From: dplatt@radagast.org (Dave Platt)
Newsgroups: sci.electronics.components
Subject: Re: unpolarized capacitor
Date: Wed, 03 Jan 2001 06:57:56 -0000

>    An old radio magazine said one could make a unpoloarized
>electrolytic capacitor by connecting two negatives or positives of
>polarized capacitor in series.. Don't remember if it was neg. or pos.
>though..

Doesn't matter whether you tie positive to positive, or negative to
negative.

>The circuit will be wired as :
>
>---||----||----
>   + -   - +
>
>I don't know how this works, because whenever you try to charge this you
>will be applying reverse voltage on one of the capacitor.. Does this
>really work and how??

It really does work, mostly.  It works because a polar electrolytic
cap with reverse voltage across it behaves a good deal like a short
circuit (as many people have found out the hard way by accidentally
hooking one up backwards, and seeing their circuit overheat and/or
go FOOM in a most impressive fashion).

Referring to your diagram above - if the input at the left side is
positive with result to the input at the right, then the left-hand
capacitor will be forward-biased (and will behave like a good
capacitor should) and the right-hand capacitor will be reverse-biased
and will behave pretty much like a short circuit (or a small-value
resistor with some miscellaneous capacitance and inductance thrown
in).  Reverse the polarity of the voltage being applied to this
composite, and the two capacitors switch roles - the right-hand one
becomes forward-biased and behaves like a cap, and the left-hand one
becomes negatively biased and acts like a chunk of wire.  In effect,
each capacitor in this composite is called upon to be a capacitor only
during half of the waveform.

The approximation that "a reverse-biased electrolytic behaves like a
short circuit" isn't a terribly good one... it may have some nonlinear
behavior, I gather.  Some people prefer to put a rectifier diode in
parallel with each capacitor, oriented so that the capacitor is never
reverse-biased by more than the forward voltage drop of the rectifier.
This can improve the circuit's approximation of a "real" nonpolar
capacitor, and reduces voltage and current stress on the caps.

>                    The capacity will halve and voltage rating will
>double just like connecting them in series the regular way right?

Nope.  Capacity and voltage rating remain unchanged in this case.

--
Dave Platt                                           dplatt@radagast.org
Visit the Jade Warrior home page:  http://www.radagast.org/jade-warrior/
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

From: dplatt@radagast.org (Dave Platt)
Newsgroups: sci.electronics.components
Subject: Re: unpolarized capacitor
Date: Fri, 05 Jan 2001 23:37:34 -0000

> Wouldn't inrush current cause one of the capacitor to blow up?

The inrush current (and the effect thereof) into the forward-biased
cap is would be the same as if it were hooked up to a DC supply
of the same voltage.

The inrush current probably has relatively little effect on the
backwards-biased half of the pair, because it's behaving much like a
near-short-circuit - the voltage across it remains low due to its low
ESR in this direction, and as a result not much power is dissipated in
it.

>                                                          What about
>rapid discharge(i.e. short circuit)?

Same issue, I believe.

In general, I'd expect the ability of a pair of caps hooked up in this
way to withstand inrush/outrush/ripple currents to be very little
different from a single cap of the same type, which was handling an
equivalent ripple current but never being reverse-biased.  Maybe a bit
less, but not (I imagine) too much less.

In general, electrolytic caps are _not_ designed or rated to handle
the sort of high ripple current levels you get if you hook them onto
the mains!  PFC-correction caps are of a special construction which
allows them to survive the high ripple currents, and the heat which
can build up as a result.

>                                  Can unpolarized electrolytic capacitor
>hooked up this way used for heavy duty cycle where charging and recharging is
>very fast such as power factor correction?

I have no real idea.  I wouldn't try it myself!

The most common application I've seen for "nonpolar" capacitors of
this composite variety is in loudspeaker crossovers (in the midrange
or tweeter high-pass section) and they're not usually handling huge
currents.

>    I am a bit unconfident with capacitors ever since an accident with huge
>capacitor.  This thing weighed like a pound and was like twice the size of
>pop can.   I bought two identical capacitors and one didn't have any label so
>I thought something as big as this could take major voltage and probably
>unpolarized.   I didn't bother reading the label on other capacitor either.
>I hooked it up to 240V mains to see what it does. This thing didn't just pop
>with "foom" , but went KA BOOM!!  Debris were everywhere in laundry room and
>I could swear I was deaf for few seconds following the boom.  Later I read
>the label on the other one and it read 500,000µF 20V DC and this revealed I
>have not only exceeded voltage rating only by 220, but also hooked a
>polarized cap to AC.  The large capacity explains the size... duh... Damn I
>got yelled so much for that.

Lucky you didn't lose a hand or an eye!  You might as well have
dropped a crowbar on the mains!

Yeah, a general _very_ good rule is "Never hook up to the AC mains, or
to DC with a high ripple-current factor, any capacitor which is not
specifically designed and rated for that application."  Violating this
rule can result in surprising, injurious, or lethal consequences!  The
mains fuse or circuit breaker will almost certainly _not_ operate
quickly enough to protect you!

--
Dave Platt                                           dplatt@radagast.org
Visit the Jade Warrior home page:  http://www.radagast.org/jade-warrior/
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

From: dplatt@radagast.org (Dave Platt)
Newsgroups: sci.electronics.components
Subject: Re: unpolarized capacitor
Date: Fri, 05 Jan 2001 23:37:34 -0000

> Wouldn't inrush current cause one of the capacitor to blow up?

The inrush current (and the effect thereof) into the forward-biased
cap is would be the same as if it were hooked up to a DC supply
of the same voltage.

The inrush current probably has relatively little effect on the
backwards-biased half of the pair, because it's behaving much like a
near-short-circuit - the voltage across it remains low due to its low
ESR in this direction, and as a result not much power is dissipated in
it.

>                                                          What about
>rapid discharge(i.e. short circuit)?

Same issue, I believe.

In general, I'd expect the ability of a pair of caps hooked up in this
way to withstand inrush/outrush/ripple currents to be very little
different from a single cap of the same type, which was handling an
equivalent ripple current but never being reverse-biased.  Maybe a bit
less, but not (I imagine) too much less.

In general, electrolytic caps are _not_ designed or rated to handle
the sort of high ripple current levels you get if you hook them onto
the mains!  PFC-correction caps are of a special construction which
allows them to survive the high ripple currents, and the heat which
can build up as a result.

>                                  Can unpolarized electrolytic capacitor
>hooked up this way used for heavy duty cycle where charging and recharging is
>very fast such as power factor correction?

I have no real idea.  I wouldn't try it myself!

The most common application I've seen for "nonpolar" capacitors of
this composite variety is in loudspeaker crossovers (in the midrange
or tweeter high-pass section) and they're not usually handling huge
currents.

>    I am a bit unconfident with capacitors ever since an accident with huge
>capacitor.  This thing weighed like a pound and was like twice the size of
>pop can.   I bought two identical capacitors and one didn't have any label so
>I thought something as big as this could take major voltage and probably
>unpolarized.   I didn't bother reading the label on other capacitor either.
>I hooked it up to 240V mains to see what it does. This thing didn't just pop
>with "foom" , but went KA BOOM!!  Debris were everywhere in laundry room and
>I could swear I was deaf for few seconds following the boom.  Later I read
>the label on the other one and it read 500,000µF 20V DC and this revealed I
>have not only exceeded voltage rating only by 220, but also hooked a
>polarized cap to AC.  The large capacity explains the size... duh... Damn I
>got yelled so much for that.

Lucky you didn't lose a hand or an eye!  You might as well have
dropped a crowbar on the mains!

Yeah, a general _very_ good rule is "Never hook up to the AC mains, or
to DC with a high ripple-current factor, any capacitor which is not
specifically designed and rated for that application."  Violating this
rule can result in surprising, injurious, or lethal consequences!  The
mains fuse or circuit breaker will almost certainly _not_ operate
quickly enough to protect you!

--
Dave Platt                                           dplatt@radagast.org
Visit the Jade Warrior home page:  http://www.radagast.org/jade-warrior/
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!