Searching \ for '[EE] DIY bipolar capacitor? (for signal filtering)' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/logic/dsps.htm?key=filter
Search entire site for: 'DIY bipolar capacitor? (for signal filtering)'.

Exact match. Not showing close matches.
PICList Thread
'[EE] DIY bipolar capacitor? (for signal filtering)'
2011\03\08@050522 by V G

picon face
This is regarding my capacitor crisis.

I want to create filters, but lack large value non-polarized capacitors.

1. I hear that charging a polarized capacitor (electrolytic, for example)
the wrong way can damage it. In this case, what exactly happens? Does the
capacitor just "conduct"? Does it become "damaged"? If so, how?

2. I also hear that you can make a bipolar electrolytic capacitor by
connecting two electrolytic caps back to back.

2.1. But wont this cause one of the capacitors to charge in reverse, thus
"damaging" it?

2.2. Does this mean, in order to make a 10uF bipolar electrolytic cap, I
would just wire two 10uF caps back to back?

3. Are electrolytic caps suitable for filter applications anyway?

4. What about mylar capacitors

2011\03\08@052136 by Richard Prosser

picon face
VG

The capacitors should be connected in series - so two 10uF caps would
result in a 5uF block. Just connect plus to plus OR minus to minus.

Electrolytics can withstand a small voltage reversal and the damaging
factor is DC current flow. So, provided this DC current is blocked by
the other cap, then the block will work OK.

Another alternative, if you have a negative supply available is to run
the opamps (or just the final amplifier) so that the DC output level
is at ground potential. Since the input impedance to an opamp can be
made almost arbitrarily high, any DC blocking requirement can be
carried out using much lower valued capacitors.

RP

On 8 March 2011 23:05, V G <spam_OUTx.solarwind.xTakeThisOuTspamgmail.com> wrote:
{Quote hidden}

>

2011\03\08@071953 by alan.b.pearce

face picon face
> I want to create filters, but lack large value non-polarized capacitors.
>
> 1. I hear that charging a polarized capacitor (electrolytic, for example)
> the wrong way can damage it. In this case, what exactly happens? Does the
> capacitor just "conduct"? Does it become "damaged"? If so, how?

It is likely to go 'bang' in some destructive manner. On a typical electrolytic capacitor with a rubber bung in one end the bung is likely to get blown out violently and much aluminium foil spread over a sizeable area, along with a horrible stink from the decomposed electrolyte.

> 2. I also hear that you can make a bipolar electrolytic capacitor by
> connecting two electrolytic caps back to back.
>
> 2.1. But wont this cause one of the capacitors to charge in reverse, thus
> "damaging" it?

Yes, and maybe, but it works. I suspect that what happens is the one that is correctly biased halts the DC current flow that would cause the electrolyte in the reverse biased one to do things it shouldn't.

> 2.2. Does this mean, in order to make a 10uF bipolar electrolytic cap, I
> would just wire two 10uF caps back to back?

No, the resultant capacity is half the value of the two series capacitors.

> 3. Are electrolytic caps suitable for filter applications anyway?

Not really, they have too large a tolerance, a typical aluminium wet electrolytic capacitor is +100/-50% tolerance.

> 4. What about mylar capacitors?

Mylar capacitors tend to be reasonably close tolerance, and non-polarised low leakage. Certainly suitable for filters, but may be physically large in the uF range.
-- Scanned by iCritical.

2011\03\08@082200 by V G

picon face
On Tue, Mar 8, 2011 at 7:18 AM, <.....alan.b.pearceKILLspamspam@spam@stfc.ac.uk> wrote:
>
>  > 4. What about mylar capacitors?
>
> Mylar capacitors tend to be reasonably close tolerance, and non-polarised
> low leakage. Certainly suitable for filters, but may be physically large in
> the uF range.
>

Whoops! I think I meant to say "tantalum capacitors", which ARE polarized,
and have large uF values. I'm looking here:
http://futurlec.com/CapTant.shtm

2011\03\08@084036 by Olin Lathrop

face picon face
V G wrote:
> 1. I hear that charging a polarized capacitor (electrolytic, for
> example) the wrong way can damage it. In this case, what exactly
> happens? Does the capacitor just "conduct"? Does it become "damaged"?
> If so, how?

First it conducts, which damages it, eventually if you persist, BOOM.

> 2.2. Does this mean, in order to make a 10uF bipolar electrolytic
> cap, I would just wire two 10uF caps back to back?

Definitely not as two 10uF capacitors in series is 5uF total.

> 3. Are electrolytic caps suitable for filter applications anyway?

In some cases.  Probably not in yours though since you're dealing with high
impedence signals.  Electrolytics have relatively high leakage.

> 4. What about mylar capacitors?

The dielectric is mylar, hence the name.  You won't find them in the high
values you are looking for.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000

2011\03\08@085145 by V G

picon face
On Tue, Mar 8, 2011 at 8:40 AM, Olin Lathrop <olin_piclistspamKILLspamembedinc.com>wrote:

> Definitely not as two 10uF capacitors in series is 5uF total.
>

Ah. I thought that the possible shunting nature of one would make its
capacitance irrelevant.


> > 3. Are electrolytic caps suitable for filter applications anyway?
>
> In some cases.  Probably not in yours though since you're dealing with high
> impedence signals.  Electrolytics have relatively high leakage.
>

I'm putting the filters AFTER the low impedance output of an opamp. Or am I
mistaken and the opamp output is high impedance?

> 4. What about mylar capacitors?
>
> The dielectric is mylar, hence the name.  You won't find them in the high
> values you are looking for.


You're right, I meant to say tantalum capacitors

2011\03\08@085353 by Olin Lathrop

face picon face
Richard Prosser wrote:
> The capacitors should be connected in series - so two 10uF caps would
> result in a 5uF block. Just connect plus to plus OR minus to minus.

It's not as simple as that.  You have to make sure the intermediate node
floats at the right voltage.

In most cases, back to back electrolytics isn't worth all the associated
problems.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000

2011\03\08@094447 by Olin Lathrop

face picon face
V G wrote:
> I'm putting the filters AFTER the low impedance output of an opamp.
> Or am I mistaken and the opamp output is high impedance?

Oh, I thought you were putting them right at the input.  If you don't, then
the DC bias will be amplified by the first stage, which may cause clipping
there.  I guess we don't really know what you're trying to do.  Show a
schematic so we can all be on the same page.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000

2011\03\08@094703 by Kerry Wentworth

flavicon
face
You could use a capacitance multiplier.

http://en.wikipedia.org/wiki/Capacitance_multiplier

Kerry


V G wrote:
{Quote hidden}

-- Internal Virus Database is out-of-date.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 267.11.13 - Release Date: 10/6/05

2011\03\08@100550 by alan.b.pearce

face picon face
> You could use a capacitance multiplier.
>
> en.wikipedia.org/wiki/Capacitance_multiplier
>
> Kerry

Good grief, there is a couple of badly laid out schematics for ya' ...
-- Scanned by iCritical.

2011\03\08@103310 by V G

picon face
On Tue, Mar 8, 2011 at 9:45 AM, Olin Lathrop <.....olin_piclistKILLspamspam.....embedinc.com>wrote:

> Oh, I thought you were putting them right at the input.  If you don't, then
> the DC bias will be amplified by the first stage, which may cause clipping
> there.  I guess we don't really know what you're trying to do.  Show a
> schematic so we can all be on the same page.


The DC bias isn't large enough to cause clipping. The gain is low enough so
that everything stays within rail to rail values

2011\03\08@104635 by Herbert Graf

picon face
On Tue, 2011-03-08 at 05:05 -0500, V G wrote:
> This is regarding my capacitor crisis.
>
> I want to create filters, but lack large value non-polarized capacitors.
>
> 1. I hear that charging a polarized capacitor (electrolytic, for example)
> the wrong way can damage it. In this case, what exactly happens? Does the
> capacitor just "conduct"? Does it become "damaged"? If so, how?

Generally the ceiling above the board will have a dent.

I've seen a cap actually embed itself in the suspended ceiling above a
workbench, it's actually quite likely to injure if you happen to have
your head over the board when the cap blows.

TTYL

2011\03\08@105616 by V G

picon face
On Tue, Mar 8, 2011 at 10:46 AM, Herbert Graf <EraseMEhkgrafspam_OUTspamTakeThisOuTgmail.com> wrote:

> Generally the ceiling above the board will have a dent.
>
> I've seen a cap actually embed itself in the suspended ceiling above a
> workbench, it's actually quite likely to injure if you happen to have
> your head over the board when the cap blows.
>
>
How much energy do I have to put in for it to do that? Approx how many
volts

2011\03\08@110537 by Herbert Graf

picon face
On Tue, 2011-03-08 at 10:56 -0500, V G wrote:
> On Tue, Mar 8, 2011 at 10:46 AM, Herbert Graf <hkgrafspamspam_OUTgmail.com> wrote:
>
> > Generally the ceiling above the board will have a dent.
> >
> > I've seen a cap actually embed itself in the suspended ceiling above a
> > workbench, it's actually quite likely to injure if you happen to have
> > your head over the board when the cap blows.
> >
> >
> How much energy do I have to put in for it to do that? Approx how many
> volts?

I can't give you exact figures since this mode of operation is a little
out of spec for the parts at hand...

That said, I BELIEVE it was only a 5V rail (it could have been a 12V
rail), but it was a high power supply, I'd say at least a few amps were
flowing before the cap went.

LEDs can also be VERY spectacular when they fail (by grossly overdriving
them). I remember as a teen blowing up a bunch of LEDs, some would just
fizzle, others would launch pieces of plastic quite a distance.

TTYL

2011\03\08@112007 by alan.b.pearce

face picon face
> > Generally the ceiling above the board will have a dent.
> >
> > I've seen a cap actually embed itself in the suspended ceiling above a
> > workbench, it's actually quite likely to injure if you happen to have
> > your head over the board when the cap blows.
> >
> >
> How much energy do I have to put in for it to do that? Approx how many
> volts?

That is not how it gets measured.

The mechanism is that the current flow converts the electrolyte into gas. The gas build up pressure inside the case. Eventually the case cannot contain the gas pressure, and will do anything from slowly split releasing a stinky smell, to a sudden explosion often caused by the sealing bung coming out of the end. The body of the capacitor will then go flying. If it is constrained by other items in the equipment then the explosive gas stream will generally spray the foil electrodes everywhere.

So it is not a case of how many volts or amps, more the time it takes for the electrolyte to turn to gas, which may be a function of current, and how well the case can withstand the resulting gas pressure.
-- Scanned by iCritical.

2011\03\08@115041 by Michael Watterson

face picon face
On 08/03/2011 15:56, V G wrote:
> On Tue, Mar 8, 2011 at 10:46 AM, Herbert Graf<@spam@hkgrafKILLspamspamgmail.com>  wrote:
>
>> Generally the ceiling above the board will have a dent.
>>
>> I've seen a cap actually embed itself in the suspended ceiling above a
>> workbench, it's actually quite likely to injure if you happen to have
>> your head over the board when the cap blows.
>>
>>
> How much energy do I have to put in for it to do that? Approx how many
> volts?
Basically any electrolytic at full woking volts.

Encapsulated Bead Tantalums smoke, glow bright red and eject pellet.  (there was a batch marked back to front).

110V AC on a 150DC 400uF is ample

or 230V AC on 275V DC 400uF

in older days they did not have indented lines on top for can to split,

In 1970s a BY127 or whatever rectifier failing short on Direct Mains Chassis TV was spectacular. The TV (probably mostly valved) operated but with Horrible Hum bar on picture and massive buzz on Sound. I ducked and bits of aluminium capacitor case embedded in breeze-block wall.
Usually a large can with a common and 3 values in the same can with four tags. Metal clip to chassis

2011\03\10@053355 by cdb

flavicon
face
I haven't read this complete thread,
but why not buy bipolar caps?

Perhaps BGMicro or similar outlet in Canada have surplus.

Here is an Australian outlet, though postage would be dear I would think.
Bipol 10uF/50v AU$0.70

http://www.altronics.com.au/index.asp?area=prod&grp=94

Colin
--
cdb, KILLspamcolinKILLspamspambtech-online.co.uk on 10/03/2011
Web presence: http://www.btech-online.co.uk   Hosted by:  http://www.justhost.com.au
 

2011\03\10@082410 by V G

picon face
On Thu, Mar 10, 2011 at 5:33 AM, cdb <RemoveMEcolinTakeThisOuTspambtech-online.co.uk> wrote:

> I haven't read this complete thread,
> but why not buy bipolar caps?
>
> Perhaps BGMicro or similar outlet in Canada have surplus.
>
> Here is an Australian outlet, though postage would be dear I would think.
> Bipol 10uF/50v AU$0.70
>
> http://www.altronics.com.au/index.asp?area=prod&grp=94


Hard to come by, and expensive when you do find them

2011\03\10@150224 by cdb

flavicon
face
:: Hard to come by, and expensive when you do find them.

BGMicro have Non Polarised 10uF/25v for US$0.10 - do you know anyone in the US or maybe Canada who might want something else that BGMicro have and help with the freight cost.
http://www.bgmicro.com/capsnonpolarized.aspx

What about the  electronics bods at your university or a local electronics clubs if such things are still in existence?

Colin
--
cdb, spamBeGonecolinspamBeGonespambtech-online.co.uk on 11/03/2011
Web presence: http://www.btech-online.co.uk   Hosted by:  http://www.justhost.com.au
 


2011\03\10@152126 by V G

picon face
On Thu, Mar 10, 2011 at 3:02 PM, cdb <TakeThisOuTcolinEraseMEspamspam_OUTbtech-online.co.uk> wrote:

> BGMicro have Non Polarised 10uF/25v for US$0.10 - do you know anyone in the
> US or maybe Canada who might want something else that BGMicro have and help
> with the freight cost.
>
> http://www.bgmicro.com/capsnonpolarized.aspx
>
> What about the  electronics bods at your university or a local electronics
> clubs if such things are still in existence?
>
>
Wow, cheap shipping too. Thanks

More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...