low side shunt on CCB021M12FM3
Hello,
I'm designing a motor controller around the CCB021M12FM3 and am wanting to do current measurement with a 3 shunt resistors on the low sides of each leg. What would be the ramifications if I float the S2/S4/S6 pins so that I don't have to have a low side power supply dedicated to each leg? In other words, each DC- line would go to the shunt for that phase, then to the low side ground.
I realize this would probably throw off the DESAT detection of the ADUM4146 I am using, but are there other concerns?
Comments
-
TBhatia Wolfspeed Admin - Contributor Level 5
Hello, thank you for your message. A member of our technical team is looking into your questions and we will respond shortly.
-
JHayes Wolfspeed Employee - Contributor Level 3
Hi andersonm,
I think the biggest impact would be to switching losses. If the kelvin-source connections are not used for each of the low-side devices, then the gate loop and power loop would share a common-source inductance composed of the power wirebonds, DBC trace, press-fit pin, PCB traces, and shunt resistor, which could be significant. This common-source inductance would increase the switching losses for these switch positions.
It is difficult to say by how much, so you will ultimately need to verify the performance for your application. But it is highly recommended to use the source-kelvin pins.
Thanks,
Jonathan
-
andersonm Contributor Level 1
OK, so I guess best practice is to either:
- Bring all the DC- points and the kelvin-source connections together and then go through a single shunt to Vbus-
- Or have 3 isolated supplies for the low sides and do a shunt on all 3 legs
Does that sound right?
-
JHayes Wolfspeed Employee - Contributor Level 3
Hi andersonm,
Yes, that is correct for low-side shunt measurements, although combining the three bridge-leg currents into a single shunt measurement may not provide the information you need for motor control.
Alternatively, you can look at using a shunt measurement on the phase outputs (UVW), referenced to the source-kelvin of switches 1, 3, and 5. This at least takes advantage of the isolated supplies required for the high-side switches and allows you to use a single supply for all low-side switches.
Of course, there might be a cost trade-off between phase shunt measurement and isolated supplies for the low-side that you will need to consider.
Below is an example of this type of measurement where an isolated delta-sigma modulator is used.
Thanks,
Jonathan