Hi,

I noticed that your circuit only had high frequency decoupling (the 100nF), which is good, but there should be bulk low frequency decoupling as well. Maybe you already have it and it's not shown.

You will notice that in the LTE-5202 datasheet, they have a large decoupling capacitor (100uF to 4700uF) on the power pin.

MOSFETs are very good at instantaneously sucking energy from the supply to drive the load. The instantaneous power delivered to the chip is limited by the power supply source impedance and the inductance of the power interconnect to the Motor Driver Chip.

The bulk low frequency capacitor acts as a "local battery", much like a large capacitor (1 Farad or more), is located near a car power amplifier to reduce the "local" source impedance that the amplifier sees. It improves the local storage for the amplifier to improve low bass performance. The midbass and high frequency energy source impedance is supplied by the decoupling within the power amplifier.

The load in your case is composed of FET RDS(on) in series with the motor DC resistance. The time frame is governed by the FET edge rate, motor inductance, and interconnect.

The capacitor will also improve the initial torque of the motor as the voltage on the FETs remains closer to the applied voltage during the transient. The FET Bridge voltage droops when the motor is started or moved from a position where it draws higher peak currents. The capacitor keeps that voltage closer (more available current) to the rail voltage over the transient time. If capacitor is not there, the FETs will be starved for energy during that timeframe.

The capacitor could improve stability & reliability too.

Ross
_________________________
In SI, a little termination and attention to layout goes a long way. In EMC, without SI, you'll spend 80% of the effort on the last 3dB.