TECHNET Archives

January 2013

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Steven Creswick <[log in to unmask]>
Reply To:
TechNet E-Mail Forum <[log in to unmask]>, Steven Creswick <[log in to unmask]>
Date:
Wed, 30 Jan 2013 18:32:57 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (191 lines)
Victor,

You definitely need to review the product data sheet and MSDS for the
specific thermal compound which you are using.

The Product Data sheet should list the dielectric constant, which can give
you an idea of what impact this material might have on your circuit's
operation.

The Product data sheet should also list the breakdown voltage of the paste.
It will likely range from 100's to 1000's of volts/mil, etc.

Hopefully, the MSDS would tell you if you have a metal filler.  Most likely
it is an oxide filler, but you need to check for sure.

Some thermal materials harden up quite distinctly, others remain 'squishy
and sticky' for a long time, as I believe it was Larry that mentioned.  If
your material is one of those that hardens, I would immediately begin to be
concerned.  One of the first things I would contemplate is doing some DSC
testing to get a feel for the material's thermal coefficient of expansion
over the temperature range that is applicable to your product.  Silicones
will typically have high CTE's [100-200 PPM/°C], but may have a low enough
modulus to not be offensive to your product.
 
I've never done it, but you might be able to suction off some of the excess
paste, if it is a gel/grease-like material.  Regardless, it will still be a
huge mess.

As has been said, we can't definitively answer your question about
functionality since we don't know the specific material, nor the circuit,
nor application.


Steve Creswick
Sr Associate - Balanced Enterprise Solutions
http://www.linkedin.com/in/stevencreswick
                         616 834 1883



-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Inge Hernefjord
Sent: Wednesday, January 30, 2013 4:17 PM
To: [log in to unmask]
Subject: Re: [TN] Gray Thermal Heat Sink Compound-paste

Hi Vic Tor,
Application of thermal paste is an art.The interface between the heat
generating component and the surrounding is often meticously calculated,
because the heat transfer capability and precision decides the life of the
component .either it's chip, a BGA, a power transistorr or else. Removal of
heat is often the most forgotten parameter. One reason why Ericsson had such
a success decade after decade is that a guy started todig deeply into this
topic. He made models, travelled around and preached his 2cooling message"
on Ericsson's facilities and fostered a whole generation of design engineers
to do better heat transferrng constructions.

The heat flow is representated in a equation as a lot of series- and
parallel coupled resistances, with addition of series/parallel coupled
capacitors. Now, let' s take a power transistor against a chassis. If you
torque the package naked, you get three main heat flows:
convection/radiation to ambient, a flow path creatid by the many
metal-to-metal contacts and radiation cross the micrometer thick non-metal.
I performed the calculations for leading away heat om a super computer. The
volume for the twenty stacked wafers was  a cube, 12 inches sides, and my
job was to remove  12,000 Watts peak power! I filled the gap between the
wafers with Helium and the heat exchange was a ,salt solution.

Now Vic Tor, if you fill the space wth 'a good thick layer' you may make a
big mistake, depending on what paste you use. The heat transfer of a paste
is not super, which many think. A thick layer can result in a worse heat
transfer, than a naked , well torqued transistor. The paste is used only for
improving the volumes of non-physical contact, i.e. something that is better
than air. If you have a fat paste or mat layer, you risk that the
metal-to-metal contacts be fewer. And there is a risk for material flow,
whereby the the important good contacts become fewer too.With other words :
use as thin fill-up as possible.

I have some real good stuff  about this, but my contract fromEricsson is
vivid as long as I live. In my opinion there is nothing that isn't found in
today's computer models, but Ericsson are extremly stringent. But you'll
manage on your own, there are lots to buy, put in your facts, press enter
and the computer does the job.

I will add some good stuff in my Dropbox, will take little time...

Inge




On 30 January 2013 16:22, Victor Hernandez
<[log in to unmask]>wrote:

> Folks,
>
>    Again I state my inquiry, " Can/theatrically excessive saturation 
> of the SMT components-BGA, etc. devices with the above stated product can
a
> leakage condition".   I don't feel I got an answer.   I am very aware that
> the placement of the paste should only be over the die surface and with a
> control volume.    I don't have the brand and/or composition make of the
> gray material before me, only the aftermath effects of the component 
> /BGA over saturation.
>
> Victor,
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Nutting, Phil
> Sent: Wednesday, January 30, 2013 9:05 AM
> To: [log in to unmask]
> Subject: Re: [TN] Gray Thermal Heat Sink Compound-paste
>
> Hi guys,
>
> In our testing the type of thermal paste does not make a difference if 
> the surfaces are really smooth (16 µin or better surface).  Where it 
> makes a difference is in rougher surfaces.  And Brian is right, more is
not better.
>
> For electrical isolation we have switch from a thixotropic phase 
> change material (Isostrate) to T-gard with great results.
>
> Just my 2 ¢
>
> Phil
>
> -----Original Message-----
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Brian Ellis
> Sent: Wednesday, January 30, 2013 9:52 AM
> To: [log in to unmask]
> Subject: Re: [TN] Gray Thermal Heat Sink Compound-paste
>
> It depends.... on its composition. If it's grey, it may rely on fine 
> metallic powder. I have some Thermaltake TG2 which the makers claim 
> has low electrical conductivity but I suspect it may have metal in it. 
> I use it for CPUs to radiators, so its electrical characteristics are
unimportant.
> Consult the manufacturer's data. BTW, don't put on an excess; if it 
> squeezes out, you're wasting a costly product :(
>
> Brian
>
> On 30/01/2013 15:41, Victor Hernandez wrote:
> > Folks,
> >
> >     Can excessive saturation of the SMT components-BGA, etc. devices
> with the above stated product can a leakage condition.   Cosmetic wise  it
> doesn't look pretty.   I am concern functionally over a long period of
time.
> >
> > Victor,
> >
> > ____________________________________________________________________
> > __ This email has been scanned by the Symantec Email Security.cloud 
> > service.
> > For more information please contact helpdesk at x2960 or 
> > [log in to unmask] 
> > ____________________________________________________________________
> > __
> >
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please contact helpdesk at x2960 or 
> [log in to unmask]
> ________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please contact helpdesk at x2960 or 
> [log in to unmask]
> ________________
>
> ______________________________________________________________________
> This email has been scanned by the Symantec Email Security.cloud service.
> For more information please contact helpdesk at x2960 or 
> [log in to unmask] 
> ______________________________________________________________________
>

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________

______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________

ATOM RSS1 RSS2