Tool/software:
I currently use DRV8300-Q1 on my custom 1 channel BLDC motor driver.
When tested for Radiated emissions, I have observed some peaks at 48MHz.
When the motor is driven, only then this noise is observed, so i am suspecting noise emitted by the driver circuit.
I have the emission observations below,
Can i get some help on how to deal with this emissions?
1. Please find the waveforms below,
GHA
SHA
GLA
I have following question,
1. Why is the rise time for high side only 8ns and low side is 228ns.
2. I tried to use snubber based on instructions in 7 Step procedure (https://www.ti.com/document-viewer/lit/html/SSZTBC7). But observed that by adding snubber amplitude of ringing noise was increasing.
without snubber (6.4V),
with C1 = 94pF (10.4V)
1. Supply voltage is up to 58.8V, with motor compatible for the same.
2. GVDD used is 18.5V
3. GHx wrt SHx is,
4. I followed the link provided, however on add C1 capacitor, the frequency of ringing reduced, which lead to a frequency ration of less than 1. So the final capacitor calculated is in negative.
5. I tried increasing the gate resistor from 22R to 60R, and following is the GHx wrt SHx,
I am not sure why there is noise in the GHx wrt SHx,
Only one of the GHx exists at a time. They do not overlap. Please check the SHA and SHB provided below,
INH1 and INH2 is below
INH2 and INH3 is below
Note:
None of the High side switches at the same time. Also low side control is given as continuous high for the respective duration of hall state high (100%) instead of PWM with duty cycle.
Hey Praveen,
Where on the PCB are you measuring GHx, and SHx from? are they being taken on the driver pins or the MOSFETs?
In this measurement the ringing in the middle, could be coming from a fast slew of SHx of the same phase (the noise is present around the miller region, where VDS slew occurs). So recommend tuning down the slew rate using larger gate resistances.
Also for RC snubber, you may need to play around with the values to find the ideal solution.
The voltage overshoot and ringing on SHx turning on could be resulting in the 48-60MHz noise.
Also noticed that using GVDD at 18V would be resulting in 18V as your gate drive voltage, if this within your MOSFET specification? Most applications we have seen you GVDD at 12V since that is enough to fully enhance the gates but not enough to damage the mosfets. Do you have your MOSFET datasheet you can provide/schematic?
Best,
Akshay
Hey Akshay,
Sorry for the delayed response.
1. The waveform recorded is on the MOSFET pins.
2. We increased the resistance from 60R to 75R and saw no change in noise, effective total rise time increased from 550ns to 740ns, however that noise is still the same. Below is the GHx wrt SHx when 75R gate resistor was used,
3. Ringing has drastically reduced with 60R as shown in the image below,
4. Mosfet Gate voltage is upto 20V, hence 18V GVDD was used.
5. Mosfet used is NVMFS6H824NT1G
1. Please find below the rise time of SHA pin wrt Driver GND.
2. Also, wouldn't there be possible shoot through if we increase the gate resistor further. Currently the GHx wrt SHx is about 550ns for 50us PWM time period.
I understand the rise time of SHx is 9.2ns even though the Gate rise time is 550ns. but i might have to use resistance of 500R or even 1k to achieve this, which doesn't sound right.
3. Also i do not understand why we need to see the values of GHx or GLx on the driver end, as the effect of gate resistance cannot be seen on this location.
