Skip to main content

Topic: Please, check these electrical components out (Read 3492 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • hlloyge
  • [*][*][*][*][*]
Please, check these electrical components out
Hi-Fi capacitors and resistors that "have sound".
Do you know anyone who uses these trinkets in making crossovers or amplifiers?

http://www.eng.sup-audio.com/html/high_components.html

This is my "find of a month".
  • Last Edit: 15 February, 2010, 01:22:30 PM by greynol

  • odigg
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #1
I don't know about these parts but it's a quite a common belief that "audiophile" capacitors can improve sound. This is speaking in terms beyond electronic theory (e.g. a low impedance load coupled with certain capacitors can lead to a high pass filter).

I've seen some enormous "black gate" capacitors put on sound cards, the EMU 0404, etc.  Of course the only way to do this is to have the capacitors hanging off the board...

Here is an example http://img505.imageshack.us/i/img1501hu4.jpg/
  • Last Edit: 13 February, 2010, 09:43:01 AM by odigg

  • hlloyge
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #2
 oh-my-god.

  • DVDdoug
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #3
I don't know how any of the following affects sound quality...  It's based on my knowledge of electronics.

The most ecomonomical high value (high microfarad) capacitors are electrolytic capacitors, and high-value capacitors are needed for crossovers.  Electrolytics do have some characteristics that make them non-ideal for use in crossovers.  For starters, they don't have tight tolerances and their values change as they age*.  (I'm pretty sure the capacitors in my crossovers are electrolytics.)

Any other type of capacitor will "measure better". But,  I don't think you'd find any measurable difference between any of the various non-electrolytics (in the audio range)...  It's electrolytics vs. "everything else".    I don't find any reason to use "audiophile grade" capacitors.  I'd just look at the value, tolerance and voltage rating.

Sometimes, a (lower value) non-electrolytic is wired in parallel with the electrolytic to compensate for some of it's other undesirable characteristics.  (It doesn't help with tolerance issues or value-drift due to aging.)  This is an easy "mod" you can make to a crossover...  But, I'd be surprised if it makes an audible difference.

With coils, there are trade-offs between air core and iron/ferrite core inductors.  Air core inductors are more linear, but they require more "turns" to get the same inductance.  More turns means more series resistance, and it means heavier gauge wire to overcome the extra resistance.  (I don't think you can fully compensate with bigger wire.)    I suppose the "best solution" is to use an air core inductor, and then use a bigger amplifier to compensate for the loss due to series resistance...

BTW - You can avoid all these crossover issues/compromises with an electronic crossover and bi-amplification/tri-amplification. 


* For most audio applications capacitor value or "quality" is not critical.  They are typically used as power supply filters, or to pass the audio signal while blocking DC.  The capacitor just has to be "big enough" to do the job, and the designer can use a value that's 2x or 5x what's needed so that tolerance and aging are not  issues.  Except for passive crossovers (where large values are needed), electrolytics are never used in filter (or equalizer) circuits where the value is related to frequency. 
   


  • hlloyge
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #4
DVDdoug, thank you for explanation. If I understood correctly, when making "passive" crossover, it is preferable to use high quality electrolytic capacitors, with low tolerance value - and buy more expensive ones just to be sure of manufacturing quality and long lasting...?
And when making "active crossover", you can circumvent these bad sides of passive crossovers? but then you have to do biamping by default, don't you?
I am asking because I find the idea of buying audiophile grade electronic components ridiculous, but I suspected there is some theory behind that idea, as I am not an electronics guy, more electric by my education

  • 2Bdecided
  • [*][*][*][*][*]
  • Developer
Please, check these electrical components out
Reply #5
This is a page from a standard electronics text book - nothing to do with audiophiles...

http://books.google.co.uk/books?id=bkOMDgw...;q=&f=false

...check out the different types of capacitor listed down the left, and the comments about accuracy, stability etc down the middle and right.

The authors aren't fans of electrolytic capacitors!


As with most electronics: skill, care, knowledge and most importantly good design are the key to good performance. You design a circuit so that the parts of the circuit which define its performance are made of high quality, stable components - and any low quality components are used in roles where their faults can have little effect on circuit performance.

Taken to its logical extreme, this means designing audio circuits so that there are no capacitors in the direct audio path. It can be done. It's not expensive.


However, if you want to make some audiophile bling, you put lots of capacitors in there, and whether they're important or not, you use big expensive looking ones!


Of course there are plenty of "good enough" designs which sit somewhere between these two approaches.

