Post Go back to editing

LT8362 coupled inductor

Dear Sir/Madam,

Does the phase polarity of the coupled inductor matter for the LT8362 configured as inverting converter?

https://www.analog.com/media/en/technical-documentation/data-sheets/8362fa.pdf

page 20

https://www.analog.com/media/en/technical-documentation/eval-board-schematic/710-DC2517A_REV04_PCA_SCHEMATIC.PDF

page 1

Regards Joe

Parents
  • May I add that for reasons of electromagnetic emission, the version with the (green) dots connected to the SW node is the one with lowest emission.

    The reason being that the dot is - by convention - the end where the wire starts to be rolled on the core and thus is below the rest of the windings. So the dot-end is being shielded by the outer windings. The SW node is the one with the largest voltage excursions so is the primary source of radiated electric fields. The non-dotted ends of the wires are kept "quiet" in this circuit due to the capacitors to gnd.

    So rule of thumb: Wire the dot to the node with voltage noise.

  •  

    While pondering what you said, I have taken note of page 24 in the LT8362 Datasheet which shows a 5V SEPIC converter design. I thought the datasheet designs were the gold standard for what we designers should pattern our own designs after; and yet, the dot on L1 leads to Vin, and the dot on L2 leads to GND. Based on what you are saying about lower EMI, one should point the dot on L1 to SW, and point the dot on L2 to the point between C6 & D1.  If you agree, would it then be reasonable to assume the current design given on page 22 to be somewhat of an oversight in terms of EMI?  Or is there a good reason it was designed that way?

    LT8362 SEPIC Datasheet Design

  • Unfortunately, it is not always the case that the dot marks the inner beginning of a coil. It really depends on the manufacturer.
    I'm afraid as good as the datasheets are, they are not a good substitute for the thinking process. Some skills are still required to use these switching regulators correctly. To that end, yes they could have put the dots differently in order to avoid this discussion.

    Best Regards,
    Mathias

  • I appreciate those thoughts very much. Thank you.

    For the sake of offering people reading this discussion even greater clarity on what I mentioned above, the example SEPIC design on page 24 of the LT8362 datasheet recommends use of a Wurth 744878002 coupled inductor, and below is how the Wurth 744878002 datasheet shows its dot location. Wurth's use of the word "Start" next to the dot informs me as a designer that is where the wire starts to be rolled on the core.  And so, if we follow the advice of @larsen above, we would need to ensure pins 1 & 4 on that coupled inductor are THE OPPOSITE of what the LT8362 datasheet design example shows, putting the dot instead toward SW, which is the node with voltage noise.
    Wurth 744878002 dot

Reply
  • I appreciate those thoughts very much. Thank you.

    For the sake of offering people reading this discussion even greater clarity on what I mentioned above, the example SEPIC design on page 24 of the LT8362 datasheet recommends use of a Wurth 744878002 coupled inductor, and below is how the Wurth 744878002 datasheet shows its dot location. Wurth's use of the word "Start" next to the dot informs me as a designer that is where the wire starts to be rolled on the core.  And so, if we follow the advice of @larsen above, we would need to ensure pins 1 & 4 on that coupled inductor are THE OPPOSITE of what the LT8362 datasheet design example shows, putting the dot instead toward SW, which is the node with voltage noise.
    Wurth 744878002 dot

Children
No Data