Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.arch.embedded > #13156

Re: Routing signals under crystals

Show all headers | View raw

On 23/08/2013 02:45, Richard Damon wrote:
> On 8/22/13 1:33 PM, Luis Filipe Rossi wrote:
>> Hello,
>>
>> We got a product that uses a low power RTC crystal and a 12Mhz
>> crystal on a 4 layer board. The inner layers are VCC and GND. Should
>> I remove the VCC or/and GND under any of both crystals? Would be a
>> problem to route or place any component on the oposite side of the
>> board? Currently I am removing the VCC layer under both crystals and
>> leaving the GND layer under the 12Mhz crystal to improve shielding
>> but removing under the Low Power RTC Crystal to avoid unwanted
>> capacitance. Is that correct? If yes, considering that there is a GND
>> layer under the 12Mhz crystal, would be a problem to route any signal
>> or place components at the oposite layer?
>>
>> Best regards,
>>
>> Luis
>>
>
> Generally, you would want the crystal, and its bias network and
> interconnect over an isolated ground plane that connects to the rest of
> the ground plane at a single point near the ground reference of the
> oscillator. You do want to double check the datasheets for the crystal
> and oscillator for guidance (i.e. is the parasitic capacitance an issue
> like you mention.
>
> I would then put a similar isolation on the VCC plane to minimize noise
> coupling, or even make it an additional isolated ground plane.
>
> Now, since you have a plane cut here, you do not want to cross that with
> critical traces (either high speed or high power) to avoid the break in
> the reference plane. You can probably allow yourself to place low
> frequency/low power circuits on the other side of the board here.
>
I usually design 4 layer boards with stacked like this:

components
signal layer
ground plane
"power" plane
signal layer
(possibly more components)

I don't remove ground under the crystals because I would rather shield 
the crystal from em radiation and the world from hf noise. If I want a 
stable clock I would use a ready made oscillator for HF.
Low power RTC crystal oscillators using external crystal are quite 
susceptible to radiated em interference.

I'm unconvinced by little areas of isolated ground plane with single 
point links to big ground planes - works great at 50Hz or even 32kHz but 
useless at 1GHz.

I rather think that since we all have different opinions it may not 
actually matter that much.

Michael Kellett

Back to comp.arch.embedded | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Routing signals under crystals Luis Filipe Rossi <luisf.rossi@gmail.com> - 2013-08-22 10:33 -0700
  Re: Routing signals under crystals Tim Wescott <tim@seemywebsite.really> - 2013-08-22 16:25 -0500
    Re: Routing signals under crystals upsidedown@downunder.com - 2013-08-23 09:35 +0300
      Re: Routing signals under crystals Tim Wescott <tim@seemywebsite.really> - 2013-08-23 12:43 -0500
  Re: Routing signals under crystals Richard Damon <Richard@Damon-Family.org> - 2013-08-22 21:45 -0400
    Re: Routing signals under crystals MK <mk@nospam.co.uk> - 2013-08-23 08:40 +0100
      Re: Routing signals under crystals Mike Perkins <spam@spam.com> - 2013-08-23 11:05 +0100
        Re: Routing signals under crystals Luis Filipe Rossi <luisf.rossi@gmail.com> - 2013-08-23 04:55 -0700
          Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-23 16:10 -0400
          Re: Routing signals under crystals Clifford Heath <no.spam@please.net> - 2013-08-24 09:55 +1000
            Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-23 20:23 -0400
              Re: Routing signals under crystals Tom Gardner <spamjunk@blueyonder.co.uk> - 2013-08-24 09:41 +0100
                Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-24 12:10 -0400
                Re: Routing signals under crystals Tom Gardner <spamjunk@blueyonder.co.uk> - 2013-08-24 17:21 +0100
                Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-24 12:53 -0400
              Re: Routing signals under crystals Clifford Heath <no.spam@please.net> - 2013-08-24 23:11 +1000
            Re: Routing signals under crystals Don Y <this@isnotme.com> - 2013-08-23 19:43 -0700
      Re: Routing signals under crystals Richard Damon <Richard@Damon-Family.org> - 2013-08-24 00:34 -0400
      Re: Routing signals under crystals Mark Borgerson <mborgerson@comcast.net> - 2013-08-23 22:16 -0700
        Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-24 12:19 -0400
          Re: Routing signals under crystals Mark Borgerson <mborgerson@comcast.net> - 2013-08-24 09:59 -0700
          Re: Routing signals under crystals Randy Yates <yates@digitalsignallabs.com> - 2013-08-24 22:27 -0400
            Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-25 16:20 -0400
              Re: Routing signals under crystals Randy Yates <yates@digitalsignallabs.com> - 2013-08-25 21:04 -0400
                Re: Routing signals under crystals rickman <gnuarm@gmail.com> - 2013-08-27 12:49 -0400

csiph-web