Why do the ouput of the inductor filter exhibit two peaks when connected to an inductive load of 150

Hi,
I am designing an inverter capable of driving 1500w load.In my design i am using Push pull Topology.In my design i am driving the primary side of the transformer using two MOSFETS.
The gates are controlled by PWM signals from PIC Controller.As of every design,on the secondary side i have a bridge rectifier and an LC filter.The output of the LC filter is connected to the H-Bridge for AC Generation.
I have attached the waveform which i obtained on the various stages of testing.I did probe all the wave forms of voltages at various pints on the Schematics.
Let me now discuss about the problem which i am facing now:
In the schematics you could see a bridge rectifier circuit and an inductor circuit attached.Before that let me give a short intro about the transformer i am using now.Its a push pull type transformer capable of taking a voltage input in the range of 24v to 40v.The leakage inductance was measured to be  14mH, at 100KHZ 1.0V by N2+N3 short circuited, and measuring  N1+N4.The expected and calculated output is about 400v.
Presently i am testing by giving PWM signals with controlled width to the MOSFETS on primary side and thereby controlling the flow of the current to the primary coil.
As designing engineers ,all of us are familiar with the output of the bridge rectifier and LC filter.
A voltage of 24v was applied to the input of the Transformer for testing purpose.The transformer output was measured to have an AC voltage of 800v peak to peak.The rectified bridge rectifier output was observed as 400v (which was an expected output). The testing was done by connecting a bulb of 200w which is an inductive load. Later an L Filter was connected to the output of the bridge rectifier and the output was observed.I was really astonished to see the output.

Before explaining more about the issue,let us just go through the working of an inductor.
When a DC voltage is applied to an inductor, it suddenly shoots to voltage equal to the applied DC voltage.The shoot can be seen in the output due to the Back EMF generated in the coil because the inductor opposes the sudden change in the current.Slowly we can see that the peak decays and finally the inductor behaves as a short circuit or a zero ohms resistance path.
But if you just check the output across the inductor in my design, you could see Two peaks corresponding to the pulses on the bridge rectifier output in contradiction to the expected output of the L filter.
In my design the value of L and C were taken to be 56uH and 33uF each.
Giving more details regarding the two peaks,
First peak seem to have a shoot up to 500v and it can be seen that the peak decays slowly.But the voltage was observed to shoot again after decaying for a certain period.Please note that normal 200w bulb which is inductive in nature is connected as a load on the secondary side of the transformer at the L filter end.i couldn't find any explanation for such a phenomenon any where.I posted similar questions in Texas Instruments Forum too, but they were also not able to help me but they tried a lot.

Later the output of the LC filter was checked.I was able to get a clean DC voltage of 400v at the end of the LC filter.

I need to find the reason for such a behavior of the inductor because when i checked the output of the transformer at the same instant i found the change in waveform even at the transformer output when a L filter is attached to it.I mean two peaks in each pulses at the output of the transformer.

.

Some of my doubts regarding the issue is noted below:

1.Is it due to some inductive kick back from the bulb which i have connected for testing at the end of LC filter?
2.Leakage inductance of the transformer be the reason causing this phenomenon?
3.Is the inductor getting saturated while testing?
4.Is it due to some type of Harmonic Distortions i am observing such type of waveform across inductor?



What could be the possible reason for such an output? Please share all of your suggestions but the issue?

For the scope waveforms showing the double pulses, please tell us exactly how you captured them, and what the vertical volts/division settings are. Are the double pulses simultaneous with the rising and falling edges of the PWM pulses?
What is the self-resonant frequency of your inductor? How is your LC circuit connected to your ground plane?
The two pulses in each double pulse have the same polarity. That is unusual.

Hi,Thanks for the Reply.Please note that the ground on the secondary side is named as AC _GND.You could see the net name PSHPL1_OUT in the schematics.PSHPL1_OUT is the Ground in the design for the secondary half.PSHPL1_OUT flows though a current sense resistor and becomes AC_GND.Just to avoid the clash in net names they were named different.Actually both are the same."For the scope waveform showing the double pulses, please tell us
exactly how you captured them, and what the vertical volts/division
settings are"Let me explain that.I have just mounted the L Filter at the end of the bridge rectifier with out mounting the Cap.Later i probed the input to the L filter and output of the L filter (without placing the Cap C) with respect to the AC_GND.
Lets go through the waveform, the first yellow channel shows the inductor output when probed wrt to ground.And Blue channel the input to L filter wrt to ground.Please note that channel 1 and 4 are active and 2 &3 are off at this instant.Both the channels are having a volt per division of 500v each.You could see the 500v mentioned in yellow and blue on the bar shown just below the waveform.The Output pulses were observed to be simultaneous to the input pulses.The part number of the inductor i am using is S14100036."What is the self-resonant frequency of your inductor?"The frequency as per data sheet was 100KHz.
"How is your LC circuit connected to your ground plane?"As mentioned above PSHPL1_OUT and the AC_GND are the same.AC_GND is the isolated ground on the secondary side of the board.Actually what happens is that PSHPL1_OUT  flows through a current sense resistor and becomes AC_GND.
"The two pulses in each double pulse have the same polarity. That is unusual."Yes that's true.