Author Topic: SMD inductors - liability to couple with external LF magnetic fields?  (Read 894 times)

0 Members and 1 Guest are viewing this topic.

Online splinTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: gb
Datasheets for SMD inductors don't seem to show LF performance below 100kHz or more, so how susceptible are they to coupling in 50/100Hz magnetic fields from nearby mains or rectified mains wiring?

Is it reasonable to assume that the core permeability, and thus inductance, don't change significantly at low frequencies for typical ferrite bead inductors below 1uH?

[EDIT] I'm referring to inductors with multi-turn winding, not the ferrite beads threaded onto a wire.
« Last Edit: October 19, 2019, 01:06:09 am by splin »
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: SMD inductors - liability to couple with external LF magnetic fields?
« Reply #1 on: October 19, 2019, 05:52:54 am »
So which is it, ferrite bead, inductor and multilayer or wound? :-//

All different things -- ferrite bead chips are always multilayer and saturate easily; ferrite beads (various styles, wound) saturate somewhat less easily but AFAIK always at some fraction of rated current.  Inductors come in multilayer or wound types, the former evidently use a low-mu or distributed gap material while the latter use a formed shape with gap built in.

- The enclosed types (multilayer, composite, molded, shielded) have good shielding, so that the susceptibility isn't much greater than an equivalent length of wire.
- "Semi shielded" types usually have some variation of a bobbin core, so the external field coupling drops off quickly with distance (relative to the height of the bobbin).  It's a dipole, but it's small?
- Unshielded types are usually rod or solenoid variants and are ordinary dipoles along the axis.
- Toroids are okay, usually having one equivalent loop turn (and corresponding wide dipole) due to the winding going around the toroid itself.  This is negligible if there are a lot of turns on the core, and can be canceled out if the winding doubles back on itself.  Few-turns toroids have further inhomogeneity due to the relatively large spaces between turns (each loop looks like a dipole, so the whole thing looks like a (n+1)pole..).

HTH :-+
Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Online splinTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: gb
Re: SMD inductors - liability to couple with external LF magnetic fields?
« Reply #2 on: October 19, 2019, 04:16:10 pm »
So which is it, ferrite bead, inductor and multilayer or wound? :-//

Sorry,  should have been more specific - I was referring to 0402 or smaller parts.  The lowest cost parts seem to be multi-layer ferrites such as the 220nH Taiyo Yuden HK1005R27J-T. I see that the more expensive LK series, such as the LK10051R0K-T, datasheet explicitly states:

Quote
Internal printed coil structure creates a closed magnetic circuit which acts as
a magnetic shield eliminating crosstalk

The HK datasheet doesn't include that statement though so is that just an omission from the DS?  Are all such multilayer ferrite chip inductors inherently shielded? Both types are listed as unshielded by Digikey but that means little.

Quote
All different things -- ferrite bead chips are always multilayer and saturate easily; ferrite beads (various styles, wound) saturate somewhat less easily but AFAIK always at some fraction of rated current.

I'm not sure how saturation is particularly  relevant here? The DC current will be very low or zero - the question was their susceptibility to induced interference from low level external magnetic fields from nearby transformers or wires.

Quote
- The enclosed types (multilayer, composite, molded, shielded) have good shielding, so that the susceptibility isn't much greater than an equivalent length of wire.
- "Semi shielded" types usually have some variation of a bobbin core, so the external field coupling drops off quickly with distance (relative to the height of the bobbin).  It's a dipole, but it's small?
- Unshielded types are usually rod or solenoid variants and are ordinary dipoles along the axis.

OK,  that appears to be the answer,  thanks.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf