Steve,

The resins used to formulate adhesives and soldermasks often have Tg's
at 15-20C. However, they typically have multiple Tg's so if that's the
source of your anomaly I would think you might see it at other
temperature ranges too. The only Tg for the polyimides I'm familiar with
is 400C.

To determine the potential influence of any one layer, use the equation
for the Thermal Expansion(x-y) of a planar composite:
Composite Strain = (Change in Temperature) x SUM(Layer Modulus x Layer
Cross Sectional Area x Layer CTE) / SUM(Layer Modulus x Layer Cross
Sectional Area)
This relationship is used by changing any layer's Stiffness, CTE or
Thickness and observing the effect on the composite response.
Calculating the Composite Strain with/without the soldermask would let
you know if it has any chance of causing the effect you observed.

Keep in mind that the relationship is only valid if the layers stay
bonded to each other, and there isn't any differential creep between the
layers. The stress in any layer is:
Layer Stress = (Layer Modulus) x (Composite Strain - (Change in
Temperature x Layer CTE))

Is your sample staying flat? Warping could be distorting your
measurements.
There's a bending moment generated during the heating of planar
composites that are not constructed symmetrically:
Bending Moment = SUM(Layer Modulus x Layer Cross Sectional Area x
(Composite Strain - (Change in Temperature x Layer CTE)) x (Neutral Axis
Distance From Origin - Center of Layer Distance From Origin))
where:
Distance From Origin to Neutral Axis = SUM(Layer Cross Sectional Area x
Layer Modulus x Distance From Origin to Center of Layer) / SUM(Layer
Cross Sectional Area x Layer Modulus)

Are you precycling the sample to eliminate potential moisure effects?

Andy Magee
Flex Guru
[log in to unmask]
(937) 435-3629

These equations are implemented in my flex circuit analysis computer
program. It's available for download at
http://www.rogers-corp.com/cmu/software.htm but has some intentional
restrictions on the materials available for evaluation.

Steve wrote,

I've been looking and not finding anything, so I'll try posting the
  question here...
  We're doing some investigation of PWB constraint, and as part are
doing
  some thermal expansion coefficient measurements.  On the reference
sets,
  (bare polyimide PWB, 6 & 12 layer unmounted and 6 & 12 layer mounted
on
  aluminum) we're seeing what looks like a hiccup in the TCE curve at
about
  15 degrees C.  (The TCE runs are from -40 C to 100 C).  The setup has
  strain gages mounted on the PWB and a titanium silicate reference
slide,
  then is being heated in at 2 degrees per minute.  From -30 to 15
degrees
  there's a longitudinal TCE of about 18 ppm/C and a transverse TCE of
  about 15 ppm/C.  From 20 to 90 C the longintudinal TCE is 19 and the
  transverse is about 16 ppm/C.  But there's this pronounced dip in the
  range of 15 to 20 C.

  It doesn't show up on the other boards, which are rigidly bonded to
low
  TCE substrates, just the soft bond to aluminum and the boards with no
  bonding at all!  There is soldermask on the panels, any chance that
that
  might have a glass transition at about 15 C?

  Thanks,
  Steve Axdal             [log in to unmask]

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################