AVR-Chat

Thanks, Graham and George.  

 

I already follow these rules with DIP packages, except for the long slits
one - I'll keep them shorter in future.   And switch to SMT ceramic
decoupling cap's (I have been using 0.1 uF tantalum thru-hole).  

 

Your mention of hi and lo in parallel brings up something I have wondered
about for years.   I typically use LP2950 voltage regulators and the data
sheet specifies a capacitor on output.  For my current design I am using 68
uF.    The Vcc power bus from this goes off to the various IC's, each with
their own 0.1 uF decoupling capacitor.  Is this the situation you are
referring to?

 

Steve

 

From: AVR-Chat@yahoogroups.com [mailto:AVR-Chat@yahoogroups.com] On Behalf
Of ecros_technology
Sent: Saturday, October 24, 2009 8:42 AM
To: AVR-Chat@yahoogroups.com
Subject: [AVR-Chat] Re: pcb layout of SOIC packages

 

  

--- In AVR-Chat@yahoogroups.com <mailto:AVR-Chat%40yahoogroups.com> , "Steve
Hodge" <steve@...> wrote:
>
> I'm a newbie to SMT devices and I'm
> doing my first pcb layout with them.
> With DIP packages I usually arrange
> them in a column straddling a Vcc
> power bus. Is it OK to do the same
> with SOIC packages?

VCC routing matters very little if you follow these two important rules.
One, connect all grounds with something as close as possible to a continuous
plane. Two, decouple each VCC pin as close to the pin as possible with a
single, low-inductance ceramic capacitor connected to the ground plane.

For SMT designs, most of the ground plane will be on the bottom. Try to
route on the top as much as possible (that's where the component pads are)
and cross to the bottom as necessary, but don't make long slits in the
ground plane.

SMT ceramic capacitors will perform better than leaded capacitors for
decoupling. Use the smallest size you can handle. 0805 shouldn't be too
hard, but I've moved on to 0603 and am starting to use 0402 in dense layouts
(but I wouldn't recommend you start there). A good value is 0.1uF. Use the
same value everywhere (using a high value and a low value in parallel was
once popular but actually makes things worse).

Graham.





__________ Information from ESET Smart Security, version of virus signature
database 4538 (20091024) __________

The message was checked by ESET Smart Security.

http://www.eset.com



[Non-text portions of this message have been removed]