Capacitor discharger
Thomi Richards
I'm looking for some electronics advice:
I have a project that is nearly finished - I'm at the point where I
need to plug it into the wall and test it (carefully) in stages. The
problem is that one of the first stages is the power supply, which
contains some fairly large capacitors. I need a way to safely
discharge them after every power-on, so i can tinker around and fix
things without the risk of electric shock. The advice I have had so
far is to build a discharging tool made from a screwdriver,
high-current resistor, and a length of wire with a clip on the end -
the idea being that one can clip the clip to ground, and poke at the
cap leads to short them out (via the resistor), thereby safely
draining them. However, I'd welcome other advice - Surely such a tool
exists already? I can't see anything like it in tool catalogs, but I'm
probably looking for the wrong thing.
Can anyone offer some advice?
Oh, and I obviously take full responsibility for whatever I end up
doing, in case that was worrying anyone.
Cheers,
Paul Campbell
> I need a way to safely
> discharge them after every power-on, so i can tinker around and fix
> things without the risk of electric shock. The advice I have had so
> far is to build a discharging tool made from a screwdriver,
> high-current resistor, and a length of wire with a clip on the end -
> the idea being that one can clip the clip to ground, and poke at the
> cap leads to short them out (via the resistor), thereby safely
> draining them. However, I'd welcome other advice - Surely such a tool
> exists already? I can't see anything like it in tool catalogs, but I'm
> probably looking for the wrong thing.
(I assume this is for your tube audio amp - so voltages are in the 350v range)
No that's the right tool - I'm sure no one sells them because they are so easy
to make - I'd actually use a wooden stick (or anything that's well insulated)
rather than a screwdriver, just to reduce the number of exposed bits that end up
with evil voltages - you don't want to just ground the caps as the high current
can cause damage to them.
For the same reason a 'high current' resistor is probably not what you want -
you want something that will take them gently down in a few seconds rather than
a few microseconds - probably in the 1k+ range - you can figure out the RC
calculation to get a good value (C is the size of the caps and you're solving
for R), then the power required (size of the resistor) from the initial peak
voltage (from V^2/R) - if the resistor is too big/expensive use a bigger one
and wait longer
Paul
Geoff Barkman
I've built a few power supplies for Ham Radio gear (Fair low voltages
though), and they usually recommend having a bleed resistor across
every big capacitor... perhaps 10 k or higher. If you have a power
available Led (with suitable dropping resistor) across the supply
output it will tell you there is still voltage in the circuit, because
the led will still be glowing for a few seconds to a minute, as the
bleed resistor slowly discharges the capacitors.
Cheers
Geoff
--
Cheers
Geoff Barkman
======================================
Thomi Richards
On 7 March 2011 08:18, Paul Campbell <pa...@taniwha.com> wrote:
>
> (I assume this is for your tube audio amp - so voltages are in the 350v range)
>
Yup, 350 - 400v, assuming I've done everything correctly ;)
> No that's the right tool - I'm sure no one sells them because they are so easy
> to make - I'd actually use a wooden stick (or anything that's well insulated)
> rather than a screwdriver, just to reduce the number of exposed bits that end up
> with evil voltages - you don't want to just ground the caps as the high current
> can cause damage to them.
So, a chopstick would be fine?
>
> For the same reason a 'high current' resistor is probably not what you want -
> you want something that will take them gently down in a few seconds rather than
> a few microseconds - probably in the 1k+ range - you can figure out the RC
> calculation to get a good value (C is the size of the caps and you're solving
> for R), then the power required (size of the resistor) from the initial peak
> voltage (from V^2/R) - if the resistor is too big/expensive use a bigger one
> and wait longer
Awesome, thanks.
I keep meaning to bring this in one Saturday, but haven't had a free
weekend in a long time. I started this project over five years ago.
It's high time I finished it!
Cheers,