boot on power-up
Skeleton Man
I'm trying to accomplish what I thought was an extremely simple task, of putting
a capacitor across the atx power switch so it will turn on as soon as power is
applied (as the bios does not have this feature).
I tried 3 different caps as follows:
1. 0.01 uF ceramic - does nothing
2. 0.47 uF greencap - does nothing
3. 10 uF electrolytic - will work the first time power is applied, but nothing
after that untill I drain it (remove and short the leads for a few seconds).
What value capacitor should I use ? Do I need a resitor across it to drain it
quickly when power is lost ?
I thought it would be a simple thing and the value wasn't important (ie. I could
use any reasonable value).
Regards,
Chris
w_tom
Jan 2005 as an Idea for Design entitled "Autostart circuit
helps ATX motherboards resume operation after power
interruption".
Skeleton Man
>Jan 2005 as an Idea for Design entitled "Autostart circuit
>helps ATX motherboards resume operation after power
>interruption".
I want just a single capacitor if possible..(I have seen many people report this
can be done).. not a 555 based circuit.. (overkill)
Regards,
Chris
kony
I happened to have a few parts out so I tried a cap/resistor
combination. 10 mfd electrolytic plus 24K Ohm resistor
works on the board I tried. You'd want to try shorting the
power-on pins with any resistor before putting it together,
to confirm the value isn't pulling it too low... at least
it would save time to do that first.
w_tom
computer. A design unique only for your machine (which was
not apparent in the first post), then use a diode connected
from capacitor to +5 volts so that when +5 volts drops to
zero, +5 volts then discharges the capacitor. This is the
same old design used for power reset circuits in the 'good ol'
days'.
This design probably will not work for short power
glitches. But it is simple.
Skeleton Man
>computer. A design unique only for your machine (which was
>not apparent in the first post), then use a diode connected
>from capacitor to +5 volts so that when +5 volts drops to
>zero, +5 volts then discharges the capacitor. This is the
>same old design used for power reset circuits in the 'good ol'
>days'.
Sounds simple enough and even with the resistors I can't get it working
currently. Nomatter what I do, I have to remove the capacitor from the circuit
entirely, discharge, and then short the leads and it works (even with the power
switch on the rear of the PSU off, it still won't let me drain the cap even by
shorting the leads)
My machine: Celeron 500Mhz system, Gigabyte GA-6OX mainboard.
Regards,
Chris.
kony
You are waiting a few seconds after the power was cut out,
before turning it back on, right? That could be necessary
for the 5VSB circuit to drain. By a few, I mean 6 seconds
should be more than enough. Ideally when implementing such
a simple hack it would be with the system already on an UPS
such that there is no time a very brief, less than a few
seconds outtage is seen. The power being out long enough
for the UPS to be exhausted, it would be pretty unlikely
that the main power came back online within the next few
seconds after being off much longer.
With the resistor and cap, measure/monitor it's voltage.
Something is certainly wrong if, with no power to the
system, the resistor is not draining the cap. You are
putting the resistor in parallel, across the cap's leads
rather than in series with it, right?
Skeleton Man
>Something is certainly wrong if, with no power to the
>system, the resistor is not draining the cap. You are
>putting the resistor in parallel, across the cap's leads
>rather than in series with it, right?
Yes that's how I'm doing it.. with power applied it measures about 4.65V, and
with no power it just flucations wildy in the milli-volt range (10mv - 50mv
usually)..
I have tried switching the power off for like half an hour ! and still no go..
only works properly when the cap is not wired in.. (ie. turn on power switch,
then short the leads with the cap).. as soon as I wire it in, it stops working
entirely..
Regards,
Chris