Three Shunt phase current measurements in case overmodulation

see proj_lab10

Unfortunately the Lab write-ups are still not complete.

But in the code you will note

SVGENCURRENT_Obj svgencurrent;
SVGENCURRENT_Handle svgencurrentHandle;

you will notice in the mainISR the new functions used

  // run the current ignore algorithm
  {
    uint16_t pwmValue_1 = HAL_readPwmCmpA(halHandle,PWM_Number_1);
    uint16_t pwmValue_2 = HAL_readPwmCmpA(halHandle,PWM_Number_2);
    uint16_t pwmValue_3 = HAL_readPwmCmpA(halHandle,PWM_Number_3);

    // run the current ignore algorithm
    SVGENCURRENT_RunIgnoreShunt(svgencurrentHandle,pwmValue_1,pwmValue_2,pwmValue_3);
  }

  // Set trigger point in the middle of the low side pulse
  HAL_setTrigger(halHandle,SVGENCURRENT_getMinWidth(svgencurrentHandle));

If you are using DRV8312 or DRV8301, 100% duty cycle is supported since the gate drive is not done with a bootstrap topology, but with an internal charge pump.

If you are using the high voltage kit, yes this uses a bootstrap topology. In this case, you can do 100% duty cycle, as long as you are producing a alternating voltage (AC voltage) to the motor, at a relatively high frequency. For instance, if your voltage output frequency is 100 Hz, even if you are generating 100% duty cycle outputs, the bootstrap will be charged at a 100 Hz rate.

In the case of the high voltage kit, charging the bootstrap takes a couple of milliseconds, since there is a 10 ohm and a 100 uF capacitor:

and to discharge would take much longer than this, since this goes to the gate of the output transistor. So I don't really see an issue, unless you are trying to generate a 100% duty cycle as a DC output, then I would suggest not doing it with the high voltage kit.

-Jorge

Thank you,Jorge

No, I don’t use all these “hundred dollars” kits from TI.If you buy one ,then you must be afraid that you will burn it or destroy somehow.

Instead, I built simple setup based on IRAM136-3063B module.You can see it on the attached photo.External bootstrap caps on the module are 2uF each.

This simple setup I connect now to “normal” LaunchPad with TMS320F28027PTT on it.

I learn Structure of the system and C functions of InstaSPIN. Even with this simple cheap setup many things are possible.

Hello Andrey,

One more thing, I think you should use TMS320F28027F if you want to use InstaSPIN code.

Best regards,
Maria

Andrey,

Granted our BOOSTXL-DRV8301 is only 24V/10A while you are using a 600V IR driver, but they are both the same price, $49.  Our Boost also includes direct connection to the LaunchPad and of course 10s of software projects with all drivers and libraries ready to use immediately. 

Probably most importantly the layout of the inverter and current sense has been optimized for excellent performance.  I really have to wonder how good your signals are and what sort of control capability you can test.

Just food for thought.

Hello,Chris

Make BOOSTXL inverter 48V/10A DC bus and all EBIKE users will be yours, I will also buy.Keep price below 100USD. For TI it is nothing/peanuts to invest. But customer base will be expanded. Small windpower PMSG can be considered as well.

Hello,Maria

Yes,its true. I still can solder two ACS712 hall current sensors on that board.

Andrey,

Yes, we could do that...but we are actually working quite closely with many existing customers in this space. So we have to be careful about how far we take an EVM....you don't want it to turn into an actual product ready for manufacturing, you know?  But in most of these e-bike controllers they are using discretes, so maybe we make a high end version with our DRV83x intelligent drivers and high efficiency NexFET MosFETs and it would be a good fit for the scooter/e-bike/helicopter enthusiasts and serve some other markets.