How to check whether a component can work in strong magnetic fields?

Question

I want to design my PCB to operate nicely even when we place it next to a neodymium magnet. How to check whether my component can work in such condition without shielding?

Edit: I haven't face any problem with my circuit when I place it next to a magnet, but people will start questioning about the stability and I have no idea how to proof it. The main component are NAND Flash memory, microcontroller, MEMS accelerometer, battery, wireless transceiver on the board.

Answer 1

I want to design my PCB to operate nicely even when we place it next to a neodymium magnet. How to check whether my component can work in such condition without shielding?

You can expect potential problems if a device contains a moving conductor, "magnetic material" or is designed as a magnetic or electric or electromagnetic field sensitive or field sensing device.

Magnetic field decreases with the inverse cube of the distance from the centre of the North-South dipole so it gets rather small rather quickly in most cases. (Field from each pole decreases as inverse square (not many people realise this)and the vector sum of the dipole pair approximates to inverse cubed at many magnet lengths away from the dipole center).

A modern high strength rare earth magnet (usually Nd2Fe14B) will produce around 1 Tesla out to half of one magnet dipole (N-S) length from the pole face. ie long (or deep) magnet = deep external field. You can pretend that means it will be about 1/8th T at 1.5 magnet lengths and 1/27 Tesla at 2.5 magnet lengths etc.

---

A MEMS accelerometer (probably) contains moving conductors and so may have some issues. You'd expect their data sheet to say so if this was important.

Any magnetic cored device that isn't shielded, and some that are shielded, could potentially be affected. For example a coil with a ferrite slug or one wound on a ferrite or iron core bobbin would have the AC BH curve moved by a DC offset value by the magnet's field and depending on magnet strength and proximity it could push a design into saturation or deeper into saturation than it otherwise would be.

A magnetic style loudspeaker or earpiece could be affected.

A Hall cell, GMR sensor, AMR sensor, and other explicitly magnetic field sensitive device 'could have fun'.

Any common mechanical meter movement could be affected (moving coil, moving iron, air core, ...)

Any electric motor (brushless DC, brushed, induction, stepper, head actuator, ...), relay or actuator using magnetic fields could be affected

Maybe:

FRAM memory, core memory

Long bow:

Light Saber, Dilithium energy cell, ...

---

Should be OK:

As long as no specifically magnetic sensitive components -

ICs, analog and digital, memory, RF (note inductor cores), .. Battery
Passives - resistor, capacitor, ...
Inductor, air cored.

Answer 2

If you're interested in how to prove it, I guess trying your typical situation and writing a piece of documentation should be o.k. Whenever I'm outside some typical or standardized situation, I try to think of a reasonable set-up with some security factor calculated in, maybe 1.5 or 2. For instance, if your application has a magnet on one side of your board, you could try to build a ferromagnetic (steel) yoke directing the field towards the components you suspect to be sensitive, or use two magnets on both sides of the board. Also, you could ask a test lab if they can check for really strong low-frequency fields.

With medical coils like these, you can create flux densities up to 5 T: Source

For the more usual stuff, there's a test set-up that's part of a standard EMI conformity test:

For low-frequency fields (like you seem to be interested in), you put your device in the middle of a large frame with a loop (magnetic coil) around it, and you put quite a lot of current through the loop, creating a strong magnetic field.

A typical test set-up looks like this: Source

This set-up looks actually quite easy and you could home-brew one - the difficult and expensive part would be the calibration. I've even been to big EMC test labs that used self-made coils for this test.

Just for the fun of it - a practical, everyday source of interference with fields about as strong as the ones tested with the device in the above picture typically looks like this: Source

or this: Source

... or like the deflection yoke in a CRT monitor: Source

Then again, with electromagnetic fields, transmitters and receivers are dual elements, so the TV set is also a receiver for external low-frequency fields - ask the guy living in the house in the picture above who watches the eight-o'clock news on a CRT TV - the picture with the red engine, not the one with ICE train; the quality of his TV's picture geometry might not be exactly stable.