TMS320F28379D: Questions about TMDSIDDK_v2.0 Code

TI-CSC

Genius 12945 points

Part Number: TMS320F28379D

Tool/software:

Hello E2E Experts,

Good day.

EPwm1Regs theoretically corresponds to phase A, why does it correspond to svgen1.Tc in the code?

Regards,

TICSC

3 months ago

0 Jiaxin Teng 3 months ago

TI__Expert 5440 points

Hi,

You are right, traditionally and theoretically in the 3/2 transformation, phase A is aligned with the alpha axis, and resultant EPwm1Regs corresponds to phase A. However, this code base uses different axis for dq transformation, thus cause this difference on the pulse width mapping.

Thanks,

Jiaxin

0 TI-CSC 3 months ago in reply to Jiaxin Teng

TI__Genius 12945 points

Hello Jiaxin,

Good day.

// Clarke transform macro (with 2 currents)

//==========================================

#define CLARKE_MACRO(v) \

v.Alpha = v.As; \

v.Beta = _IQmpy((v.As +_IQmpy2(v.Bs)),_IQ(ONEbySQRT3));

#define PARK_MACRO(v) \

v.Ds = _IQmpy(v.Alpha,v.Cosine) + _IQmpy(v.Beta,v.Sine); \

v.Qs = _IQmpy(v.Beta,v.Cosine) - _IQmpy(v.Alpha,v.Sine);

I still have questions. In the codebase, the 3/2 conversion seems to be phase A aligned with the alpha axis, how to understand what you are talking about the codebase for DQ transformation using different axes?

Regards,

TICSC

0 Jiaxin Teng 3 months ago in reply to TI-CSC

TI__Expert 5440 points

Hi,

If you are using the exact same hardware, you may find the the current sensing only sample the V and W phase and also takes phase V (not phase U) align with alpha axis no matter if it is SDFM sensing or LEM sensing.

This is different with the FOC implementation in the universal motor lab solution, where the phase A/U is set to align with the alpha axis.

Schematic of IDDK hardware