>
>Rich says:
>
> >True, however the typical 5ua reverse current ability charges the
>tiny
> >capacitances with aplomb.
>
>Now, where does the 5uA current come from?
No one said there was 5uA. I said that the ability was 5uA. My guess is
that, at 50Hz, no more than 1uA is needed to charge the tiny
capacitances.
>Only when the diode goes into
>reverse leakage, i.e. when the reverse voltage is exceeded.
'Is approached' would be a bit closer. The reverse potential knee on
modern diodes goes pretty gradualee. .
>So it appears
>that to get your distribution, the 5uA leakage flows in the 101volt diode,
>but not the 190 volt diode.
>
Sans-"equalization", the reverse current in the 101v diode is the same as
the reverse current in the 190v diode.
>
>Roll over Kirchoff!
>
>We're leaving the equalised diodes out of this. With equalised diodes, if
>you exceed the reverse breakdown voltage of the diode with the lowest
>breakdown voltage, you are in trouble. (I think Rich agrees on that one).
congrats
>The cure for that is correct design and tolerancing.
>
>Where we are arguing about is non-equalised diode strings using diodes that
>are not guaranteed to be matched.
The diodes in the sample problem are absolutely, positively guaranteed to
have the stated pivs.
>This is where I say that the dangers
>occur, and Rich has to explain:-
The aforementioned example served to explain why unmatched diodes fare
well until the total piv for the string is exceeded. When the string's
reverse current much exceeds c. 5uA, domino-failure undoubtedly occurs.
This example also shows that "equalization" on an unequal string causes
unequal reverse currents that can cause destruction when the reverse
current in the least diode exceeds c. 5uA. . This is precisely what the
Handbook author was talking about in regards to "destructive current".
>
>a) how the voltage distribution when exceeding the total rated reverse
>voltage of the stack does not lead to excess volts across one or more diodes
>(and thus leakage current through the diode)
>and
>b) how the voltage distribution across the stack can be guaranteed to be
>such that an individual diode breakdown voltage is not exceeded with the
>possible spread of parameters in unmatched diodes (including, but not
>limited to trr, Vbr, capacity, and the varying V/I characteristics when in
>reverse breakdown).
As soon as the reverse knee is approached for the least diode, the
reverse potential starts to shift to the more capable units. As long as
the total piv is not exceeded, there is no problem.
Regardinng C considerations: We are talking about 50Hz. For 100kHz
switchers, it's a whole nuther ball game.
>
>Don't need insults for this, Rich - I expect everybody is getting bored by
>it.
>
perhaps they're taking naps.
later, Peter
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|