Other Parts Discussed in Thread: DRV8705-Q1
I have trouble passing CISPR11 EMI testing with a board designed for 3 drives with DRV8701 ICs. I looked at two other similar posts and gone through the same tests. The test fails with the motors disconnected, it fails without pwm, the only way to pass is when the drive is in sleep. The Idrive is at 12.5 / 25 mA with 33kOhms, which is as low as it goes with the fets I am using. The capacitors are set as per the datasheet and it is a 4 layer board with ground directly underneath, the caps are close to the chip. Using near field probes I find the same frequencies that were found in the chamber at 3 meters. The peaks are found when the probe is directly over the AVDD and DVDD capacitors as well as the charge pump capacitors. I cannot determine for sure that the noise is comming from the regulators or the charge pump. I tried adding smaller capacitors in parallel to the AVDD DVDD to no avail, adding capacitor to the charge pump output did not solve it either. I am near certain the DRV8701 is the cause of the 100Mhz noise, I think it couples to VM or Ground and is emitted through the battery leads that are long enough to act as an antenna.
Is there any way to filter the noise the regulators or charge pump?
Results of far field:
Results of near field:
