• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 163 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LMG2100R044: Question on DC-Link Capacitor Sizing and Regenerative Energy Handling in TIDA-010936

Shimazu Toyohiro
Prodigy 10 points

Part Number: LMG2100R044
Other Parts Discussed in Thread: TIDA-010936

Dear TI Engineer,

We are reviewing the reference design TIDA-010936 and would like to better understand the design assumptions regarding the DC-link (Vbus) capacitor, as well as how regenerative energy is considered.

1. DC-link capacitor sizing (reference-design / hardware perspective)

In TIDA-010936, the DC-link capacitor on Vbus is implemented as six 10 µF ceramic capacitors in parallel (total nominal 60 µF).

Could you please clarify the design rationale behind selecting this value?

Specifically, from a hardware and reference-design perspective, was this capacitance determined based on factors such as:

allowable DC-link ripple voltage,switching frequency, or motor phase current, or is it mainly intended as a representative value for evaluation and demonstration purposes, rather than a system-level energy buffering requirement?

2. Regenerative energy handling assumptions

First, from the reference-design / hardware perspective, how is the energy generated during reverse torque or regenerative operation expected to be handled in TIDA-010936 itself?

In addition, from a system-level perspective, for practical motor drivers used in humanoid or robotic joint applications, could you comment on how regenerative energy handling is generally considered today?

We are particularly interested in current industry trends or best practices, such as:

battery backfeeding, use of brake resistors or active clamp circuits, or coordination between DC-link capacitance and controlled regeneration at the system level.

We would like to better understand how the assumptions used in TIDA-010936 differ from those typically adopted in real robotic systems.

Thank you very much for your support.

Best regards,

  • 0
    TI__Prodigy 120 points

    Hello,

    1) DC-link capacitor sizing (reference-design / hardware perspective) - Yes, as you have mentioned, we choose the capacitor sizing based on the

    a. Allowable DC voltage ripple as larger caps help in achieving smaller voltage ripples.

    b. Switching frequency because as the switching time of the FET becomes shorter with increasing frequency, the amount of charge required by the capacitor becomes smaller, so using a higher PWM switching frequency can reduce the required bus capacitance value.

    c. Cap derating as we need to make sure that at the operating voltage levels, the derated value of cap is sufficient enough to limit voltage ripples to permissible range.

    d. Cost and area constraint because as the capacitance value increases, it's size and cost also increase.

    We choose a tradeoff between the above factors to choose the optimum value of capacitance at DC bus, which is chosen 60uF here.

    As for placing 6 10uF caps instead of a single large 60uF cap, the objective is to lower the effective ESR (equivalent series resistance) of the capacitors as the ESRs of the 6 caps are in parallel and hence effective ESR is reduced by 6 times.

    The rationale for using ceramic caps instead of electrolytic cap is because ceramic caps come in smaller sizes, are more stable and durable, but this comes at a cost of limited capacitance value. However, tests reveal that at higher frequencies, the DC bus voltage ripples are similar for same value of ceramic and electrolytic caps. Hence, we opt for the ceramic cap.

    2) Regenerative energy handling assumptions - Since our reference design is for 48V input, usually we use battery to supply the input. In this case, when there is reverse torque or regenerative operation, the half bridges inverters, being bidirectional, act as boost converters to feed power back to the battery. So, we don't need any extra regenerative energy handling as there is battery back feeding already.

    However, is one uses AC/DC power supply for input, we will need extra components for regenerative energy handling - implementing a bidirectional DC-DC converter to save this energy into a specific battery is a possible solution to achieve high system efficiency. If efficiency is not a major concern, then for a simpler design, we can use a simple discharge resistor to safely discharge the DC bus capacitors to avoid any threats to the system.

    Regards,

    Shruti

  • +1 in reply to Shruti Mathpal
    Prodigy 10 points

    Hello Shruti,

    Thanks for the detailed explanation.
    The DC-link capacitor sizing and regenerative energy handling assumptions are clear and make sense to us.

    This helps align our system assumptions with the reference design.
    Thank you.

    Best regards,