FWIW, in extremely cheap modern electronics, it's usually very low quality capacitors which cause equipment to fail within a year. In price sensitive markets, eastern manufacturers are using capacitors which simply aren't up to the job. I'm not talking about "low quality" as in "it doesn't sound that good" - I mean "low quality" as in "is goes BANG after 6 months!".

Cheers,
David.

Please, check these electrical components out
Reply #6
Hi-Fi capacitors and resistors that "have sound".
Do you know anyone who uses these trinkets in making crossovers or amplifiers?

http://www.eng.sup-audio.com/html/high_components.html

This is my "find of a month".


I've designed, built and repaired crossovers ever since I was 13, but I've never bothered to waste money on audio jewlery like that.  It is for people with more money than brains.
  • Last Edit: 15 February, 2010, 07:23:40 AM by Arnold B. Krueger

Please, check these electrical components out
Reply #7
DVDdoug, thank you for explanation. If I understood correctly, when making "passive" crossover, it is preferable to use high quality electrolytic capacitors, with low tolerance value - and buy more expensive ones just to be sure of manufacturing quality and long lasting...?


Actually, it is always best to avoid electrolytic capacitors, as they are more like necessary evils than anybody's component of choice. They are among the electronic parts with poor tolerances, poor stability, and proor other characteristics, such as extraneous series resistance, inductance and parallel parasitic capacitance that effectively has resistors in series with it. The only reason that anybody bothers with them is that they are relatvely small for their voltage rating and capacitance.

The size issue is a big one, because if you ever see a component where the standard application of electrolytics has been replaced with other kinds of capacitors, the capacitors become huge, bulky, and very expensive.

Despite all the nasty things I said about them, electrolytic capactiros are often the best solution for many situations. And they have become highly perfected. There are standard guidelines for the use of the various types of capacitors, and if you follow them,you end up with a product that at least has no serious faults from the standpoint of capacitors.

Quote
And when making "active crossover", you can circumvent these bad sides of passive crossovers?


Yes, First and foremost, there is a world of techniques that you might want to apply to a crossover that are impractical unless done with so-called active circuitry. In this day and age, any really no-hold-barred approach to building a loudspeaker crossover will involve digital signal processing. Only cost and engineering time keeps all loudspeakers from being built with crossovers that run in the digital domain. Can anybody even imagine a digital crossover implemented 100% passivly?

Quote
but then you have to do biamping by default, don't you?


That's both the good news and the bad news. Active crossovers pretty much put you in the business of using one power amplfiier per driver type. In general there are economies of scale with building power amps, and the cheapest way to get to a certain power level usually involves minimizing the number of power amps.  The counterpoint is that there are also economies associated with putting the whole power amp (or several of them) onto one chip, and there seem to be practical limits to the size of good power amps that can be built this way.

It doesn't stop with biamping. Thursday night I was listening to some very expensive speakers that are primarily sold as do-it-yourself projects. The full enchelda runs about $10K.  The designer is a very smart guy who retired from a large electronics firm, and who has a very elegant and well-documented web site. I am told that one pair of speakers requires 10 channels of power amplification with the concept of analog active crossovers taken a goodly distance down the road. I believe I saw (and listened to) an 8 channel amp, and a 2 channel amp and a well-filled box of op amps, resistors, capacitors, and the like.

Another example are some of the very technical modern speaker arrays used in live sound.  There are two different technologies - one which is composed of a stack of identical speaker modules which each often have at least 2 channels of amplification, and another approach where there is a very large and complex DSP array that drives several dozen smaller drivers, each with its own power amp to create a synthetic loudspeaker with very specific but diverse directivity characteristics.  In either case there can easily be dozens of power amplifiers for in a single speaker array.

Quote
I am asking because I find the idea of buying audiophile grade electronic components ridiculous, but I suspected there is some theory behind that idea, as I am not an electronics guy, more electric by my education


I see the primary theory behind audio jewelry as being more psychology and sociology than electronics. Yes, simple components like capacitors, resistors, and inductors have technical properties that need to be chosen if a product is to work well and be economical and reliable. However, you don't need hyper-expensive components to do the job well. Diminishing returns are all over the place. There are many places where things like resistor and capacitor nominal values don't mean much, and a wide range of values can work well. There are places where 1% and even 0.1% parts can make sense. However, putting a 0.1% part where a 10% will do doesn't necessarily make the equipment 100 times better, or even make it perform any different.

Most places where high-tolerance parts would actually make a difference are disappearing into the digital domain where parameters can be controlled to an extent that is pretty much impossible to do in the analog domain for any price.

