Re: gEDA-user: basic anti-EMI design q

[Xtian Xultz <xultz@xxxxxxxxxxxx>]

Hi John

Sorry if my question is dumb, but sometimes I see shields in printed circuits 
made by lines that crosses, forming a mesh. And sometimes I see shields made 
by a polygon. 
What is the difference between both concerning EMI? What cases one are best 
and the other worst?

Thanx for any help


> Well, the general way to do this involves wave optics with complex
> refractive indices, but here (normal incidence, thick medium relative
> to skin depth, extreme index mismatch) we can simplify.
>
> Using handbook formulas, I get skin depth of 7 microns and sheet
> resistance of 3 milliohms. Empty space acts like it has a sheet
> resistance of 377 ohms, so you've got high Z driving low Z, loss
> roughly .003/377 or ~10^-5 in amplitude. 200 microns is ~30 skin
> depths, amplitude loss is ~exp(-30), or about 10^-12. On the far side
> of the sheet, low Z drives high Z so there's negligible amplitude
> loss. Overall we have an amplitude loss of ~10^-17 or 340 dB.
>
> Even very thin metal foils are extremely good RF shields where
> contiguous. It's leakage though holes, joints, and especially wires
> going through that makes trouble.
>
> John Doty              Noqsi Aerospace, Ltd.
> jpd@xxxxxxxxxxxxx