I gotta' question - Rubycon

0 Members and 1 Guest are viewing this topic. Read 14002 times.

David Ellis

  • Full Member
  • Posts: 1044
    • http://www.ellisaudio.com
I gotta' question - Rubycon
« on: 3 Mar 2003, 08:19 pm »
This might become a little long.  I'll offer some preamble first.

I admit almost total ignornace with respect to electrolytic capacitors.  About the only thing I know is what I have read in the Art of Electronics (Horowitz), and a little bit of study on the Panasonic and Rubycon pages.  I really tried to find some significant measurable difference between the various types, but the differences tend to be VERY small.  The computer grade stuff might be larger, but it doesn't seem to offer anything special.

I understand a little about metallized poly capacitors and the film/foil too.  I understand how the stuff is offset wrapped, end sprayed... I understand the basic consruction.  I don't know "squat" about the construction of an electrolytic.

I will also admit some curiosity when visiting with Kevin Becker.  We popped the lid off his preamp.  There were a few things that struck me.    Among them were Rubycon electrolytics.  I was suprised for a second, then I realized the dynamics of the situation.  Frank VanAlstine has been tweaking stuf for 30+years, tested everything, and makes his living from putting good stuff in his electronics.  The killer sound from his stuff is not due to dumb-luck.  Rubycon is about the most expensive common electrolytic available.  Certainly a cheaper unit could be used, but it isn't.

I will also admit that not everything is measurable/tangible.  Such could surely be the case with "better" electrolytic capacitors.  If this is true with Rubycon electrolytics, I am okay with that.  However, I gotta' hunch that Mr. VanAlstine has some substance behind the Rubycon decision.

My question then:  What is the difference in an electrolytic capacitor, and/or how did you arrive at Rubycon?

Please don't feel compelled to answer quickly.  I do understand that some typing will be involved.

Dave Ellis

avahifi

I gotta' question - Rubycon
« Reply #1 on: 3 Mar 2003, 10:57 pm »
I hate to disappoint you but the use of Rubycon capacitors in that unit was just random chance (almost).  They were what was available in the case size, voltage, capacitance rating, at a rational price at the time we needed to reorder.  Attach no musical significance to them at all.  They are working, after all, at all of 60 Hz and are not part of the signal path, just power supply parts.  We build with parts that provide long term durability and pick parts that are appropriate to the intended use (for example, no polarized capacitors in AC applications where the voltage could be reversed such as a blocking capacitor in a signal path not biased on).  We would not choose large soft film inductive wound capacitors in any signal path because they are microphonic and generate high frequency spikes on transients - perhaps the "ambiance" and "detail" you paid extra for?  We would not use polystyrene capacitors in a tube unit as they are heat sensitive - value changes with temp - we are building audio amplifiers after all, not sonic thermostats.  We use tight tolerence parts to have matching channels, high quality parts to avoid unnecessary warranty repairs and long term service life, but we have yet to hear a part that "sounds good" --- most are very quiet when you hold then up to your ear and listen closely.  FVA

Rob Babcock

  • Volunteer
  • Posts: 9298
I gotta' question - Rubycon
« Reply #2 on: 3 Mar 2003, 11:05 pm »
Bravo!  I love the way at least one electronics mfg'r injects some sanity into the equation!

David Ellis

  • Full Member
  • Posts: 1044
    • http://www.ellisaudio.com
Thank you
« Reply #3 on: 3 Mar 2003, 11:30 pm »
"most are very quiet when you hold then up to your ear and listen closely. FVA" :lol:

Thanks for your very through and candid answer.  I expected nothing less.

Dave Ellis
 :wink:

Lou

  • Jr. Member
  • Posts: 43
DC is just DC, or not all DC is born equal....
« Reply #4 on: 18 Mar 2003, 09:19 pm »
Hi Frank,

With due respect, I take issue with what you said above;

"Attach no musical significance to them at all. They are working, after all, at all of 60 Hz and are not part of the signal path, just power supply parts."

All amplification is based upon this insignificant DC. GIGO demands that the better we filter the DC the more focused/correct, and better the output/signal/product. Soft recovery diodes have a tremendous impact precisely because they do impact the signal trough their more properly rectifying the DC supply voltage. The DC becomes part of the signal path necessarily as it is the bias used for amplification, it cannot not impact the signal. Low DC, lower amplification, too high a value, grain, dirty DC, confused and muddy output.

You would do well to pay more attention to the foundation upon which all else is built, DC. Then again, designers’ ignoring such is what keeps modifiers like me busy tinkering. :-)

As for “we have yet to hear a part that "sounds good"” I have certainly heard some that sound horrid in circuit… Some may even rattle when held to your ear and shaken, relays with missing springs and such…

avahifi

I gotta' question - Rubycon
« Reply #5 on: 19 Mar 2003, 03:50 pm »
Hi:

Thanks for your comments on designing proper power supplies.

We take CONSIDERABLE effort in engineering the power supplies used in our products.  All of the supplies are engineered to be higher capacity than what is needed by the audio circuits.  We design our power supplies to work in the "real world" where users are faced with brown outs, power surges, dirty power lines, etc.

