1uf Elec Vs 1uf Metal film

[markm]

When it comes to large-value coupling caps, for me, it's either film or non-polar/bi-polar electro ONLY. They're so cheap, why even bother messing with anything else? And I think it's a grievous sin to ever use a polarized cap for coupling, unless it's two together back-to-back because you didn't have a bipolar available. I mean c'mon.... it's not like cost is any kind of an issue here.

Out of curiousity.....why?

[MKB]

It really does depend on the application.  After years doing hi-fi stuff, it was drilled in our heads that the worst coupling cap on the planet was a tantalum.  But building a "modded VLOD" to botique specs (with tantalums in coupling spots) really spun my head around on that one.

I was working on a guitar amp that came with electrolytics as cathode bypasses, and did the obvious tweak of replacing them with film caps.  In this application (a Yamaha T100C), the film caps hurt the sound.  After much headscratching, and putting the electros back in as a last resort, the tone came back.  So I'm constantly surprised at what cap types sound like in equipment.

One other thing for sure, the simpler an effect is, the more you'll hear the tone changes of the individual parts.  So it's best when tweaking to try different types to see what effects they have.  One fun experiment is to get all the TI TL0X2 IC's you can, and swap them out in a dual op amp effect.  They all sound different.

[soulsonic]

When it comes to large-value coupling caps, for me, it's either film or non-polar/bi-polar electro ONLY. They're so cheap, why even bother messing with anything else? And I think it's a grievous sin to ever use a polarized cap for coupling, unless it's two together back-to-back because you didn't have a bipolar available. I mean c'mon.... it's not like cost is any kind of an issue here.

Out of curiousity.....why?

Distortion. A polarized cap distorts the signal because it's designed to pass a voltage that has a certain polarity, and an audio signal has both polarities, so half the signal is not going through as it should - the sound still gets through, but it's been distorted from it's original form. Same goes with tantalums and other polarized components. I prefer my distortion to come the signal being overdriven, not from the signal being harmed by inappropriate components.

Here is a great article which clearly shows how the different caps affect a signal.
http://members.aol.com/sbench102/caps.html
There are links in the article to separate pages with curves for polarized electro/tantalum caps and also comparing different ceramic caps including those really awful monolithic "bead" caps.

[markm]

All very interesting indeed but, this still leaves me convinced this polarized cap = "bad" is only really audible to incredibly finicky
listeners under "bedroom level" playing conditions.
Seems under real-world playing conditions it would be highly unlikely that anyone could really tell the difference.

[brett]

A polarized cap distorts the signal because it's designed to pass a voltage that has a certain polarity, and an audio signal has both polarities, so half the signal is not going through as it should

This is incorrect because many (most?) coupling caps have a DC bias that ensures they work correctly. e.g. the input cap of a fuzzface has DC ground on one side and the DC base bias for Q1 on the other.

All very interesting indeed but, this still leaves me convinced this polarized cap = "bad" is only really audible to incredibly finicky
listeners under "bedroom level" playing conditions.

You may be greatly over-estimating the effect to think that it is even close to audible, even in the silence of deep space.  There was a topic here a year or so ago that had clips of: (i) pure sine waves, and (ii) sine waves with distortion (1% thd, I think).  Nobody could hear any differences.

cheers

[analogmike]

I'll repeat myself, not only the type but the brand and model of a cap makes a difference. Some electros sound terrible and some (not always expensive!) sound great.

When it comes to large-value coupling caps, for me, it's either film or non-polar/bi-polar electro ONLY. They're so cheap, why even bother messing with anything else? And I think it's a grievous sin to ever use a polarized cap for coupling, unless it's two together back-to-back because you didn't have a bipolar available. I mean c'mon.... it's not like cost is any kind of an issue here.

if the circuit has voltage on one side (most do) why not use polarized? The NP caps are simply two of the same thing back to back, why would 2 sound better than one?

[Ronsonic]

Some schmuck named Rupert Neve built a lot of stuff using tantalum coupling caps and most of his gear still works and is generally considered to sound good.

It is much more about how you use it.

I remember reading a white paper from Sprague years ago where the author pointed out that people have been inventing new types of capacitors since the time of Ben Franklin - most of them intended to replace an older type and still every type of cap ever invented remains in production and in use for one purpose or another. It is about horses for courses and knowing how to use what is available, not about one being "better" than another.

Ron

[RedHouse]

I've used both and in listening tests I feel IMHO the Film are noticeably better.

Not like a night-n-day contrast, just a little better.

Here's a pic of some of my Univibe boards with both types:

[soulsonic]

It takes alot more than a couple volts of bias to smooth out the polarity issues of a polarized cap.... Bench's article shows this clearly, but whatever, it's not like it matters anyway. It doesn't make any difference to me if someone uses Nichicon Muse polarized caps for coupling or Kemet tantalums. What an individual uses based on personal taste is certainly not something worth arguing over. I use what I use for the reasons I have and it makes what I build special to me and my customers. The sound and feel of what I build is something many musicians find very pleasing, and that sound is the result of both the design and the component choices I have made. I don't care if you don't want the same sound I have - I prefer that you didn't anyway.

[markm]

I've used both and in listening tests I feel IMHO the Film are noticeably better.
Not like a night-n-day contrast, just a little better.

There's that term again......"better".   
Never an explaination of how it's better, or what is better.....just "better". 
BTW, why so many Univibes?

[RedHouse]

I've used both and in listening tests I feel IMHO the Film are noticeably better.
Not like a night-n-day contrast, just a little better.

There's that term again......"better".   
Never an explaination of how it's better, or what is better.....just "better". 
BTW, why so many Univibes?

