[SI-LIST] Re: decoupling

  • From: Zhiping Yang <zhiping@xxxxxxxxx>
  • To: si-list@xxxxxxxxxxxxx
  • Date: Thu, 19 Jul 2001 10:46:52 -0700

Hi,

Here is my $0.02 input.

Where to put the decoupling caps is a complicated problem.
The PCB stackup is a big factor to consider. No matter where
you put it, the goal is to minimize the total loop inductance
between IC and caps.

Here is what I would recommend:
1. If you can place IC and caps pretty near each other, then
placing them on the same side is better than the opposite side.
2. If IC and caps are on the same side and the power layer is near IC,
then it's better to place cap near GND pin. If the ground layer is
near IC, then it is better to place cap near power pin.  Here is a plot
of suggested placement

      - IC -       -cap-
     |            |       |         |
--*----------*------- GND
                  |       |
------*--*-----------PWR

3. I would recommend to connect IC and cap PWR/GND pins directly
to PWR/GND layer through vias.

Thanks.



johnni.friis@xxxxxxxxx wrote:

> Hi folks
>
> Where do you place the caps ?
> The caps and the IC is on the side of the PCB, or ?
>
> 1. Close to Vcc pin
> The GND connection of the IC and the cap are made directly to GND-plane=
> .
> The Vcc connection is made from Vcc pin to cap to via plane. (See below=
> ).
>
>                        ___
> Vcc pin o-----|      |----o via to Vcc plane
>                       |cap|
>                       |___|-----o via to Gnd plane
>
> 2. Close to GND pin
> The Vcc connection of the IC and the cap are made directly to Vcc-plane=
> .
> The GND connection is made from GND pin to cap to via to plane. (See
> below).
>
>                        ___
> Gnd pin o-----|      |----o via to Gnd plane
>                       |cap|
>                       |___|-----o via to Vcc plane
>
> Some say 2 is best. The current from cap to IC will generate a noise
> voltage in the plane (1. in GND-plane and 2. in Vcc-plane).
> It's best to have the noise in Vcc-plane.
> (Our baords is 4 layers,  plane seperation of  1 mm (40 mill), F=3D40 M=
> Hz,
> singel ended signals, tr/tf  ~ 1nS. )
> What do you think ?
>
> Best regards,
>
> Johnni Friis
> G=E5sdal Bygningsindustri A/S
> Decoration & Sunscreening, Electrics and Shutters
> B=E6kg=E5rdsvej 40
> DK 6900 Skjern
> E-mail: johnni.friis@xxxxxxxxx
> Phone +45 9980 6666
> Fax +45 9980 6260=
>
> ------------------------------------------------------------------
> To unsubscribe from si-list:
> si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
> For help:
> si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
>
> List archives are viewable at:
>                //www.freelists.org/archives/si-list
> or at our remote archives:
>                http://groups.yahoo.com/group/si-list/messages
> Old (prior to June 6, 2001) list archives are viewable at:
>                http://www.qsl.net/wb6tpu
>

--
   Zhiping Yang, Ph. D.
   Hardware Engineer
   Cisco Systems
   270 West Tasman Drive
   Mail Stop:SJCG/2/2
   San Jose, CA 95134             |          |
   email: zhiping@xxxxxxxxx      :|:        :|:
   Tel  : 408 525 5690          :|||:      :|||:
   Fax  : 408 526 5504       .:|||||||:..:|||||||:.
*****************************************************




------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                //www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
  

Other related posts: