Either way you lose, kony. If the 300v buss discharges "in a
few seconds" as you insist, your advice of pushing the ON/OFF
switch will have no effect (since the buss was already discharged.)
Wrong, either way I'm right:
The moment the PSU is unplugged from AC, it begins
discharging. If the system load is light, it will take a
very few (typically under 10 seconds) to discharge enough
that the system logic won't be functional.
If the system puts a heavier load on 5VSB, the inteval in
which the system logic is still functional decreases.
Either way, you're wrong. It will NOT be discharged within
a few dozen ms, nor will it not drain, it will do as I
described, drain over a period that, IF the system power
switch were pressed before it had drained enough, would
cause an attempt to power on the rest of the PSU circuitry
which rapidly drains the remaining charge in the caps. They
would have drained anyway, within a few seconds, but do so
even sooner.
This a VERY easy to see, again a situation where anyone who
has experience and bothers to take voltage readings can see
it clearly.
You are clueless.
If the +5vsb is 60hz transformer based,
Talk about an idiot.
You continue mentioning 60Hz transformer when that is not
used anymore and was rare even in the beginning.
it will drop to below the
Vcc cutoff point rather quickly (the ATX specs say 17ms, minimum)
so the PS_ON circuit will quickly be impotent. Pushing the ON/OFF
switch will have no effect here, either.
IF it were based a that design which nobody uses, yes you
would be correct. That isn't used, and your initial error
in assuming it was, made your following presumption about
what would happen, erroneous.
I have never claimed that "the psu won't drain" (I never use the
word drain), I merely stated that 300v bus discharge time varies
with design. You have been caught twice now making erroneous
statements. Care to try for three?
You wrote "I always kept a 47K bleed resistor handy when
playing with _any_ SMPS, AT or ATX."
That makes it very clear that you have no clue it isn't
necessary to drain them because by the time you've reached
over and grabbed your bleed resistor, there is nothing to
bleed!
Never said it would. I merely stated that was the minimum time based on
the ATX specification. The actual time would be dependent on +5vsb load
and +5vsb circuit capacitance.
BS, you were basing your argument around it.
So are you stating that you have performed these tests, and that
pushing the ON/OFF switch after unplugging an ATX PSU actually
affected the 300v buss? I think you're "embellishing" again, kony.
Never embellished in the first place, you were just too
thick to accept that there is more to it than you had
guessed. You did what far too many do, made a false
presumption then built a house of cards upon it.
Further, you now assume everyone else makes up BS as they go
along. Isn't it a realy strong clue that when I'd suggested
all along that you do the test, that I already knew what the
result would be through same test?
Since it seems you are completely hopeless, I see no reason
to carry on with this thread. Naturally you will try to
claim I'm wrong but it seems of no concern since you don't
even have a clue about the very points we're aguing. Just
to end the thread on a positive note I include the following
data:
=======================
Channel Well CWT-420ATX12 Power Supply
ATX rev. 2.03, 420W
AKA- Antec 400W PP-412
System- miniATX, ECS P6STP-FN board, P3-933 skt 370 based.
119.6 ACV, 326.4 DCV
5VSB with AC connected, no load (no system)= 5.11V
AC Plug Pulled, time till collapse = 11 seconds
5VSB with AC connected, system as load = 5.08V
AC Plug Pulled, time till collapse = 7 seconds
5VSB remained at 5.08 until then.
When AC plug is pulled and 1 second allowed to elapse before
the system (case) front power button pressed, the PSU 12V
rail is now first energized as evidenced by the rear exhaust
fan beginning to spin from prior off state and less than one
second more elapsed until voltage collapsed, when it would
have otherwise taken 7 seconds as observed above with no
change in test or measurement method, only pressing front
power button. 326V DC also quickly collapsed, a moment
(fractions of a second) before 5VSB did. Without any
logging equipment in front of me I can't give an exact ms
figure but deviation is definitely far less than 1/2 second,
probably under 1/4 second.
Above is described AS I did it, not some hypothetical
belief. Since the system used was only a light load on
5VSB, it is expected one more modern, particularly with 5VSB
jumperd to power PS2 or USB devices on the board would
drain faster, decreasing the interval when pressing system
power button would make a difference but not eliminating it
from doing so.
Maybe you know some who aren't hands on, but that does not
mean all are. This IS a hardware forum and hands-on is what
it's all about.
I invite you to gather your own data, I see no point in
further bickering over this so digest the data however you
want, it seems pretty clear to me and I will not argue this
any longer.
