current sensing using three shunt resistors

Hi Shady,

It will be dependent to the control method. We will try to find more resources and get back to you later.

Best regards,

Thank you Wilson Zuo for your reply

I use Direct Torque Control method of control .  This method is considered a competitive to Field Oriented Control for high performance torque control without current control and don't use PWM nor axes transformation and requires less parameter of the machine.

I am waiting for your resources and helpful replies

Thanks

Hi Shady,

Don't know how you can do the control without PWM. Anyway, the three shunts will all go to the OpAmps and outputs to controller. The three current shunts signal will be enough to calculate the three phase currents.

If using trapezoidal control, normally anytime you have only one current goes into UV, VW, or WU. There is one phase floating. So the one of the shunts will give you the related phase current.

In FOC, or other sinasoidal control, anytime all the three phase will have current and all the three shunts will have signals output. For example you will get UV, UV+WV, UW on the three shunts, You can always get the phase current by simplely checking two of the three shunts to get all the currents of UV, WV, UW. And the sum of the three will be zero.

There is a section in the following doc on how the shunts work in instaSPIN.
(Shunt Current Measurements)
www.ti.com/.../spruhj1f.pdf


Best regards,

if you want to use low-side shunt measurements you will need to be able to take those measurements in the center of a low side PWM on event to get an average current while current is flowing through the low side transistors. this may not be possible with the modulation scheme used by DTC (or at least much more challenging).

most drives I've seen using DTC don't use shunts, but instead use direct phase current measurements using hall effect sensors (usually LEM brand).

Thank you ChrisClearman for your reply 

but the DTC sheme i used is the switching table type . in this scheme the 8 basic voltage vectors of the inverter are only used to derive the motor (6 active and 2 zero voltage vectors) and no PWM technique is used here.so, is the method of three shunts under the lower switches be possible?

only if you are able to simultaneously sample an average current from 2 of 3 low side shunts.