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.

LMG1020: Feed-Through decoupling capacitor connection

Part Number: LMG1020

Hello,

I'm carefully reading the decoupling recommendations of the LMG1020.

In the datasheet :

Figure 16 presents a layout of LMG1020 with a 0.1 µF feed-through capacitor (C1) and a larger 1uF capacitor (C3) for decoupling. In this design, the feed-through capacitor C1 is placed in a shunt-through manner for lower noise decoupling, and C3 is placed next to C1.

After reading some documentation about the "shunt-through manner" to use a feed-through capacitors (https://product.tdk.com/en/techlibrary/solutionguide/3tf04.html), is appears to be a good idea. But it's just impossible to do that and place both capacitors terminal very close to the LMG1020.

The Figure 16 is effectively showing a feed-through capacitor connected as mentioned, but it looks like the capacitor is inverted : the central pad (commonly used as GND) is the Vdd, and the 2 externals pads are GND. I have never seen this usage of this kind of capacitor. It solves the layout challenge but is it really a good idea?

The devkit use a 0603, 470nF feed-through capacitor rated to 4A, used in the classical manner.

Any light appreciated :)

Thanks

  • Hello Nicolas,

    The LMG1020 draws a lot of current from it's bypass capacitors. Close placement of the capacitors is critical. I'd recommend following the EVM layout as an example compared to the datasheet.

    As for the datasheet layout, I'd think the feed through capacitor works as two capacitors in parallel in this case. This probably does not work as well as normal. Figure 15 should work as normal.

    Best Regards,
    Ethan Galloway