SBAA666 February   2025 AMC0106M05 , AMC0106M25

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
  5. 2Design Challenges
  6. 3Design Approach
    1. 3.1 AMC0106Mxx Functionally Isolated Modulators
    2. 3.2 Circuit Design and Layout
    3. 3.3 Sinc3 Filter Design
  7. 4Test and Validation
    1. 4.1 Test Setup
    2. 4.2 Digital Interface
    3. 4.3 DC Accuracy, Noise, and Effective Number of Bits
    4. 4.4 PWM Rejection
      1. 4.4.1 DC Phase Current Measurement Over One PWM Cycle
      2. 4.4.2 AC Phase Current Measurement at 100kHz PWM
    5. 4.5 Bootstrap Supply Validation and AVDD Ripple Rejection Tests
      1. 4.5.1 LMG2100R044 Bootstrap Supply With Low Voltage-Ripple
      2. 4.5.2 Discrete Bootstrap Supply With High Voltage-Ripple
  8. 5Summary
  9. 6References

4.4.1 DC Phase Current Measurement Over One PWM Cycle

For this test, the AMC0106M05 high-side supply (AVDD) was powered from the LMG2100R044 GaN-FET bootstrap supply which is the default configuration shown in Figure 3-4.

Figure 4-5 and Figure 4-6 show the PWM-switched phase voltage W at 3A phase current, measured with a current probe. The phase-W voltage equals the common mode voltage of the AMC0106M05 analog ground (AGND) versus the digital ground (DGND). The higher slew rate of 13V/ns during the transition from 0V to 48V is when the high-side GaN-FET is hard-switching, the lower slew rate of 2.7V/ns is during the transition from 48V to 0V, when the low-side GaN-FET is soft-switching (zero voltage switching). For more information on the switching characteristics of the GaN-FET, refer to the LMG2100R044 data sheet.

Phase W Voltage Rising Edge at 3A CurrentFigure 4-5 Phase W Voltage Rising Edge at 3A Current
Phase W Voltage Falling Edge at 3A CurrentFigure 4-6 Phase W Voltage Falling Edge at 3A Current

Figure 4-7 shows a scope plot of the phase-W voltage, the phase current W (measured with a current probe) and the supply voltage AVDD for the AMC0106M05. The average phase-W current is 3A and the peak-to-peak current ripple is 0.4APP, a function of the servo motor's stator time constant.

Phase W Voltage and Current Over One PWM Cycle Figure 4-7 Phase W Voltage and Current Over One PWM Cycle

For the first test, a static 3-space vector PWM with a constant phase (330 degree) and voltage magnitude was impressed. The voltage magnitude was configured to force constant motor currents. For example, phase current W was set to IW = 0A or 3A, while phase U and V currents were IU = IV = –0.5 * IW.

To study the impact of PWM switching on the measurement accuracy of the AMC0106M05, the start of conversion of the sinc3 filter was increased by 60ns for each new PWM period. The PWM frequency was 10kHz, the corresponding PWM period is 100µs. The effective sampling point of the first conversion started 14µs after the start of the first PWM cycle. For the seconds PWM cycle, the effective sampling point was 14.06µs. For the 600th PWM period, the effective sampling point was 50µs (center aligned to the PWM). And the last sampling point, on the 1200th PWM cycle, was set to 86µs. In a last step, the 1200 results were overlayed and plotted as a single PWM cycle.

The results for 3A phase current are shown in Figure 4-8. Results for 0A phase current are shown in Figure 4-9. In both cases the OSR was 32. Phase current is plotted on the left axis, phase voltage on the right axis.

AMC0106M05 Phase Current IW with OSR=32 at 3A Over one PWM CycleFigure 4-8 AMC0106M05 Phase Current IW with OSR=32 at 3A Over one PWM Cycle
AMC0106M05 Phase Current IW with OSR=32 at 0A Over one PWM cycleFigure 4-9 AMC0106M05 Phase Current IW with OSR=32 at 0A Over one PWM cycle

At zero phase current there is no visible effect on the AMC0106M05 output signal during PWM switching of the phase. Neither the falling edge of the PWM with a low slew rate of 2.7V/ns, nor at the rising edge with a high slew rate of 13V/ns disturb the phase current measurement. This confirms the high common mode transient immunity (CMTI) of the AMC0106M05.

The AMC0106M05 phase current measurement was compared to a measurement with a current probe to validate the result. Both measurements are overlayed in Figure 4-10. Note that transient noise is seen with the current probe measurement (black trace), but not with the AMC0106M05 measurement (red trace).

AMC0106M05 Phase Current IW with OSR=32 at 3A versus Current Probe Measurement Over one PWM Cycle Figure 4-10 AMC0106M05 Phase Current IW with OSR=32 at 3A versus Current Probe Measurement Over one PWM Cycle

The test was repeated with a 64-times oversampling ratio. As expected, the measurement noise with 64-times oversampling is significant lower. Even with this high-resolution measurement there is no effect visible of the PWM switching on the AMC0106M05 measurement accuracy.

AMC0106M05 Phase Current IW With OSR 64 at 3A Over one PWM CycleFigure 4-11 AMC0106M05 Phase Current IW With OSR 64 at 3A Over one PWM Cycle
AMC0106M05 Phase Current IW with OSR 64 at 0A Over one PWM CycleFigure 4-12 AMC0106M05 Phase Current IW with OSR 64 at 0A Over one PWM Cycle