RF attenuator ground plane question

Hi all I'm KJ4PTD, I have designed a crystal controlled RF attenuator to use in "radio directional finding" (using a receiver to locate an RF source) when you get close to the source the signal can get too strong to pinpoint, thus the need for an attenuator. The PCB I have designed to hold the components uses 6V-DC to oscillate the crystal and therefor has a DC ground. Also on the board is the antenna input as well as the receiver output and the ground for the RF signal.

Being a PCB that handles RF I have included a ground field plane, QUESTION is should I bond it to the DC ground or the RF ground?

Help?

Thanks
KJ4PTD

For such simple low power circuit you can use a common solid ground for both, DC and RF.

PS: Impressive home brew CW key in your shack
73's

Hi KJ4,
I think you has so & so only one GND in your system? Whats your RF.GND pls, is it a shild & over a capacitor on DC GND?
73 de Karesz

Depends on. You can possibly detect traces of the digital clock with a high sensitive receiver or spectrum analyzer. If it
matters, you may want to think about an option to stop the clock.

I don't see anything in the initial post about digital traces or digital ground. Is specified very clear that is about DC and RF grounding.

I think, "traces of the digital clock" should be self-explanatory related to RF problems. You may prefer the term "spurious signals".

It can be a complicated mess to figure out where grounds should be.

If you are using most modern MMIC chips, you will find that the chip ground is also the RF ground. As such, I usually have the RF ground plane also be the DC ground plane.

But there are probably lots of scenarios where you might want separate DC from RF grounds. For example, if there are large DC currents flowing, such as in a motor control.

In any event, IF your DC ground is also a ~ solid ground plane, you HAVE TO connect the RF ground to the DC ground in multiple places. The reason is that if you have two ground PLANES, you are capturing a big sandwich of dielectric between them. WHen RF energy gets into that sandwich, it can resonate at certain frequencies, and that resonance can cause all sorts of havock! Amplifiers can start to oscillate, noise levels can be very high, etc. So you want ground via holes between the two ground PLANES at around λ/8 spacing, or smaller.

If you don't have a DC ground PLANE, but just a bunch of DC traces, things are a lot easier. In that case, you might only have to add a ferrite bead here or there to stop RF from popping up in the wrong place.

vfone said:

For such simple low power circuit you can use a common solid ground for both, DC and RF.

PS: Impressive home brew CW key in your shack
73's

Click to expand...

Thanks for the complement, I am just teaching myself CW and had to have a key. I figured I could buy one at a ham fest for $10 or so or build one with stuff I already had in the shack. The key sounds great and makes good positive contact, I have received good reports on the audio from it. I think the bands may be opening up for us finally, Horay!

Thanks for all the replys, it has given me stuff to think about and consider, I think I will just run the DC and RF grounds together, after all the premise of how my attenuator works is by introducing the 1Mhz signal from the crystal into the RF signal through a pot causing intentional interference in order to attenuate the source. I will mostly be using it for Fox-Hunts where audio quality isn't important just the S units.

73's all de KJ4PTD