Subject: Ringing on PGND/GND

[elguerrero]

Hi everyone,

I’m currently developing a DC motor controller and I’m observing high-frequency ringing on the ground path.

System Specification:

• max 29,4V (7s Battery)
• Brushed DC Motor, PWM frequency at 16 kHz
• using Gate Driver with Rg = 47R
• Measurement: Probing between PGND (Power Ground) and GND (Signal/Logic Ground).

It seems that the Rg is fine since no ringing available on Mosfet Gate (low side) and Vds.
The ringing is existing only if I switch on the Motor.

Here is the measurement PGND<->GND:

Here is the roughly schematic:

My Questions:

1. Is a 400 mV ground bounce considered "acceptable" for a 28V system, or does this indicate excessively high parasitic inductance in the ground return path?
2. From an EMC perspective, at what amplitude does this type of ringing (in the 30 MHz range) typically become a "fail" during radiated emission testing?
3. What measures can I implement to prevent these ringings (RC Snubber on Mosfet, on Motor? , GND Star already exist, not in perfect way but improvement is planned)

[Paul_KD7HB]

@elguerrero

It seems that the Rg is fine since no ringing available on Mosfet Gate (low side) and Vds.
The ringing is existing only if I switch on the Motor.

Doesn’t resemble ringing to me, which will be a single DC pulse which you have an AC pulse.

It seems normal for a system with separate grounds. Why are they not connected? That signal is what motors do when turned on.

[jim-p]

@elguerrero

GND Star already exist, not in perfect way but improvement is planned)

Well that could be the total problem.

[Wawa]

@elguerrero

Here is the roughly schematic:

How about an accurate schematic, showing the actual ground layout with wire lengths, diameters and types of capacitors, and where you actually have measured. HF ringing can usually be killed with ferrite beads. You sometimes see one slipped over the drain pin of a mosfet.
Leo..

[elguerrero]

I found out that the issue is not from ground bounce.
The issue I am getting when the motor is switched ON and I am very near with the probe (ground clip is connected to probe tip) to the motor ~5cm. At 10 cm distance from probe to motor I am not getting this issue. The issue is also not there when motor is switched off!

What does it mean with regards to EMC? Does these EMI from Motor needs to be supressed or is it acceptable.

Remark:
There are no interference suppression capacitors at motor poles

I have done additional measurements just for curiosity of EMI. I have put the probe tip at GND (probe ground clip floating) of pcb and measured:

Any comments on this? Are the Interference of the DC Motor on Baseline expected?

Baseline / Idle State (Motor OFF) (50Hz Environment Interference)

Baseline + Motor ON

Zoom In of Screenshot 2 (Baseline + Motor ON) -> 16kHz (equal PWM Frequency)

Probe Tip directly on Motor

[jremington]

Have a look at this tutorial: Pololu - 9. Dealing with Motor Noise
This tutorial on noise in power leads may also be helpful: Pololu - Understanding Destructive LC Voltage Spikes

[PerryBebbington]

@elguerrero

I have done additional measurements just for curiosity of EMI. I have put the probe tip at GND (probe ground clip floating) of pcb and measured:

This doesn't make sense. So you don't have a defined ground input to measure against? Voltage is measured between 2 points, if you don't define one of those points then what you measure at the other is meaningless.

Please show a schematic or something to illustrate what you are doing.