• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 435 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MC33063A-Q1: External P-Chan FET switch for more current! Inductor Current?

BobL
Intellectual 330 points
Part Number: MC33063A-Q1
Other Parts Discussed in Thread: MC33063A, ALLIGATOR

Support,

I'm using an external P-chan FET for higher supply current. I worked through the formulas to determine my minimum inductance and output capacitance needed for my application. What should I be seeing for the current going through the inductor.

The datasheet shows a consistent sawtooth starting at 0A and rising to about double my required DC current requirement. My circuit actually shows the the sawtooth bouncing around and teeth beginning in the middle of another teeth. Some teeth don't start or drop to 0A. I'm driving a brushless DC fan (~1.6A running) and am thinking my layout (circuit ground and current paths may not be adequately controlled) and the fan noise may be coupling into the MC33063.

What should the inductor current look like in a properly working switcher? Should the sawtooth always start a 0A and end at 0A? Will there be gaps between teeth? Should the peak amplitude of all the teeth be consistent? This is a PWM controller. The inductor waveform frequency should be consistent (set by Ct), and the output voltage is controlled by on/off time? Is that correct? The Oscope waveform below is a result of placing a 0.01uF cap across the lower feedback resistor, otherwise the waveform looks way worse.

Thx in advance!

  • 0
    TI__Guru 62565 points

    Hello Bob,

    Thank you for designing with the MC33063A.  I am not sure if your driver circuit for the external PFET is correct.  Note that FET requires a large but short pulse current, while the Bipolar which the MC33063A intends to work with just needs a smaller but continuous current to drive. Anyway, could you monitor the Vgs of the PFET along with the inductor current?

    Thanks,

    Youhao Xi, Applications Engineering

  • 0 in reply to Youhao Xi
    Intellectual 330 points

    Ch1 is Gate

    Ch2 Inductor Current

    Ch3 is Source

     

    The Oscope ground leads are the long alligator clip leads.

  • 0 in reply to BobL
    TI__Guru 55765 points

    Hello BobL,

    It seems that the current waveform looks different in your second measurement compared to the first one. Did you change anything else than just adding some probes?
    I expect with Source you mean the voltage on the SC pin, correct?

    It seems that the device needs relatively long to turn on the MOSFET (clearly observed Miller plateau at turn on and turn off) which seem to cause a lot of noise in the system.

    Could you please observe the voltage on the current limit resistor as well?

    Please try to use short GND connections for the measurement of the current limit resistor at least.

    I wonder if you could check with a bipolar instead of the MOSFET externally.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Intellectual 330 points

    I have the current limit resistor jumpered out with a 18 AWG bus wire. The Source trace is the Source pin of the FET. I added a 0.01uF across the lower resistor in the feedback circuit. That helped stablize the circuit better.

    I highly suspect that my ground needs work. I'm going to respin the board with ground and power planes and try to do a better job of controlling current paths and noise, and also keep the feedback line short and isolated. I wanted to use a FET to keep the power dissipation of the board down. Unfortunately, FET on resistance is inversely proportional to FET capacitance.

  • 0 in reply to BobL
    TI__Guru 55765 points

    Hello BobL,

    Thank you very much for the feedback. You can find a lot information on layout and GND connection in a lot of different app notes and power supply design seminar topics on www.ti.com support and training tab or technical documents. Grounding is very important, but please do not forget that you check not only the DC current loops. It is most important that you first layout the AC current loops, like the output side of a boost converter or the gate driver path and return.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Intellectual 330 points

    Brigitte, some things are unclear to me from the datasheet:

    1) CT sets the oscillator time. Should the current flowing through the inductor rise and fall within that period?

    2) Should the current thru the inductor start at 0A and end at 0A within a cycle?

    3) should there be spacing between the teeth in the inductor current sawtooth?

    I'm seeing current sawtooths that last for multiple cycles. I'm also seeing runt sawtooths and sawtooths within sawtooths. I'm wondering if I have a control loop instability issue. The datasheet for this part doesn't mention control loop stability, but other switch controller datasheets do.

    Some things I have tried:

    1) capacitor across lower resistor of feedback

    2) more Cout capacitance

    3) ceramics for high-frequency noise and such at various points in the circuit.

    4) larger inductor

    5) Using a fixed DC load instead of a fan load. Circuit still seems to have stability issues, looking at the inductor current waveform. The output voltage seems to regulate correctly at 16V though.

    6) slowing the switching frequency, increasing the switching frequency.

    Thx in advance for your help!

  • 0 in reply to BobL
    TI__Guru 55765 points

    Hello BobL,

    1. CT sets the maximum rise time. The current through the inductor is commanded on by the output voltage (if it is too low) and stopped when the oscillator has reached the upper threshold, see figure 8 in the datasheet.

    2. Yes, if the cycle is started because the output voltage reached its threshold.

    3. You need to observe the output voltage as well as the inductor current, see figure 8 in the datasheet. Please send measurements if we should review in detail.

    The controller is somewhat specific because of the turn on and turn off behavior. If you can send some observations when you changed the different components, this would be helpful.

    Best regards,
    Brigitte