Hi Bulent,
please see below an answer from an application engineer I have contacted.
DESRIPTION
The best V+ protection will depend on the load and operating parameters. The input filter capacitor is valid for protecting against the avalanche condition caused by a total loss of power during normal operation. In such a case, the capacitor should be large enough to absorb all of the energy in the load without exceeding the breakdown voltage of the MC33883. If the assumption is made the the load is primarily inductive, then the inductive energy will need to be converted to the rise in voltage between VPWR and VPWR_max.
Since the energy in the inductor is 1/2*L*I^2 and the energy in the capacitor is 1/2*C*V^2, the size of the capacitor can be easily calculated using the formula:
C=L*(I/V)^2 : where I is the load current and V = (Vpwr_max - Vpwr)
Two problems with relying on the filter capacitor only are:
1. If the Vpwr is out of range, or the current is greater than accounted for, the peak voltage on the capacitor could exceed the maximum rating.
2. The size of the calculated capacitor may be prohibitive, either due to physical size or cost.
In this instance, using a zener in parallel with the filter capacitor can be used to dissipate the energy which would have resulted in an over-voltage condition.
The thought of using TVS across the output terminals could work, but in general the tolerances for the TVS voltages are not sufficient to insure the voltage range on the input is not exceeded while keeping the losses due to leakage from causing problems. Those losses are cycle by cycle, while the zener across the filter capacitor are one time during the fault condition.
The schottky diode arrangement described simply provides a bypass for the lowside body diodes in the MC33883. They will have minimal impact since the body diodes are already performing this function, just with a little higher voltage drop.
With Best Regards,
Jozef
