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.

DC-Link Capacitor Selection for PMSM motor control

Other Parts Discussed in Thread: DRV8301, INA240

Hi,

I would like to know how to decide the DC-link capacitor for PMSM motor control. I found this link on one of the TI forums for a similar topic:

www.ecicaps.com/wp-content/uploads/IEMDC_2009_11310_Final_Rev_4.pdf

It is a very good paper in explaining how ripple voltage, ripple current and type of capacitor influences the capacitor selection.

But I have a few doubts:

1. While using VL = Ldi/dt, the paper does not take into account the back emf when substituting for VL. How will it influence the calculations?

2. Another case is regen current. If the battery cuts-off due to some fault, the capacitors should be able to take the regen current during flux weakening and withstand the voltage. How do we account for that? 

I know its a tricky thing to select a DC-link cap. But any suggestions or references to papers/thesis will be helpful. I tried finding more on this topic in literature but did not get much information on it.

  • Hello Rushi,

     Please refer to this E2E post: https://e2e.ti.com/support/motor-drivers/f/38/t/619948   .  Do you have a design already, and are you using TI devices in this design?  If so, could you please share that design, just the portion relevant to the question at hand. 

    Regards,

    ~Leonard 

  • Hi Leonard,

    I have already gone through the e2e post shared by you and other e2e posts on similar questions.

    We do not have a schematic ready yet but we are in processing of doing so. Yes, we will be using TI parts (F280049 controller, INA240 for current sense, and may be some more - mostly trying to upgrade the DRV8301 kit).

    At this time, I just want to come to an optimum value of DC-link Cap (to know which caps to use, do I need a combination of different types, and how much board space it will take).

    I can share you the part of the design once it is done for your suggestions.

    It would be great if you can help understand how to come to a value of DC-cap.

    The formula mentioned in the refernced e2e post, gives me DC-link value of only 260uF. I think that is too less compared to other controllers that I have seen of the same power rating. I do understand choosing electrolytic caps affect the capacitor value selection based on ripple current.

    But I believe there are other factors like back emf and regen which have not been taken into account in the link shared by me.

    I want to be sure that I have taken into account all parameters before I finalize the DC-link Cap value.

    Any help from your side would be appreciated.
  • Typically, the load current decides the dc voltage ripple and dc cap ripple current. Calculations based on dc voltage ripple and load current will provide the minimum capacitance (Cmin) at the dc link. But the ripple current rating of the capacitor will decide the final value as it may decide the N number of caps of value C to be put in parallel, such that N.C >= Cmin

    Hope it helps..

  • Oh, btw, regen or bemf is not a typical concern on the cap, except in the case of a PM motor. In the FW mode, when the motor happen to coast, there could be a higher back voltage on the dc cap that can be risky. You may have to consider sufficient margins if you see such a possibility.
  • Thanks Ramesh for your suggestions.

    I was looking for more of a mathematical explanation on how to decide my DC-link capacitor value.

    If you have any doc that can help me, let me know.
  • Nope, I don't have one. You can refer engineering text books for the capacitor design based on load current and ripple voltage. Googling may help too.