Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

pcb failed ul dielectric test on power connectors

1 view
Skip to first unread message

hartman1

unread,
Jan 15, 2006, 7:40:58 PM1/15/06
to
can anyone recomend what I should do to fix the following,
I have designed a pcb which uses 12-16VAC input and have a common ground
connected on the neg side of power diodes, my panel has failed 1000vac
dielectrical test between ground and ac connector. Im told this is
because i connected directly on other side of diodes, how can I fix this??

no_one

unread,
Jan 15, 2006, 8:31:45 PM1/15/06
to
use a transformer?

"hartman1" <hart...@bellnet.ca> wrote in message
news:tYByf.6541$xk1....@news20.bellglobal.com...

TimPerry

unread,
Jan 15, 2006, 9:04:02 PM1/15/06
to

"hartman1" <hart...@bellnet.ca> wrote in message
news:tYByf.6541$xk1....@news20.bellglobal.com...


if i understand you correctly, you are testing between either of the
transformer primary connections and ground.

nothing on the secondary side of the transformer should have any bearing or
effect on this test. use a transformer with better primary to secondary
isolation.

Palindr☻me

unread,
Jan 16, 2006, 4:07:53 AM1/16/06
to


If the test is being performed between the 12-16VAC input connector of
the panel and ground - I suggest that you point out that the socket is
for voltages <50V and ask them to re-examine their test criteria.

If the test is being performed between the panel mains supply input
connector and ground - then there are at least two scenarios that could
be present:

1)
You need to look to the circuitry prior to the 12-16VAC connector on you
pcb. Either you are using the wrong type of filter components,
transformer,etc - or you are using them incorrectly. Nothing to do with
your power diodes at all.

2) Your circuit design provides supply isolation after the diodes (eg a
switch mode supply with 12-16VAC input - say from an autotransformer -
and an inverter transformer following that provides the isolation).

In which case, the "common ground connected to the negative side of the
power diodes" should not be a common ground at all.

Or, your inverter transformer or feedback components (eg opto isolator)
may be providing a path to ground, through incorrect component
specification, layout or use.

Or, your pcb layout, pre-inverter transformer, is incorrect and you have
brought tracks too close to either a mounting bolt, the case or
something else providing a path to supply ground.

Or, etc.

Quite simply, the information that you have provided is insufficient to
even begin to pin down where the problem could be and hence how it could
be fixed. It may be as simple as needing longer stand-offs for mounting
your pcb, or something as drastic as to need a complete redesign.

But I hope that these suggestions might help, somehow..

--
Sue

0 new messages