RF components distance

[MortenSkelmose]

Hi,

I was wondering if there is any minimum distance that I can place two IC in my PCB design. I can't seem to find any reason except for coupling issues?
But if I wanted to place a VCO and an amplifier - I guess these two could be placed right next to each other?

Cheers!
Morten

[German_EE]

Normally it's not the IC that causes coupling problems, it's the stuff attached to it, especially inductors. Every circuit generates two fields, an electrostatic field and a magnetic one.

The electrostatic field is easily dealt with by using a small amount of shielding around the circuit, often you can get away with just a vertical shield without a top. The magnetic field is the tough one and in the past engineers used something called mu metal to shield things like oscilloscope tubes from the magnetic fields inside the case (especially from the power transformer).

As a rough guide I would have a minimum space around a shielded inductor of 1cm with a 3cm space around an unshielded coil. Things can be improved by laying out your PCB so that adjacent inductors are at right angles to each other, the following picture illustrates this well:

http://wimo.com/bilder/om-6bpf-open-2-l.jpg

Finally, wiring. Wires can also radiate a field and the best way of dealing with this is to route all of your wires along the sides of the shields and the chassis rather than leave them floating in mid air. You then have the option of covering them with a layer of copper tape for 100% shielding.

Two ICs? just put them next to each other.

[Richard Head]

The VCO is one part of the radio that must be shielded.
I have never had a problem with the magnetic field created by low level VCOs exiting the shielding. The shielding, if correctly designed can offer way in excess of 100dB attenuation between adjacent compartments.
I recently designed a diplexer that has something like 120dB isolation between TX and RX in a cigarette box size.

[AF6LJ]

Toroidal inductors are self shielding, this is a big asset in complex and dense RF designs. Other inductors depending on their physical construction should mounted at right angles to each other to minimize coupling. (the plane of the winding should be 90 degrees from another nearby inductor.) Some inductors don't offer you much of a choice and have the plane of their winding in the same plane as the circuit board, then distance or shielding are your friend.

It's complicated, studying the designs of others and your own experience is for the most part what you have to go on.

[MortenSkelmose]

Hi all,

thank you all for your inputs! I'll look to have enough shielding around in all cases!


Cheers!
Morten

[csmithdoteu]

Toroidal inductors are self shielding, this is a big asset in complex and dense RF designs. Other inductors depending on their physical construction should mounted at right angles to each other to minimize coupling. (the plane of the winding should be 90 degrees from another nearby inductor.) Some inductors don't offer you much of a choice and have the plane of their winding in the same plane as the circuit board, then distance or shielding are your friend.

It's complicated, studying the designs of others and your own experience is for the most part what you have to go on.

Does this apply to toroids flat against a ground plane? I'm still struggling on mounting options for mine. I did consider a plastic M3 bolt through the middle with some plastic washers but currently the things I have built are pretty much just flat on the board:

(excuse the kynar - ran out of enamelled wire. Don't use kynar for this - it works but it sucks. It's springy so the inductance changes if you move it)

[German_EE]

To fix one of those toroids in place use some large nylon washers including one between the PCB and the coil. Putting a toroid so close to a copper ground plane does nasty things to the Q.

[csmithdoteu]

Perfect. Thanks for the tip.

[AF6LJ]

Toroidal inductors are self shielding, this is a big asset in complex and dense RF designs. Other inductors depending on their physical construction should mounted at right angles to each other to minimize coupling. (the plane of the winding should be 90 degrees from another nearby inductor.) Some inductors don't offer you much of a choice and have the plane of their winding in the same plane as the circuit board, then distance or shielding are your friend.

It's complicated, studying the designs of others and your own experience is for the most part what you have to go on.

Does this apply to toroids flat against a ground plane? I'm still struggling on mounting options for mine. I did consider a plastic M3 bolt through the middle with some plastic washers but currently the things I have built are pretty much just flat on the board:



(excuse the kynar - ran out of enamelled wire. Don't use kynar for this - it works but it sucks. It's
springy so the inductance changes if you move it)

The short answer;
Toroids are self shielding. The magnetic field is contained within the geometry of the toroid. That also means they are more immune to outside interference from magnetic fields.

[AF6LJ]

To fix one of those toroids in place use some large nylon washers including one between the PCB and the coil. Putting a toroid so close to a copper ground plane does nasty things to the Q.

This is quite true, and another thing. if you place a metal screw or bolt rhough the center of a toroid it must be insulated from any metal said toroid may be mounted to,