2
\$\begingroup\$

I am designing a board, which has a high current (~2A) buck-topology switch mode power supply on it, which uses an unshielded SDR0805 inductor. About 4cm from it, there is a magnetometer (combined mag/acc LSM303DLH.) Would problems be encountered with this setup? Would I need to shield the inductor?

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Have you read the appnote? It talks a lot about proximity to constant and transient ("hard" and "soft") EM sources and calibration. \$\endgroup\$ Commented May 9, 2011 at 18:04

1 Answer 1

3
\$\begingroup\$

The datasheet says this in the application hints.

High current wiring effects

High current in wiring and printed circuit traces can be the cause of errors in magnetic field measurements for compassing. Conducto-generated magnetic fields add to earth’s magnetic field, creating errors in compass heading computation. Keep currents that are higher than 10 mA a few millimeters further away from the sensor IC.

2 A is way more than 10 mA, and an inductor is a obviously going to create a stronger magnetic field than a straight wire. A good guesstimate is that you're going to have interference problems even if you shield the inductor.

\$\endgroup\$
5
  • \$\begingroup\$ Hmm, this is worrying. Now, could I apply the inverse square law in this situation? If 10mA is okay at, say, 5mm, then 640mA should be okay at 4cm... but that's still way less than 2A. Ugh. \$\endgroup\$
    – Thomas O
    Commented May 9, 2011 at 18:06
  • \$\begingroup\$ If you put a magnetic shield cage around the SMPS, you might be able to further reduce the interfering magnetic effects. \$\endgroup\$
    – Toybuilder
    Commented May 9, 2011 at 18:09
  • 3
    \$\begingroup\$ Inverse square law should work in this case, however keep in mind that the 10 mA figure is for DC current through a single wire/trace. The inductor will both produce a stronger magnetic field and be a nastier source of interference, because of the AC ripple from the coil when it's being charged and discharged at a high frequency. \$\endgroup\$
    – nitro2k01
    Commented May 9, 2011 at 18:23
  • \$\begingroup\$ Looks like shielding is the way to go. What are my options for shielding a power supply or the magnetometer? \$\endgroup\$
    – Thomas O
    Commented May 9, 2011 at 18:53
  • \$\begingroup\$ Shielding will help with the e-field but not with the magnetic field. If your are having magnetic coupling problems increasing the distance or changing orientation is your best bet. The magnetic field size will be bigger with larger physical inductors and higher currents. The inductor is a antenna so it will be directional and you are worried about the near field where the magnetic field dominates. \$\endgroup\$
    – Rex Logan
    Commented May 9, 2011 at 19:40

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.