Electrolytic and ceramic capacitors cause some distortion at low frequencies when used as coupling caps due to their dielectric absorption.
Also their value changes with the instantaneous voltage across them.
These problems with those types of capacitors causes even-harmonics distortion when the capacitor is operating as a simple high-pass filter. Metalized plastic film caps are best for coupling caps.

An example of dielectric absorption:
1) Charge an electrolytic cap.
2) Discharge it by shorting it for a couple of seconds.
3) Measure its voltage with a high impedance voltmeter. It will show a voltage even though you thought it was discharged. :lol:

__________________
Uncle $crooge

 

Thanks for clarifying the first 2 points but still as far as 3rd question is concerned can you help a bit more.
My understanding of capacitors is that electrolytic capacitors are not used for ac applications and capacitor may blast if it connected wrong way and so the question. but it seems for low impedences electrolytic capacitor may be used for ac as well. Is there any criteria for low impedence?

 

Electrolytic caps are usually polarized so must not have a reverse voltage across them so they don't blow-up!

To couple AC frequencies to a low impedance like a speaker, cheap amplifiers use an electrolytic capacitor. The capacitor always has a DC voltage across it so its voltage never reverses.

Even cheap two-way speakers use non-polarized electrolytic caps to keep the woof out of the tweeter and to keep the tweet out of the woofer.

__________________
Uncle $crooge

 

Capacitors (of ANY kind) essentially pass AC and block DC, it doesn't matter if they are electrolytic or not. All you need to do for an electrolytic coupling capacitor (apart from ignore Audioguru's scare mongering 8) ) is ensure it has a DC potential across it the correct way.

Usually a coupling capacitor in a transistor circuit will connect from the collector of one transistor to the base of the next, both of these points will have a DC voltage on them - you just need to connect the positive of the electrolytic to the most positive of the two points.

There's also nothing particularly bad about speaker coupling capacitors, some of the most highly respected HiFi amplifiers in the world used them, and they still in great demand today!.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Thanks every body for taking pains to explain this point

 

Hee, hee. :lol: Good quality amps and speakers use direct-coupling with dual-polarity supplies or bridged amps and metalized-plastic film coupling caps for their very low distortion. :lol:

In an amp or in a speaker's crossover network, a cheap 6.8uF/100V electrolytic coupling cap costs pence, but a good film one costs a Euro. In addition to causing distortion, a cheap electrolytic cap has a wide tolerance that is unusable in a frequency-determining circuit, and its value reduces as it dries-up or leaks.
Even for my 6V Ultra-bright Chaser project I used two 1uF 5% film caps in parallel so they all spin at the same speed.
Electrolytic caps are fine in a power supply circuit. :lol:

__________________
Uncle $crooge

 

Modern amps usually use those techniques, but as much for cost reasons as quality ones? - older amps used speaker coupling capacitors because the components weren't really available to do direct coupling (at least with decent quality).

The Quad 33/303 system is still a highly sort after amplifier, which uses electrolytics throughout (as did ALL amplifiers of their day), including speaker coupling.

The vast majority (99.99+%) of high quality amplifiers these days will still use electrolytics for internal coupling (but NOT for the speaker), only the really silly priced ones will use expensive special capacitors (which is reflected in their silly prices). The difference is probably up there with oxygen free cable etc. :lol:

But are you really concerned in the difference between 0.01% distortion and 0.005% distortion?. Your posts give the impression that electrolytics make an amplifier so distorted it's unusable?, whereas the reality is it's probably barely even measureable?.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Hi Nigel,
There is quite a debate about whether electrolytic capacitor distortion can be heard. One guy lists many tests using experienced audio people to vote on distortion that they couldn't hear. Then he says his designs try to reduce the distortion:
http://www.angelfire.com/ab3/mjramp/caps.html
"In my own recent amplifier designs the speaker coupling capacitor was included in the overall feedback loop to minimise distortion, and the low measured distortion figures included the effect of this and the other capacitors. Only the input filtering capacitors are expected to add significant distortion if badly chosen, these being outside the feedback loop. Looking at published distortion measurements by Bateman and others there seems general agreement that at 1nF good types are polystyrene and NPO ceramic. The 1uF input coupling capacitor is more of a problem, one of the best is polypropylene, but these are both large and expensive. More convenient in both respects are polyphenylene sulphide. (Available in the UK from Farnell.) My measured distortion was with a polyester 1uF input capacitor. The Bateman tests suggest that these are quite variable, but even his worst distortion level is still very low."

Another guy (not me) says: "Electrolytic capacitors have no place in the high fidelity audio chain anyway. They are nonlinear and create distortion. Their only real purpose is as energy storage devices in power supplies. They can also be used in bypass applications were the effect of their nonlinearity is unimportant."

Like them, I don't like having distortion when I can easily eliminate it. :lol:

__________________
Uncle $crooge

 

Except you don't know if you have distortion or not?, nor do you know if changing the capacitors has really improved things?.

Then of course, the recording you're listening to has most probably been through hundreds of electrolytic coupling capacitors during the recording and mixing process 8)

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Hi Nigel,
I haven't played vinyl records for a while and can't even remember where they are (I kept them somewhere). They were recorded and made through many caps, some probably electrolytics.
I don't think my CDs were recorded through electrolytic coupling caps.

While testing the 1st few versions of my mod-4 FM transmitter, I saw low frequency distortion on my 'scope from an electrolytic cap as a pre-emphasis maker. I guess the circuit's rising frequency response emphasized the distortion. My circuit uses a very low supply voltage and the distortion from electrolytic caps is increased with low bias voltages. When I replaced the electrolytic with a poly one, there wasn't any visible distortion.
10% distortion is difficult to see but plainly heard.
I built only a single one, and didn't try a replacement elco, so maybe Mr. Murphy came by for a visit.

I just finished listening to a series of tones from the web, some pure and others with 1% second harmonic distortion added. The 1st time I played them I was distracted by my wifey asking me to turn it off (I should have used headphones), then each time I repeated it I got a steady tone.

My 1st amp had tooobes (valves) and when I took it to the store for their MacIntosh Free Amp Test, its distortion was about 15%! It crept up as the output tooobes mis-matched themselves and I didn't even notice!

My latest amp is a fairly inexpensive Japanese one (Yamaha), has "High Dynamic Power Low Impedance Drive Capability", very low distortion and only a single coupling cap (poly input) in its DC-coupled design.

I used poly coupling and frequency-determining caps in my bass-EQ circuit. My complete system sounds excellent!

__________________
Uncle $crooge

 

What?, you don't believe mixers have electrolytics?.

I presume you don't seriously believe that you can hear any possible distortion from an electrolytic coupling capacitor?. If you were getting audible distortion you had done something wrong somewhere, like the cap polarised the wrong way and upsetting the bias conditions?.

As you say, 10% is well audible - yet people swear by their high distortion valve amplifers?.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Hee, hee. :lol: I have seen some new very expensive valve amps with the valves only for show. They are powered but not connected in the solid-state circuit. :lol:

__________________
Uncle $crooge

 

That's how some guitar preamps are built as well :lol: just stick in a valve (on the outside so you can see it), wire the heater up so it lights, and charge the dumb ba****ds four times as much money!.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

I've seen new valve amps with the high voltage causing them to appear with a purple light inside. Their plates don't get red-hot and they aren't connected to the solid-state amp's circuitry.

__________________
Uncle $crooge