCM |