Non-polarized electrolytic cap

"ozzy"

** Yes.

Your app is the normal way NP electros are used -

ie where the polarising voltage changes polarity.

.... Phil

Phil,

Thanks for the reassurance. Just did'nt want to order the wrong component.

Osman

Not so often, actually. Coupling caps are not really being used in AC operation.

BP or NP electrolytic is probably the cheapest solution and that is exactly what they are intended for (occasional reversal of a substantial voltage). You could probably also use a ceramic cap (or two 10uF in parallel) if the voltage isn't too high. 0805 10uF/10V run less than 4 cents each in volume.

A non-polarized one, no; they have no "forward" or "reverse" voltage.

An ordinary electrolytic, Yes, even if it's just a very short period of time - they blow up, literally.

Use a non-polar.

Hope This Helps! Rich

They have both and they are equal.

You can safely use Sprague TE series up to about half rated voltage in reverse DC bias condition, and about that much AC, low currents. Or safely use any standard non-polarized 'lytic at half label rating.

An NP EL is TWO caps in series opposed configuration, in the same package.

One cap operates shorted as the other takes charge and if the polarity is reversed, the other cap shorts and the partner charges. So any "reversed voltage" is impossible because the cap is made such that either excitation gets handled by a properly polarized cap.

Not really. It looks almost exactly like a regular aluminum electrolytic in construction, except that it has oxide layer and etched roughness that are similar on both both plates rather than just the anode.

A regular polarized cap has only the natural oxide layer thickness on the cathode (maybe 2nm thick), so it can only withstand a small reverse voltage.

All "polarized" electrolytics are bipolar, just quite asymmetrical in their voltage capability.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

It IS PHYSICALLY TWO EL caps in series opposed configuration. The way it gets wrapped causes it, and the result is EXACTLY what I stated. TWO SERIES OPPOSED CAPS in the same package.

The outer appearance of the can has not a goddamned thing to do with it, you dingledorf. They are typically twice the size of the same value polarized version.

They also have TWICE the wraps.

Now, do you need you hand held further?

There is exactly twice as many wraps in it. Maybe even a plus one on one side or the other. The result is exactly what I said it is.

"Wraps"??? You really don't understand, do you?

ALL aluminum electrolytic capacitors are effectively two capacitors in series.

There are two aluminum plates, each with an oxide layer, separated by paper soaked in a resistive electrolyte solution. One or both may be thicker and deeply etched prior to anodizing in order to increase the effective surface area by a factor of maybe 50 over the area of the overall xy dimensions.

The actual dielectric of each capacitor is the thin layer of AL2O3; the electrolyte is just a conductor.

A polar capacitor has only a very thin layer on one side- maybe 2nm, which results in a very high capacitance (recall that capacitance is proportional to the dielectric constant and inversely proportional to the thickness ofnthe dielectric) and not much voltage capability (the reverse voltage capability of the polar cap). The thickness of the layer on the anode is dependent on the intended voltage rating of the capacitor- Nichicon says typically 1.4nm/volt (dielectric constant of Al2O3 is about 9). So a 25V cap would have about 35nm of oxide.

The total capacitance is the series combination of the two capacitors.

A nonpolar cap has the same etching and oxide thickness on both plates.

POLAR CAP (voltage rating +10VDC (and maybe -1 VDC)

Cathode Anode Electrolyte

|//| ___ |/////| -|//|--|___|--|/////|- |//| |/////|

C1 C2

Ceq = C1*C2/(C1+C2) ~= C2, since C1>>C2

NON-POLAR CAP (voltage rating +/-10VDC)

Plate A Plate B Electrolyte

|/////| ___ |/////| -|/////|--|___|--|/////|- |/////| |/////|

C1 C2

Ceq = 1/(1/C1 + 1/C2) ~= C1/2 ~= C2/2, since C1 ~= C2

Obviously, the volume of the non-polar cap will have to be about double for the same capacitance, all other things being equal, since the voltage capability of the dielectric is proportional to the thickness of the dielectric.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Nope. RAW number of WRAPS, dumbfuck. Half of which run the other way.

Spehro, it is hopeless to try to educate AlwaysWrong. If you disagree with him in any way, it is like nectar to a troll.

Of course not. He's AlwaysWrong.

Hope This Helps! Rich

You can't educate pork ;-)

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)

You misspelled 'Dork'.

--
You can't have a sense of humor, if you have no sense.

So, as a rule of thumb, can polarized caps be used at some small percentage of their rated voltage in an AC circuit? Say a 35 volt cap operating at 5 volts AC?

I've noticed sometimes 2 polarized caps are used back to back for AC operation and sometimes diodes are used to bypass the reverse voltage on both caps. What are your thoughts about diodes and back to back polarized caps?

-Bill

"The Journey is the reward"

eff.com

I wouldn't use it over half a volt, but then I'm rather conservative, particularly when it comes to things that could fail in the field. Nor would I put AC on it, just occasionally reversing DC. I have done this quite a bit and have never had any problems.

Diodes seem safer. I've never had to do this, so have not given it a lot of thought. There are different scenarios.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

"Bill Bowden"

So, as a rule of thumb, can polarized caps be used at some small percentage of their rated voltage in an AC circuit? Say a 35 volt cap operating at 5 volts AC?

** No way.

The AC voltage across a regular electro must not be more than 1 volt.

Polarised electros are regularly use in audio circuits with no DC voltage - but sized so that there is very little signal voltage ACROSS the cap.

Their purpose is to remove offsets at the inputs and outputs of op-amp stages so that pots and switches can operate silently.

I've noticed sometimes 2 polarized caps are used back to back for AC operation and sometimes diodes are used to bypass the reverse voltage on both caps. What are your thoughts about diodes and back to back polarized caps?

** That trick is done where one wants to remove a DC offset of maybe 0.5 to 1 volt from the AC supply - back to back electros alone would do the job, but for the massive inrush surge.

... Phil

The way I heard it, the two 'lytics in series-opposing is the way to go, because in the reverse direction the cap acts like a diode. The thing that saves it from blowing up is the fact that there's another cap in series with it so nothing more than its charging current flows.

I've contemplated diodes, and they certainly couldn't do any harm, but I've never really heard of a situation where they've been necessary.

And plus, who was it that said that a non-polar 'lytic essentially has two cathodes?

Cheers! Rich