1
\$\begingroup\$

I'm courious how strong is magnetic field around step-down conveter? My plan is to use one LM2596 @ 12V(1A at maximum load) and one LM2596 @ 5V for STM32. I Know that coils generate MF around itself and in, lets say normal use I don't have any problem with magnetic field. But in this project I want to use MPU9250 for compass. I'm pretty sure magnetic field interferes with magnetometer, am I? Is magnetic field strong enough to disrupt MPU9250? Is there any way I can protect MPU9250 from magnetic field? Anti magnetic field shield? Enough space from step-downs and MPU9250? I hope that exists way to use step-downs and MPU9250 in same housing. I can separate them cca 15-20cm max. Will it be enough?

Simple: Is it possible to use magnetometer(MPU9250 module) near two step-down converters?

|improve this question
\$\endgroup\$
  • 3
    \$\begingroup\$ What is "MF"? I assume you don't mean the thing that Samuel L Jackson's movie characters are always calling people. \$\endgroup\$ – The Photon Mar 4 at 18:08
  • 1
    \$\begingroup\$ @ThePhoton Don't get it? MF is magnetic field \$\endgroup\$ – SilvioCro Mar 4 at 18:09
  • 1
    \$\begingroup\$ @SilvioCro Don't get it? TEBIMF is The Earth's Built-In Magnetic Field. \$\endgroup\$ – Elliot Alderson Mar 4 at 18:38
  • 1
    \$\begingroup\$ Your STM32 microcontroller may draw varying DC current, depending on what code is running. You may have to mount MPU9250 far from it (and all other current-carrying wires/traces), when your magnetic sensor is capable of resolving microTeslas. \$\endgroup\$ – glen_geek Mar 4 at 18:49
  • 3
    \$\begingroup\$ It's possible to buy a "shielded" inductor surrounded by a ferrite material to contain the external magnetic field. It increases the size of the part, and possibly the cost. Whether the shielding is adequate for your application I don't know. \$\endgroup\$ – The Photon Mar 4 at 19:17
5
\$\begingroup\$

The magnetic field near circuitry carrying about 1 amp could be a problem 20cm away, but for a compass application you can get there if a few steps are taken in the design. I would wind toroids for the buck inductors (distributed air gap) and keep them well out of saturation to keep the field contained in the core, and would lean toward continuous mode to keep discontinuities at a minimum. Make sure your current paths and the corresponding return path follow close and parallel traces; keep the separation at a minimum so that you don't create large current loops. Use twisted wire from your power source and out to any load. If you take these precautions, you can easily keep the field associated with the circuit below 100 nT 20 cm away - a small but not insignificant error contribution since the earth's field is around 50 uT, about 2%. You might be able to do better than this.

|improve this answer
\$\endgroup\$

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