[SI-LIST] Re: vias and current pathes
- From: "O. Laney" <olaney@xxxxxxxxx>
- To: gnuarm.2006@xxxxxxxxx, si-list@xxxxxxxxxxxxx
- Date: Sat, 21 Jul 2012 06:27:41 +0000
An essential piece of information is that plane overlap capacitance is
distributed, and distance and time are related by the velocity of
propagation in the local medium. Thus, the higher the frequency, the
more local the effective interplane capacitance. The total capacitance
at low frequencies is simply not applicable as the effective radius
shrinks with frequency. This smaller effective local capacitance can
have appreciable reactance even if the planes are huge. Two more
effects come into play. One is that the change in plane depth as a
signal crosses over a plane transition reveals itself as a step change
in transmission line impedance, causing signal reflection. Another is
that the interplane cavity can resonate, possibly strongly enough to
cause problems elsewhere on the board (or EMC compliance failure for
the system as a whole).
Stitching techniques and other common treatments are a practical matter
based on how electrons actually behave, as shown on a VNA or TDR.
Suffice to say, constant impedance construction and intimate return
paths (aided by local vias) are classic high frequency treatments
because they work. It's between you and your boss if you want to bet
your design on an alternative theory.
Orin
------ Original Message ------
From: "Rick Collins" <gnuarm.2006@xxxxxxxxx>
To: si-list@xxxxxxxxxxxxx
Sent: 7/20/2012 8:34:21 PM
Subject: [SI-LIST] Re: vias and current pathes
>The OP asked about RF effects. I believe that for RF, unless you
>have a very unusual board, overlapping ground planes will have near
>zero impedance between them. So there does not need to be any
>special vias located near the signal path. The return current will
>flow directly between the two ground planes as if they were connected for RF.
>
>This is also true of interruptions in the continuity of a single
>plane, either ground or power. If a signal crosses a break in the
>plane, but the plane on both sides of the break adequately overlap an
>adjacent, not interrupted plane, the return path will couple from one
>ground plane to an overlapping power or ground plane and back to the
>other half of the interrupted ground plane. For RF all the
>overlapping power and ground planes are like one.
>
>Rick
>
>
>At 05:28 PM 7/20/2012, Aaditya Kandibanda wrote:
>
>>
>>Herbert,
>>Whenever a trace changes layers, the return current flows through the vias
>>that stitch the adjacent ground planes of those two layers. If those two
>>planes are not stitched, there would be a discontinuity for the return
>>path.
>>
>>Thanks and regards
>>Aaditya
>>
>>On Fri, Jul 20, 2012 at 2:19 PM, Herbert <afu@xxxxxxxxxxxxxxxxxx> wrote:
>>
>>> Hello to all,
>>>
>>> I am new here and I have searched about the real
>>> current path for RF on a combination
>>>
>>> trace - via - trace
>>>
>>> On the trace there is current flow more or less identical
>>> on both sides of the trace ( top - bottom ).
>>> When flowing through the via , I would think, there is a more or
>>> less equal corrent on inside surface and on outside surface.
>>> That currents pass through the via and flow then on top and bottom
>>> of the trace on the second side.
>>> What about the interchnge of current between the sides and which influence
>>> has the pad ring of the via ?
>>> What about the idea of having less skinn effect when having
>>> that situation with a ground plane on the second side ( means a grounding
>>> via ) and using so called "thermal reliefs" ?
>>> I have the opinion, that the thermal releafs have a negative effect
>>> on the overall impedance of the via construction WITH thermal reliefs.
>>>
>>> Waiting for good ideas
>>>
>>> Herbert
>>>
>>> ------------------------------------------------------------------
>>> To unsubscribe from si-list:
>>> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>>>
>>> or to administer your membership from a web page, go to:
>>> //www.freelists.org/webpage/si-list
>>>
>>> For help:
>>> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>>>
>>>
>>> List forum is accessible at:
>>> http://tech.groups.yahoo.com/group/si-list
>>>
>>> List archives are viewable at:
>>> //www.freelists.org/archives/si-list
>>>
>>> Old (prior to June 6, 2001) list archives are viewable at:
>>> http://www.qsl.net/wb6tpu
>>>
>>>
>>>
>>
>>
>>------------------------------------------------------------------
>>To unsubscribe from si-list:
>>si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>>
>>or to administer your membership from a web page, go to:
>>//www.freelists.org/webpage/si-list
>>
>>For help:
>>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>>
>>
>>List forum is accessible at:
>> http://tech.groups.yahoo.com/group/si-list
>>
>>List archives are viewable at:
>> //www.freelists.org/archives/si-list
>>
>>Old (prior to June 6, 2001) list archives are viewable at:
>> http://www.qsl.net/wb6tpu
>>
>>
>
>
>------------------------------------------------------------------
>To unsubscribe from si-list:
>si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
>
>or to administer your membership from a web page, go to:
>//www.freelists.org/webpage/si-list
>
>For help:
>si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
>
>List forum is accessible at:
> http://tech.groups.yahoo.com/group/si-list
>
>List archives are viewable at:
> //www.freelists.org/archives/si-list
>
>Old (prior to June 6, 2001) list archives are viewable at:
> http://www.qsl.net/wb6tpu
>
>
>
>
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
