[Elecraft] K3 Shields - Theory of Operation?

[Alan Bloom]

On Fri, 2008-05-16 at 12:50, Dan Levin wrote:
> Naïve question...
> 
> I'm looking at the shields on the inside of my K3 (the one that separates
> the RF compartment from the front panel, and the one that segregates the
> KPA3 from the rest of the system), and I'm wondering why they work :-)
> 
> My questions fall into three areas:
> 
> 1) These shields are full of holes and slots.  

As long as the longest dimension of the hole is much smaller than a
wavelength (less than 1/10 wavelength is the rule of thumb) it doesn't
degrade the shielding very much.

> 2) The PC board forms the "bottom" of the shield box in both cases.  Yet the
> shields aren't connected electrically to the boards.

It looks to me like the PA shield is indeed connected to the RF board
(motherboard) and also to the top cover.  (Notice how the paint has been
removed on the top cover to make a good connection.)  It looks like the
front-panel shield is connected to the front-panel board through the
chassis.

> 3) The connectors that pass through the shields, especially the ones that
> pass through the front panel shield, don't appear to be RF proofed in any
> way.  

That can indeed be a path for RFI to couple through the shield.

I think the general answer to your question is that the best shielding
is a solid aluminum box with all power and signal lines going through
coaxial feedthrough capacitors and all coax cables using connectors
mounted to the box.  Sometimes you have to do that kind of thing if you
are designing something like a microwave spectrum analyzer (I used to
work for HP/Agilent) where (1) the wavelengths are very short, and (2)
all detectable spurs must be surpressed over a very wide bandwidth.  But
for an amateur transceiver operating in the HF range, the requirements
are far less stringent.

Al N1AL