Ed,

 I have a good bit of experience designing small transformers, winding by
hand (protos) and having thousands built in off shore.

 Most of these parts are vacuum impregnated with various varnishes, so of
which are thermosets. Unwinding by hand can be quite destructive and thus
mask the true problem.

 The first question is what types of failure parameters were you seeing?
Functional failure in a circuit is probably not a good indication.
Inductance changes, winding resistance and loss are what you want to
measure. 

 Transformers are so simple I seldom see any failures. Broken cores (dropped
units) and bent pins are the norm. Also the magnetic properties of the
materials varies wildly. Perhaps you got a batch wound on the wrong material
and they are all close to failing? Wrong material can be hard to verify
especially with the power ferrite materials. It is loss at different flux
densities and frequency that is difficult to measure. Check the electrical
resistivity and color to previous old batches.

Bob K.

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ed Popielarski
Sent: Wednesday, October 30, 2013 1:04 PM
To: [log in to unmask]
Subject: [TN] SMT transformer failures

Good Morning Esteemed Colleagues,

We have recently been experiencing increasing failure rates of an SMT (5.2 X
5.2 X 1.2 mm) transformer. The first build in September showed about 6%
fallout. Subsequent builds from the same reel have increased to 25% most
recently. These components have been stored in a normal ambient environment.
I have verified and re-verified reflow conditions are well within
manufacturer's specifications (confirmed by the manufacturer).

When this problem was first identified, I "dissected" one of the failing
devices and found the wire pair (primary and secondary) in good condition on
the outer windings, but as I continued to unwind down to about 1/3 deep, the
insulation began to stick both side by side and to subsequent layers which
would "fuzz" as it was pulled apart with exposed copper implied.

I am beginning to suspect "cold flow", a problem that was prevalent in the
days of wire wrapped backplanes.

Has anyone experienced a similar condition? The manufacturer has been
"working on it" since mid-September and has yet to provide any root cause
and/or solution.

Any insight would be appreciated.


Ed Popielarski
Engineering Manager

[Description: FullLogo]
                               970 NE 21st Ct.
                              Oak Harbor, Wa. 98277

                              Ph: 360-675-1322
                              Fx: 206-624-0965
                              Cl: 949-581-6601

https://maps.google.com/maps/myplaces?hl=en&ll=48.315753,-122.643578&spn=0.0
11188,0.033023&ctz=420&t=m&z=16&iwloc=A


______________________________________________________________________
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] 
______________________________________________________________________