SLVAF66 June   2021 DRV3255-Q1 , DRV8300 , DRV8301 , DRV8302 , DRV8303 , DRV8304 , DRV8305 , DRV8305-Q1 , DRV8306 , DRV8307 , DRV8308 , DRV8320 , DRV8320R , DRV8323 , DRV8323R , DRV8340-Q1 , DRV8343-Q1 , DRV8350 , DRV8350F , DRV8350R , DRV8353 , DRV8353F , DRV8353R

 

  1. 1 Introduction to High-Power Motor Applications
    1. 1.1 Effects of a Poorly-Designed High-Power Motor Driver System
    2. 1.2 Example of the High-Power Design Process
  2. 2 Examining a High-Power Motor Drive System at a High Level
    1. 2.1 Anatomy of the Motor Drive Power Stage and How to Troubleshoot
    2. 2.2 Troubleshooting a High-Power System
  3. 3 High-Power Design Through MOSFETs and MOSFET Gate Current (IDRIVE)
    1. 3.1 MOSFET Gate Current
      1. 3.1.1 How Gate Current Causes Damage
      2. 3.1.2 Gate Resistors and Smart Gate Drive Technology
        1. 3.1.2.1 Gate Resistors
        2. 3.1.2.2 Smart Gate Drive and Internally-Controlled Sink and Source Gate Currents
        3. 3.1.2.3 Summary for Gate Resistors and Smart Gate Drive Technology
      3. 3.1.3 Example Gate Current Calculation for a Given FET
  4. 4 High-Power Design Through External Components
    1. 4.1 Bulk and Decoupling Capacitors
      1. 4.1.1 Note on Capacitor Voltage Ratings
    2. 4.2 RC Snubber Circuits
    3. 4.3 High-Side Drain to Low-Side Source Capacitor
    4. 4.4 Gate-to-GND Diodes
  5. 5 High-Power Design Through a Parallel MOSFET Power Stage
  6. 6 High-Power Design Through Protection
    1. 6.1 VDS and VGS Monitoring
      1. 6.1.1 Turning Off the FETs During an Overcurrent, Shoot-Through, or FET Shorting Event
    2. 6.2 Passive Gate-to-Source Pulldown Resistors
    3. 6.3 Power Supply Reverse Polarity or Power Supply Cutoff Protection
  7. 7 High-Power Design Through Motor Control Methods
    1. 7.1 Brake versus Coast
      1. 7.1.1 Algorithm-Based Solutions
      2. 7.1.2 External Circuit Solutions
      3. 7.1.3 Summary of Brake versus Coast
  8. 8 High-Power Design Through Layout
    1. 8.1 What is a Kelvin Connection?
    2. 8.2 General Layout Advice
  9. 9 Conclusion
  10. 10Acknowledgments

4.1 Bulk and Decoupling Capacitors

Figure 4-1 Bulk Capacitor Example

Bulk and decoupling capacitors have the primary purpose of providing instantaneous charge into a system so the main power supply does not have to. To be more specific, current ripple within the supply, and voltage spikes from parasitic inductance created from wire and traces are the result of insufficient charge in the supply. Since the power supply is be physically located far from the motor drive circuit, there is quite a lot of inductance in the path from the supply to the MOSFETs.

Small-valued capacitors can be emptied and filled with charge relatively quickly, where larger valued capacitors can store a lot of energy, but do not react as quickly. This is why most data sheets show recommended components with large and small capacitors placed in parallel on power supplies. In the context of the power stage, millifarads or hundreds of microfarads electrolytic or ceramic capacitors are used in combination with singular to tens of microfarads ceramic capacitors.

In addition, there are times where the motor can act as a generator where the bulk and decoupling capacitors store energy from the motor to prevent rising voltage on the drain of the high-side FET, or VDRAIN as shown in Section 4.1.

In summary:

  • Because low value capacitors can provide some charge quickly and high value capacitors provide a lot of charge over time, it helps to reduce voltage ringing and voltage spikes in the system
  • It is highly recommended to always use them. A few 100-µF to 330-µF capacitors in parallel with a few 1 µF to 2.2 µF is a good starting point with some more footprints that can be replaced later.
  • A common rule of thumb is 2 µF/W; however, actual system results vary significantly

Truthfully, this is nebulous advice. This does not describe the process of estimating parasitics for a given layout and simulating their effects through SPICE to get an optimal bulk capacitor value. As a result, there are no equations or hard math. However, we want to highlight this as practical advice. There is a lot less effort in the design process to test the system in reality or rely on past system knowledge in combination with the data sheet. If the performance is not good enough, then designers add more capacitors or change the bill of materials so an existing capacitor is replaced with a capacitor of a different value to fix the problem.

In summary, planning to implement a general rule to get a baseline capacitor value, but testing a system in reality, might result in good performance with no other changes needed, or bad performance where an experimental and iterative process fixes the performance issues.