I have a circuit designed to run a electro piezotransformer at resonance frequency of around 80kHz. As a controller was used microcontroller Attiny 25, which by measuring feedback current generates PWM with the resonant frequency to drive the piezotransformer. Operating voltage is 4V and current around 1.5-1.6 A. The filter inductance is overheating due to the saturation and output current drops significantly, till piezo stops operating. Before was used 4.7 uH inductance with saturating current 4A, now I tried 6.8 uH @ 3.6 A from Murata, but the result is the same. What could be done to overcome that problem? Any help is very appreciated. Thank you very much in advance. Thanks again for the response. I have another question regarding output voltage and current of the driver. Voltage measured on the input of the piezo has a square wave shape on the positive part and sine wave form on the negative side. But the current shows very strange shape and frequency about 160 kHz instead of 80 kHz. Please, I need your help to understand the current shape reasons and to calculate the output power of the driver. It is very important for me. Thanks in advance.[The blue one is a voltage, and yellow is a current2
Isat must be avoided by current limit sensing cycle by cycle to avoid thermal runaway problem. As L is rated at -10% for Imax, it continues to drop rapidly for increasing current, which in turn reduces output voltage and thus demands more current and L spirals down.
Your circuit filters out Imax peaks with a LPF so you must regulate to average current at a lower value, such that Ipk does not exceed rated current. Increasing L tends to reduce Imax limit for same size.
Consider output power you need and what current required at 3.3V Choose 10A choke like this http://www.coilcraft.com/xal1010.cfm