  • odigg
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #8
DVDdoug, thank you for explanation. If I understood correctly, when making "passive" crossover, it is preferable to use high quality electrolytic capacitors, with low tolerance value - and buy more expensive ones just to be sure of manufacturing quality and long lasting...?


There is most certainly good reasons for using good quality capacitors.  I've seen (as 2Bdecided as pointed out) capacitors burst (via leaks, not explosions)  in cheap DVD players, TVs, and other electronics right after they were out of warranty.  But there is a BIG difference between quality and "PUT THE BIGGEST CAP YOU CAN FIND" like in that EMU 0404 USB.

Perhaps what's astonishing is that even *some* audio engineers support this.  I once met a respected (in the audiophile world) amp maker who uses measurements, but makes final decisions by "listening" to the various parts and making selections accordingly.  I was just shocked that somebody solders in a capacitor, listens, then swaps in another one, etc, and listens.

I agree with Arnold when he says so much of this is just psychology and sociology.  When you buy a really expensive piece of eqiupment like a $500 headphone amp that has no functions other than amping and a gain switch, you have to use "audiophile" grade parts in there just to justify the prices and the "audiophile" moniker on the eqiupment.
  • Last Edit: 15 February, 2010, 12:31:49 PM by odigg

  • rpp3po
  • [*][*][*][*][*]
  • Developer
Please, check these electrical components out
Reply #9
Can anybody even imagine a digital crossover implemented 100% passivly?


Probably using memristors. They should allow you to emulate a transistor's switching capabilities (without the amplification). So basic discrete logic circuits should be possible, thus also ADCs, digital filters, and DACs. This would possibly be a nightmare to implement but you just asked about imaginability...

Please, check these electrical components out
Reply #10
I agree with Arnold when he says so much of this is just psychology and sociology.  When you buy a really expensive piece of eqiupment like a $500 headphone amp that has no functions other than amping and a gain switch, you have to use "audiophile" grade parts in there just to justify the prices and the "audiophile" moniker on the eqiupment.


To expand on this, I think that anxiety drives a lot of this. People do a project and stake a lot of their ego on its sucessful outcome. Of all the things audio to build, a loudspeaker has to be the trickiest play around because of its interactions with the listening room. IOW, you can build a great speaker but if it is the wrong speaker for the room, it will never sound right in it. People hear someone claim that a certain exotic part is required for acceptable sound, and their egos force them to invest.

I know a number of people who build speakers for fun and have done so for years. They do try to do their best, but at some level they have a "whatever" attitude about the outcome of any particular project. After all if they don't like this speaker, maybe they will like the next speaker better, and besides they always seem to find someone who wants to buy their last speaker so they at least get their parts cost out of it.

  • sq225917
  • [*]
Please, check these electrical components out
Reply #11


Walt Jung has distinct opinions on the audibility of different types of film caps and how measurements of parameters other than ESR and capacitance might back this up. He kinda knows his stuff.

  • pdq
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #12
He kinda knows his stuff.

Based on what?

Lots of people have opinions. What kind of evidence does he provide to back them up?

At HA we set a very high standard for statements that people make. Many people, when trying to meet that standard, fall short.

  • Woodinville
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #13
Beyond electrolytics, which have been covered, and outside of RF circuits, there appears to be little reason to reject the lumped-model hypothesis.

Now, that's today. There WERE some film caps a while ago that had quite a startling amount of hysteresys. It was rather startling, rather like when you changed sign of the derivitive the capacitance would "flicker" in a most odd way.

Nobody uses that kind of thing anymore, I hope.
-----
J. D. (jj) Johnston

  • Roseval
  • [*][*][*][*][*]
Please, check these electrical components out
Reply #14
Only cost and engineering time keeps all loudspeakers from being built with crossovers that run in the digital domain.


If I tell you there is an active system

Digital input over the network
AD converter for analogue input
Equalization in DSP
4th-order Linkwitz-Riley crossover in DSP
4 DAC’s + 1 for the woofer
4 class D amplifiers
Headphone amp
2 woofers + 2 tweeters
You don’t expect it to cost $300, don’t you?
http://wiki.slimdevices.com/uploads/a/ad/L...udio_Design.pdf

Active speakers (active crossover+ 1 amp per driver) are very common in the pro world.
I think it is most of all audiophile beliefs that prevent actives becoming popular.
Amps INSIDE the box, no vibration damping, no silver speaker cables, nothing to tweak. Brr….






TheWellTemperedComputer.com

  • rpp3po
  • [*][*][*][*][*]
  • Developer
Please, check these electrical components out
Reply #15
Very interesting! When you include the price tag, that's one of the most innovative audio products I have seen in a while.