AVR-Chat

The data sheet says that the large pad under the device "should" be 
soldered to ground.. 

Does anyone have direct knowlege of wether is HAS to be?
Assume all perimiter ground pins are in fact grounded.

Dave VanHorn wrote:

> The data sheet says that the large pad under the device "should" be 
> soldered to ground.. 
> 
> Does anyone have direct knowlege of wether is HAS to be?
> Assume all perimiter ground pins are in fact grounded.

Not direct knowledge, no.

Speaking generally, these type pads have three roles--thermal
conduction, EMI reduction, and lowered ground inductance.

The first is hard to get away from, unless you can guarantee the device
will be operated at a reduced ambient temperature or at a reduced power
consumption, or with a beefier heat sink.  Some parts specify different
limits with or without the pad soldered.

Having a ground shield under the die can improve noise immunity,
especially in sensitive circuits like ADCs or other mixed analog-digital
chips.  You can get something like this with a circuit board ground
plane underneath, but it's not as close to the die as the pad is in the
package.

Finally, the large pad can serve as a low-inductance ground return that
helps avoid ground differentials in the chip.

Personally (not with AVRs, though), for home brew stuff, I've put in a
ground plane on both sides and an array of plated vias inside that,
filling the vias with solder and hoping at least some of it gets wicked
up onto the pad surface.

Your mileage may vary, of course!

-Johnathan, AE6HO

My T26 data sheet (12/03) does not seem to say anything
about it, or even that it IS solderable. TI usually notes
things about a thermal-pad quite carefully; nothing seen
here. In fact, this data sheet only has the word "solder"
once, and that refers to the tip region of SOIC pins.

Jim


On Wed, 02 Nov 2005 15:47:04 -0800
 Johnathan Corgan <jcorgan@aeinet.com> wrote:
> Dave VanHorn wrote:
> 
> > The data sheet says that the large pad under the device
> "should" be 
> > soldered to ground.. 
> > 
> > Does anyone have direct knowlege of wether is HAS to
> be?
> > Assume all perimiter ground pins are in fact grounded.
> 
> Not direct knowledge, no.
> 
> Speaking generally, these type pads have three
> roles--thermal
> conduction, EMI reduction, and lowered ground inductance.
> 
> The first is hard to get away from, unless you can
> guarantee the device
> will be operated at a reduced ambient temperature or at a
> reduced power
> consumption, or with a beefier heat sink.  Some parts
> specify different
> limits with or without the pad soldered.
> 
> Having a ground shield under the die can improve noise
> immunity,
> especially in sensitive circuits like ADCs or other mixed
> analog-digital
> chips.  You can get something like this with a circuit
> board ground
> plane underneath, but it's not as close to the die as the
> pad is in the
> package.
> 
> Finally, the large pad can serve as a low-inductance
> ground return that
> helps avoid ground differentials in the chip.
> 
> Personally (not with AVRs, though), for home brew stuff,
> I've put in a
> ground plane on both sides and an array of plated vias
> inside that,
> filling the vias with solder and hoping at least some of
> it gets wicked
> up onto the pad surface.
> 
> Your mileage may vary, of course!
> 
> -Johnathan, AE6HO
> 
> 
> 

---------------------------------------------------------------
The Think Different Store
http://www.thinkdifferentstore.com/
For All Your Mac Gear
---------------------------------------------------------------

Atmel produce a demo board for their ATR2406 radio IC which includes an 
ATmega88 in this package and they did not put a large pad under the 
mega88. They do say they will change it when they up issue the board but 
the board works fine and they are not too worried since the processor has multiple GND pins. The GND pad does make that package more difficult to 
track on a two layer board particularly because the package is so small. 

Dave Pallant | Hardware Design Engineer

Promethean Limited | TDS House
Lower Philips Road | Blackburn | Lancashire | UK | BB1 5TH
Please note new telephone number

Tel
: +44 (0) 1254 290826 
Fax
: +44 (0) 1254 291673
E-mail
: dave.pallant@prometheanworld.com
Web
: www.prometheanworld.com


AVR-Chat@yahoogroups.com wrote on 03/11/2005 00:29:43:

> My T26 data sheet (12/03) does not seem to say anything
> about it, or even that it IS solderable. TI usually notes
> things about a thermal-pad quite carefully; nothing seen
> here. In fact, this data sheet only has the word "solder"
> once, and that refers to the tip region of SOIC pins.
> 
> Jim
> 
> 
> On Wed, 02 Nov 2005 15:47:04 -0800
>  Johnathan Corgan <jcorgan@aeinet.com> wrote:
> > Dave VanHorn wrote:
> > 
> > > The data sheet says that the large pad under the device
> > "should" be 
> > > soldered to ground.. 
> > > 
> > > Does anyone have direct knowlege of wether is HAS to
> > be?
> > > Assume all perimiter ground pins are in fact grounded.
> > 
> > Not direct knowledge, no.
> > 
> > Speaking generally, these type pads have three
> > roles--thermal
> > conduction, EMI reduction, and lowered ground inductance.
> > 
> > The first is hard to get away from, unless you can
> > guarantee the device
> > will be operated at a reduced ambient temperature or at a
> > reduced power
> > consumption, or with a beefier heat sink.  Some parts
> > specify different
> > limits with or without the pad soldered.
> > 
> > Having a ground shield under the die can improve noise
> > immunity,
> > especially in sensitive circuits like ADCs or other mixed
> > analog-digital
> > chips.  You can get something like this with a circuit
> > board ground
> > plane underneath, but it's not as close to the die as the
> > pad is in the
> > package.
> > 
> > Finally, the large pad can serve as a low-inductance
> > ground return that
> > helps avoid ground differentials in the chip.
> > 
> > Personally (not with AVRs, though), for home brew stuff,
> > I've put in a
> > ground plane on both sides and an array of plated vias
> > inside that,
> > filling the vias with solder and hoping at least some of
> > it gets wicked
> > up onto the pad surface.
> > 
> > Your mileage may vary, of course!
> > 
> > -Johnathan, AE6HO
> > 
> > 
> > 
> 
> ---------------------------------------------------------------
> The Think Different Store
> http://www.thinkdifferentstore.com/
> For All Your Mac Gear
> ---------------------------------------------------------------
> 
> 
> ------------------------ Yahoo! Groups Sponsor --------------------~--> > Get Bzzzy! (real tools to help you find a job). Welcome to the Sweet 
Life.
> http://us.click.yahoo.com/A77XvD/vlQLAA/TtwFAA/dN_tlB/TM
> --------------------------------------------------------------------~-> > 
> 
> Yahoo! Groups Links
> 
> 
> 
> 
> 
> 



Promethean Technologies Group Ltd and or associated and or subsidiary companies :

The views expressed in this communication may not necessarily be 
the views held by Promethean Technologies Group Ltd and or associated and or subsidiary companies.

This e-mail is for the exclusive use of the addressee(s). Unauthorised 
disclosure, copying or distribution is prohibited.

This e-mail message has been swept for the presence of computer viruses.

Promethean Technologies Group Ltd and or associated and or subsidiary companies accepts no liability for any loss resulting from this email transmission.



*****************************************************************
This email has been checked by the e-Sweeper Service
*****************************************************************

Well, it turns out that YES it does have to be soldered.
It's a mechanical issue though.  

If you don't solder it, then the pad solder joints will end up being 
cracked by any small amount of board flex that might occur in thermal 
cycling, or a drop. 

Many thanks to atmel for failing to put this information into the data 
sheets.  It is circulating within atmel, in the form of a report 
attached to an email.