• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 98 subscribers
  • Views 392 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.

Can TPS65070 L3 CVDD inductor be very close to CPU?

Paul Singh
Expert 2750 points
Other Parts Discussed in Thread: TPS65070

This is also posted in the PMU forum.

Hi,

When using TPS65070 to power an L138 board, can the VDCDC3/L3 (supply CVDD 1.2V) be very close to the CPU?

In TPS65070 datasheet SLVS950C pin 31 L3 (“pin functions”) is the switch of DCDC3, and it would connect to an inductor with the other end of the inductor providing power directly to the CVDD network. We follow the datasheet “Table 4. Tested Inductors” recommendation and use a LPS3015 2.2uH inductor.

If the board layout is compact, could this inductor’s output end be very close to the CPU, be as close as 1mm distance? Inductor basically suppress high-frequency current so its output should be flat, therefore we don’t think (and haven’t observe on our current board) that any significant voltage ripple will present. However, the inductor dissipates electronic energy by propagating EM wave to outside space, and when this inductor is only 1mm far from the chip supplied by the CVDD voltage, is it possible for such EM wave to be strong enough to affect L138 operation?

  

 

Paul

  • 0
    Expert 1950 points

    Hello,

    Another free advice:

    Coilcraft's LPSxxxx are shielded inductors, with vertical axis (magnetic field normal to board). To my opinion, for digital designs, conducted EMI problems raise far more frequently (with bad PCB routing or decoupling capacitors quality/number) than induced ones.

    Check for the common misusage : such inductors should be mounted with the marking (point) on the chopper side (ie the inner ring wire to the TPS6507x). This permits to reduce EMI/RFI by making the outer rings (CVDD side) shield the inner ones.

    Jakez

  • 0 in reply to Jakez
    Expert 2750 points

    Jakez is again so awesome‼

    Just asked an FAE of coilcraft who confirmed that the point mark side is where the winding starts, and the other side is where the winding ends. He agreed that in principle the better practice is to solder the inner side with the noisier voltage and the outside to the powered point. But he also added that there is ferrite elements/composites in the casing so the shielding is very good and practically the need on the orientation is not that great.

  • 0 in reply to Paul Singh
    Expert 1950 points

    Hello Paul,

    Thanks, soft-soaping is often good politics... So even more free advice:

    Inductor shielding refers to magnetic shielding (with magnetic material as ferrite, which permits to reduce coupling between near inductors and then between the 3 supplies regulation loops in the TPSxxxxx case), while the outer rings shield deals more with electric field shielding (the chopper-side rings are 'in' a copper box made by the outer rings).

    Jakez

  • 0 in reply to Jakez
    Expert 2750 points

    Jakez,

    I was not quite sure does the outer ring work similar to a Faraday cage in terms of shielding the inner ring? The models seems very different for an isolated cage and winding wires flowing continuous current.

    Paul

  • 0 in reply to Paul Singh
    Expert 1950 points

    Hello,

    Free advices are sometime wrong...

    More advised opinion can be found here (and probably some other elsewhere):

    http://www.power-mag.com/pdf/feature_pdf/1319729361_Murata611_Layout_1.pdf

    Jakez

  • 0 in reply to Jakez
    Expert 2750 points

    Jakez,

    Just read through the note. We have been quite careful with the placement and have avoided the extreme close placement, but still need to test after assembly.

    Paul