INA240: current sensing

Hi Mohammad,

Can you please post related questions to the same thread and not opening new thread for each? Please click on “this resolved my issue” to close the other thread if you decide to abandon it.

Regards, Guang

Thanks for response,
I have another question,
can you give me a example of bandwidth calculation for a arbitrary number of poles, RPM and settling time?
Regards, Mohammad

Hi Mohammad,

I’ll leave it to your to look into the relationship among motor speed, poles and PWM switching frequency. There are online documents/pages you can refer to, such as this one. http://www.ti.com/lit/an/slyt692/slyt692.pdf

The PWM frequency determines how much time the current sense amplifier has to react to a change in current when cycle by cycle current control. With a 10KHz PWM and 50% duty cycle, the upper limit is 50uS. This is doable with almost all devices.

If the PWM becomes 50KHz, the upper limit for the settling time is now 10uS. In this case, you probably want to select that device that has a slew rate at least 1uS/V and can settle to a reasonable accuracy within a few uS. 

Regards, Guang

Hi Guang,

Guang Zhou said:

The PWM frequency determines how much time the current sense amplifier has to react to a change in current when cycle by cycle current control. With a 10KHz PWM and 50% duty cycle, the upper limit is 50uS.

TI has test the 240 in-phase monitor up to 100KHz PWM 50% duty via TIDA-00909. If I recall it compares 50% PWM duty cycle via scope captures 40KHz - 100KHZ. That hooked this fish easily until later discovering the modulation used was specialized Space Vector PWM reference SPRA524.  

BTW the 240A1 (low side) is very quiet via 12 pole BLDC motor and trapezoidal 20KHz PWM, output settles <2.5us. We tested same 12 poles to 40KHz with the same PCB produced similar results @24VDC.

Hi ,
About the duty cycle:
1-)Isn't the duty cycle between almost 0% (for no load ) and 100% (for full load)?
2-)I think the amount of duty cycle is just important in the low side current sensing (not in inline) because in low side current sensing we should read the current whenever the bottom switch is on and this time is depend on duty cycle and we don't have this constrain in inline sensing.
Please correct me if I am wrong.
Regards, Mohammad

Hi,

You can still have large current in the motor under no load conditions, where the speed will increase. When loaded, the speed will decrease and torque will increase if you keep the current constant.

You’re correct however, there are situations where the duty cycle can be almost 0%, in a PWM modulated sinusoidal drive. The saving grace is that at any given moment only two phase currents are required in order to derive the motor speed and position information. If one phase is too difficult to measure, this information can be derived from the other two windings. However the current sense amplifier does determine the top PWM frequency. As a hypothetical example, if the CSA settles to 1% in 5uS, then the top PWM can never be more than 200KHz because at that point the CSA needs 100% of the duty time just to settle.

Low side or inline current sensing doesn’t make a big difference in terms motor control itself. However, in addition to motor safety and health information, inline sensing provides the direct winding currents which does reduce the amount of math the controller has to deal with.

Regards, Guang