0
\$\begingroup\$

I've recently started working in microwave field, so I don't have big knowledge except for basic theory.

What I've noticed is that, in rf and microwaves printed circuits (for instance for microstrip lines, patch antennas etc) the substrate material is generally a non - magnetic material (relative permeability about equal to 1).

This tendency has become so standard that most of people (and even substrates datasheets) never mention the relative permeability, since it's known to be 1.

Here and Here you may find a list of typical substrates materials, which are all characterized by non-magnetic properties, and which are choosen mainly because of their loss tangent (dissipation) and dielectric constant.

So, my question is: why not using materials with relative permeability greater than 1?

High electric permittivity materials are used when we need to:

  1. Increase the capacitance between two metal objects;

  2. Decrease the wavelength (for a fixed source frequence). This allows to make the structure (patch antenna, microstrip line etc) seem bigger with respect to the wavelength, and so its size may be reduced. This allows to minimize devices sizes at the cost of higher parasitic effects.

Well, even choosing a material with high magnetic permeability can reduce the wavelength and provide the advantage 2), since:

enter image description here

The role of magnetic permeability on reducing the wavelength is exactly the same of electric permittivity. No difference. But why magnetic materials are rarely used?

\$\endgroup\$
4
  • \$\begingroup\$ .. such as? What candidate materials exactly? What manufacturing process requirements does that change? \$\endgroup\$
    – pjc50
    Feb 15 at 10:54
  • \$\begingroup\$ Like Andy said, the extra losses are a show stopper in many RF/microwave applications. Some designers would sell their mothers for 0.5 dB lower loss <wink>. Material chemists & designers spend years coming with better low-loss materials to use as a substrate. \$\endgroup\$
    – SteveSh
    Feb 15 at 12:53
  • \$\begingroup\$ Non-standard process. It's a bit of a circular argument though, because if there was a big enough advantage, a process allowing e.g. high permeability PCB substrates would quickly become standard. Until someone shows such an advantage and develops such a process, it'll be expensive (in time at least) to manufacture. (May be cheap in materials. If I had the need I'd be mixing epoxy resin with ferrite powders as well as glass weave) \$\endgroup\$ Feb 15 at 16:27
  • \$\begingroup\$ Magnetic permeability tends to be strongly frequency dependent, and drops off rapidly at high frequency. I'm not even sure you can get materials like you're thinking at microwave frequencies. If you could, they'd probably have very limited range of working frequencies, which would limit their utility. \$\endgroup\$ Feb 15 at 16:35
4
\$\begingroup\$

But why magnetic materials are rarely used?

A few thoughts: -

  1. Extra cost
  2. Extra losses (eddy current losses especially as frequency rises)
  3. An extra process
  4. Another variable
  5. Changes in permeability with temperature
  6. Changes in permeability with frequency

Another thought

  • Having a magnetic substrate across the whole PCB will mean extra undesirable properties on PCB tracks that should not have inductance increases. To avoid that would mean that the magnetic substrate will need to be implemented in some areas and not others.

The only benefit I can envisage is that you can thicken up tracks for 50 Ω impedance and get more current throughput. Or, you can make boards thinner whilst still maintaining the same characteristic impedance.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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