Hello from San Diego:
A few weeks ago, I asked for anecdotes of failure of
basic insulation after passing the production-line
hi-pot test. Here is a summary of the responses.
I've classified the failures into three broad classes,
mechanical failures, thermal failures, and electrically-
caused failures.
MECHANICALLY-CAUSED FAILURES.
PHYSICAL DAMAGE.
Leads breaking through sheet insulator.
Heat sink extrusion flaw penetrated insulation.
Cord abrasion at sharp edges and strain-relief.
Motor winding failure due to mechanical damage.
Diode stud cut wire insulation.
Cord insulation failure due to crush under furniture.
Pinched between appliance inlet and metal frame.
Magnet wire outside slot insulation.
Caught in metal-to-metal joint.
(Some of the physical damage incidents were manifested
over time due to the cold flow of plastic material as
a function of time. Initially, the insulation passed
hi-pot, but failed at a later time.)
BRIDGING.
Bridged by leaking capacitor electrolyte (2 reports).
Bridged by loose screw.
Bridged by metal shavings (2 reports).
THERMALLY-CAUSED FAILURES.
Overheated components caused carbonized current paths.
Overheated transformer caused insulation failure.
Transformer thermal protector installed with shorted leads.
Arcing in circuit-breaker heated the terminal insulation.
ELECTRICALLY-CAUSED FAILURES.
(No reports.)
My motive in requesting this information was to determine
whether or not the electric strength of basic insulation
is adequate. A follow-on question is whether hi-pot
testing is contributing to early failure of insulation.
HP has studied the electric strength of insulation as a
function of time and temperature in our optocouplers.
The results of this study are published in the HP
Optocoupler Application Note 1074. This note can be
found on the web:
http://www.hp.com/HP-COMP/isolator/app_index.html
(This is an Adobe Acrobat file.)
There is lots of circumstantial evidence to conclude:
Hi-pot test voltage, hi-pot test waveform, and
hi-pot test time are not critical to solid
insulation or Y-capacitor lifetime.
Electric strength requirements in safety standards
are at least adequate, and may be quite conservative.
The electric strength of modern insulating materials
is very much greater than the electric strength
required by safety standards.
On the other hand, we don't do a good job of identifying
mechanical threats to basic insulation.
Best regards,
Rich
