troubles with inverter currents

Discussion in 'General Electronics Chat' started by cachapo, Aug 24, 2009.

  1. cachapo

    Thread Starter Member

    Aug 24, 2009
    12
    0
    Hi everybody,

    I´ve got a three phase inverter circuit to control a BLDC motor, with 6 MOSFET as switching devices. I´ve put current sensors in the DC bus and in two of the three phases. The DC bus sensor shows a current that is the same as the power supply ( for example 2A), but the two phase sensors show a current that is much higher ( about 5A).

    Does somebody have any idea about what is happening? Thanks very much

    (Sorry for my english)
     
  2. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    I'd guess it's regeneration effects.

    The transistor bridge will (should?) have diodes from the motor terminals back to the DC side, effectively forming a three phase bridge rec.
    (These are often internal in the power transistors).

    Each time a motor winding is switched off, the current in it 'flywheels' through these diodes back to the DC storage caps.

    Because of this, the AC current can be rather higher than the DC.

    Many inverters (and AC servos) can put current back into the DC side while braking the motor to a stop.

    The really fancy ones have an extra inverter to feed back to the mains supply, but most just have a big resistor they switch in if the DC side starts to go overvoltage due to regeneration.

    Keep an eye on this & make sure you don't overvoltage things while stopping the motor, especially if it ends up with a significant inertial load.
     
  3. cachapo

    Thread Starter Member

    Aug 24, 2009
    12
    0
    Hi Robert,

    First of all, thanks for your reply.
    You are right, power MOSFETs have an internal diode.
    Do you know any paper explaining regeneration effects?
    And just another question, which current have I to use to calculate motor torque? DC link current (about 2A) or phase current (about 5A)?

    Thank you very much.
     
  4. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    It's basic inductor theory, offhand I don't know of any specific links.

    The basic idea is:
    When you apply a voltage across an inductor, the current builds up slowly as the magnetic field around it gains strength. The changes in the magnetic field oppose the *change* of current. ('Slowly' is relative, could be sub-microseconds or seconds depending on values).

    If you disconnect the supply after the current is established, the energy stored in the inductor has to go somewhere. Think of it a bit like a flywheel, spun up to speed.

    Whenever you switch an inductive load on a DC supply (like a relay coil), it's standard practice to fit a 'flywheel diode' across the coil.
    This allows the current to continue circulating without doing any harm, the stored energy dissipates as heat due to circuit resistance.

    If you do not provide an external current path for this stored energy, the current in the coil still tries to keep flowing, and the voltage across the coil increases to (theoretically) around infinity... Think Tesla coils..

    In practice, things tend to arc over with a voltage somewhere lower than infinity.

    With small inductors, often the internal losses mean nothing much visible happens.
    With larger and higher power inductive gear, either you get sparks on the terminals or, typically in motors & transformers etc., the (very thin) winding insulation fails within the coils, shorting them and the motor or whatever is wrecked.
     
  5. cachapo

    Thread Starter Member

    Aug 24, 2009
    12
    0
    In order to a better understand, here you have the schematic of the power circuit. I think regeneration current should flow trough the DC link resistor, and charge DC link capacitor.

    The current flowing through that resistor is the same that I can read in the power supply. My problem is with the low side resistors. The current flowing trough them seems to be much higher.

    Do you think it would be recirculation effect? Can the recirculation current be so high?

    Thanks.
     
  6. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    How are you reading the current? If you are just using a meter, you will likely be seeing the average rather than the peak.
    With a meter, 10A pulses at 20% duty cycle may read as 2A.

    If you look at the waveform across one of the low side resistors and also the drive waveform with a 'scope, you should get a much better idea of what's happening.

    On the DC side, you could find that you have something like 6A forward and 4A regenerated; you will only read the 2A difference.
     
  7. cachapo

    Thread Starter Member

    Aug 24, 2009
    12
    0
    I´m going to upload some 'scope screenshots:

    RED signal is the voltage in one of the lower resistor, amplified x33. This gives a current of 4.7A ( sense resistor is 0.005ohms).

    BLUE signal is the current that flows into the motor, measured whith a hall effect chip. It gives 100mV/A, so this current is 4.4A.

    In the power supply I can read 1.8A.

    The second screenshot is just a zoom. Duty cycle is aprox 70%.

    I´m not sure about if it could be regeneration current, because my power supply don´t says FAULT. I think that inductor forced currents recirculate trough the low side MOSFETs and diodes, but I thought those currents wouldn´t be so high. What do you think?

    Thanks for your time.
     
Loading...