1
\$\begingroup\$

I have three buck converters on my PCB.

  1. 12V to 0.82V @ 25A -FP1308-R21-R, Fsw = 650 kHz, TPS548D21

  2. 12V to 1V @ 8A - IHLP2525CZERR47M01, Fsw = 1 MHz, TPS548A20

  3. 12V to 5V @ 3A - Inductor is XAL5030-332ME, 400 kHz (can be increased), TPS568215

Due to thermal issues and heatsink optimization, we need place all three of these inductors very close (3-5 mm) to each other.

All the inductors are shielded.

What will the inductor shielding effectiveness be? There is no specification on distance vs field.

Can we place all three inductors close to each other? If we can place them close together, what considerations do I need to take care of?

Added a PIC of this.enter image description here

\$\endgroup\$

1 Answer 1

4
\$\begingroup\$

What will the inductor shielding effectiveness be? There is no specification on distance vs field.

You can do a simple test - use 100 kHz (from a signal generator) applied to one inductor and see how much induced voltage is in another inductor when placed close to the driven inductor. I wouldn't want more than (say) 2% coupling (2% voltage induction). Rotate orientation to find least coupling and then design your PCB.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Note for the asker: Lowest coupling will probably be when the inductor coils are at right angles to each other. The magnetic shielding might affect this, though. \$\endgroup\$
    – Hearth
    Dec 12, 2019 at 15:02
  • \$\begingroup\$ @Andy aka: we are in placement phase right now. I can' t do this. Can this be simulated? whether we will get inductor models which have this much details. Inductor vendors doesn't have an answer they say their inductors are best in that size. i am dumb, from 2% coupling can i use L*di/dt to calcualte coupled voltage approx. i know which inductor carries how much current(traingular pk-pk). \$\endgroup\$
    – user19579
    Dec 13, 2019 at 3:56
  • \$\begingroup\$ @Andy aka & Hearth : added approx placement picture. we have a slot in the heatsink to accommodate these component. In that i am distributing inductors so, that they will least interact at the same time 3 buck loops are proper. \$\endgroup\$
    – user19579
    Dec 13, 2019 at 4:06
  • \$\begingroup\$ The placement picture looks OK in terms of low coupling. I've never found the info to do this in simulation, it's always been a real test. 2% coupling means that if 1 volt p-p were on one inductor, it would induce a voltage of 20 mV p-p on the other. You can't do this theoretically because you don't know the mutual (shared or coupled) inductance. \$\endgroup\$
    – Andy aka
    Dec 13, 2019 at 8:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.