Some of them were for blind testing the audible diference between caps, that's how I can say (IMO) they sound "better". Others were for clients.

My friend and I find we also don't see things the same when folks try to vebalize the qualities of sound, so we do what we call "Pepsi Challenge" tests where we build similar thinks with only the tested components differing, this gives us a better ear for the difference. I have a amplifier chassis with tons of switches added to test different mods too.
(too much time elapses when one tries to solder in the differnt parts for ear testing)

[brett]

blind testing

IMO blind testing is essential. 

Why? My experience (incl a PhD) tell me that the almost all subjective statements in the forum are about what people WANT, HAVE BEEN TOLD, NEED or IMAGINE to be true. Blind testing takes these options away.

Many people, when blind testing, are desperate for "anchoring" or "reference" information, and often have low skill at picking differences because they think along lines such as "this could be an electrolytic cap, because people say that they sound like this, so I'll say it sounds bad".  Research shows that most people, including experts, are overconfident about their ability to detect differences and changes.

A handy book on this is "The psychology of judgment and decision-making" by Plous. (If you gamble, it's a money-earner)

[Michael Weidenauer]

I agee with analogmike that not only the type, but also the brand or voltage rating makes a difference.
Sometimes it's hard to tell if you are fooled by your ears or if there is a real difference, so i did some experiments, measured different types of caps to get the same value and built pedals where i could switch between two differnt brands (or types) via footswich. Sometimes there was quite a differnce, sometimes not.

In some applications I like to put small mica-caps (or sometimes ceramics) parallel to film- or electrolytic-caps and sometimes it really doesn't matter. You have to try!

Another thing to try (an old tube engineer once told me): Use 10V or 25V Sprague electrolytics as cathode bypass in your tube amp and you'll notice that lower voltage ratings (same company!) will give you more compression than higer ones. Don't know why - but it works. Haven't figured out what that means for 9V pedals yet.

I still like WIMA film-caps best (small, not too expensive and good sound)!

[Plinky]

All very interesting indeed but, this still leaves me convinced this polarized cap = "bad" is only really audible to incredibly finicky
listeners under "bedroom level" playing conditions.
Seems under real-world playing conditions it would be highly unlikely that anyone could really tell the difference.

What he said. Most people in a bar are more worried about their next drink or their next...you get the idea.

[brett]

Hi

Use 10V or 25V Sprague electrolytics as cathode bypass in your tube amp and you'll notice that lower voltage ratings (same company!) will give you more compression than higer ones. Don't know why - but it works.

This might be due to the ESR or capacitance changing as a % of the maximum voltage of the cap.  Sometimes in the forum it has been suggetsed to avoid using 50V and 63V electros in 9V pedals because the % of maximum voltage is too low.

I'll do some research and find out.
cheers

[Michael Weidenauer]

Use 10V or 25V Sprague electrolytics as cathode bypass in your tube amp and you'll notice that lower voltage ratings (same company!) will give you more compression than higer ones. Don't know why - but it works.

This might be due to the ESR or capacitance changing as a % of the maximum voltage of the cap.  Sometimes in the forum it has been suggetsed to avoid using 50V and 63V electros in 9V pedals because the % of maximum voltage is too low.

For my last FF-clone (wonderful sound wit two OC139's) I've tried loads of 22uF electros (I once built a Test-Face with sockets for all parts ) and a quite big one with a 100V rating sounded best. In most builds I use modern sub-miniatur electros with a rating of 63V and they work well.

This leads me to another question:
Is there any easy method for testing old electros if they are dried out (my DMM only works up to 20uF)
I've got so many of them and rarely dare to use them because I fear they might spoil the build.

[MKB]

I've read in an article by a Japanese audio guru (forgot who it was) that electrolytics sound much better when you bias them up as hot as possible.  But you have to be careful that the peak AC superimposed on the bias doesn't go over the peak cap voltage.  The leakage of the cap goes WAY up and the cap heats up when you do that.  When the cap heats up the leakage goes up, thus more heating.  Thermal runaway.

I wonder if you could make a clever clipping circuit by overbiasing an electrolytic cap; might clip very softly.  Hmmmm.....

Finally, I just replaced the Sprague polypropylene and foil caps in my stereo with Russian Teflon film caps off ebay.  BIG BIG BIG HUGE difference.  Staggering.  I doubt that they'd help pedals very much though.

[soulsonic]

I've always heard that Teflon caps are something special...

[Mark Hammer]

Finally, I just replaced the Sprague polypropylene and foil caps in my stereo with Russian Teflon film caps off ebay.  BIG BIG BIG HUGE difference.  Staggering.  I doubt that they'd help pedals very much though.

Said it before.  I'll say it again.  What makes a difference with reproduction of recorded music (that normally incorporates multiple sound sources mixed down to stereo, hence much wider bandwidth and dynamic properties) can be very different than what makes a difference with music production, where the signal path of musician/instrument/processing/amplilfier generally consists of a single instrument with moderate dynamics and restricted bandwidth.  One is naive to confuse the one context with the other.  It's like thinking that whatever changes you make to your tires on a Mazda Miata would make a similar improvement in performance and handling in an 18-wheeler.

[mac]

I avoid electros whenever it's possible:

ESR, wide tolerance, high freqs, aging.

IMHO polys sound better. My GE FF has only 2 x 6.8uf polys bypassing the drive pot, and they sound gainy, bassier and richer than any good quality 22uf electro. The input and output caps are also polys. But this is my personal taste.

Since this is a very subjective topic my advice is, buy polys, ceramics, tantalums and electros (panasonic, sprague, nichicon, etc, avoid cheap chinese), a breadboard, a cap tester ... and have fun changing caps here and there.  And do a blind test.

mac