Hi Steve - the physical act of "smashing" the solder paste doesn't match up
with a technical reason for being a process improvement. In general, the
taller the solder joint, the better thermal cycle solder joint reliability
you can expect so if "smashing" results in shorter solder joints, that
seems backwards. Have you X-rayed the thermistor solder joints to see the
density and location of any voiding:? Are you printing a contiguous block
of solder paste or are you using a checker board pattern? I would guess
that your material combinations are resulting in a fairly healthy CTE
mismatch which is working against the solder joint integrity.

Dave

On Fri, Feb 13, 2015 at 2:30 PM, Steven Kelly <[log in to unmask]> wrote:

> Hi Dave,
> Good question – I am assuming both but I have not seen any parts back from
> my customer to analyze which is giving up first. I will do so I can look at
> the joint as well as the thermistor. The joint I know how to look at – so I
> guess my question is how do I look at the thermistor? Regards Steve
>
> From: David Hillman <[log in to unmask]>
> Date: Friday, February 13, 2015 at 3:11 PM
> To: IPC <[log in to unmask]>, "Steven A. Kelly" <[log in to unmask]>
> Subject: Re: [TN] Voids in solder joints
>
> Hi Steve - yes, the BGA void papers we did don't directly apply to your
> case but there are some parallels. When you say "must pass 15000 thermal
> cycles", do you mean having the thermistor electrically function correctly
> or are you describing solder joint integrity?
>
> Dave Hillman
> Rockwell Collins
> [log in to unmask]
>
> On Fri, Feb 13, 2015 at 1:31 PM, Steven Kelly <[log in to unmask]> wrote:
>
>> Hi All,
>> I have some queries that is a little off the wall- at least for us. We
>> are soldering an aluminum nitride thermistor onto a 2 layer flex. Under the
>> thermistor there is mostly a ground plane and there are vias through the
>> ground to the other ground layer (basically thermals) . We also reflow a
>> .015 solid copper stiffener/heat sink on the other ground layer  of the
>> flex.
>> I have read Mr. Hillman’s papers on voids but in this case we seem to
>> find if we “smash” the solder paste to reduce the voids when we run the
>> finished unit we pass more thermal cycles – I think in reading Mr.
>> Hillman’s papers that makes sense ( I hope).
>> Solder paste is water soluble SAC 305.
>> We also find if we glue the thermistor down and not “smash” the solder
>> paste we pass more thermal cycles- I think this also makes sense to me.
>> We are going to run one more set of tests doing both the smashing and
>> gluing to see what happens.
>> We need to pass 15000 thermal cycles - -40 to + 85 but we think we are
>> seeing deterioration in the thermistor- any thoughts on how we look for
>> that – what kind of analysis can be done for this?
>>
>> Thanks. Steve Kelly
>>
>> ______________________________________________________________________
>> 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] 
______________________________________________________________________