What Frank is referring to is that we do not design our power supplies, or our audio circuits, around a particular "brand" of part.  Our designs are based on proper audio engineering, including proper power supply design.  Audio by Van Alstine has never been about using fashionable parts (yes we tried them, but they made no difference).  We are about designing and building high quality and reliable audio components based on sound engineering and mathematics.  If changing one brand of part can make a huge difference then I would question the overall design.

Thanks for you interest in AVA products!

Larry Jenkins

Lou

  • Jr. Member
  • Posts: 43
Thanks!
« Reply #6 on: 19 Mar 2003, 09:24 pm »
Hi Larry,

Thanks. I completely agree that a sound design is based on sound engineering, not on a particular product which may, or may not, be around next week. I suppose it was just the way it was phrased that struck me so.

Larry also said;

"yes we tried them, but they made no difference"

This I find truly interesting. I’ll agree with you 100% if you say that good parts cannot make a bad design sound good. OTOH, Black Gate caps, for example, can make a good sounding design sound as good as it can, as opposed to Panasonic FC’s etc.

Older Elna caps in the DC actually seem to add grunge to the DC resulting in high frequency distortion which ends up sounding like the component is bright. I figured that out when modifying a Forte F-44 preamp. It was ok as long as all the electrolytics were Elnas. Originally it sounded as though a blanket had been tossed over my speakers, but with a slight high frequency edge to it. I replaced a few of the Elna’s, which proved to be a mistake, with Nichicons and Black Gates and it sounded absolutely horrid. I should have replaced all the Elna's I later figured out. I had too do some serious head scratching to figure out what was going on to be sure.

The easiest and least expensive test I know of is taking a Sound Valves SV-100 preamp, usually about $150 on the used market, and about $300 new, and replacing the 1N4004 diodes with General Semiconductor UF4007’s, and the Illinois Capacitors in the amplification stage with Black Gates of the same value. My 80 year old Uncle who has a hearing aid noticed a tremendous improvement when I did this to the SV-100 I had given him. I cannot believe if he can hear it, that almost everyone cannot also hear the difference.

I do know, or have been told I should say, that Frank seems to think zip cord and dime store interconnects are fine. I listened to Julian Hirsch in Stereo Review on that one too. Then a store in St. Johns, Newfoundland, offered to allow me to return some speaker cables if they did not perform as promised. I was stationed at the submarine listening post we used to have in Argentia at the time.

Trust me, I was in the Navy and did not have any real money, I REALLY did not want to spend the $200 plus they were asking for these cables, and tried them more to humor them than because I believed them. I still have those cables buried around here somewhere.

I have been running electrostatic speakers for many years, so while I have been able to tell the difference between junk and decent cables, due, I believe, to the design of the estat speakers, I have not been able to hear great differences in boutique speaker cables. I believe they exist because I have heard too many people whom I absolutely trust tell me they have heard these differences.

ICs have been a completely different story. There is no question but that they can make a difference. Again, a good stable design can minimize those differences, but it cannot eliminate them.

I’ll tell you what, I’d really like to get an amp and preamp from you guys, mod it out, use sheathing on cables so no one could tell which is which, and using the same type of source and source material, and do an A/B at say, the Chicago Audio Society. I’d be willing to make a small wager, though I do not generally bet, that those present would by a 2 to 1 majority, if not much higher, prefer the system I have tweaked. Say a dollar wager?

avahifi

I gotta' question - Rubycon
« Reply #7 on: 24 Mar 2003, 04:34 pm »
I am curious if, when you replace capacitors, you use a precision meter to select capacitors of exactly the same value as the ones you replace.  The value of any given part is what it measures, not what is printed on it.  Changing part values randomly will, of course, change the sound as pole points are moved and phase gain differences occur in the two channels.

Or is it more like testing tires without a pressure gauge and wheel balancer?   :)

Lou

  • Jr. Member
  • Posts: 43
Checking Caps
« Reply #8 on: 24 Mar 2003, 07:13 pm »
I do not have a dedicated capacitance checker, though I believe the tollerance on my, now orphened, Tektronics TX-3 DMM is pretty tight. I really don't worry too much about the value of the pulled caps as they are supposed to match the value on the schematic, but I like to match the new caps closely. Sure, that can cause some very minor phase shifts in most circuits.

AuriCaps, for instance, always work/sound better than AudioCap Thetas, and both always sound better than the standard Solens. In this instance phase would not have anything to do with it, because Thetas dull the sound, at least in amps and speakers, compared to AuriCaps, which is where I have used both Thetas and AuriCaps.

This is not to say that every cap has an impact, it depends upon the circuit. Coupling caps and caps in the feedback circuit are where the most obvious differeces are to be realized.

Lou

randytsuch

I gotta' question - Rubycon
« Reply #9 on: 24 Mar 2003, 07:38 pm »
Quote from: avahifi
I am curious if, when you replace capacitors, you use a precision meter to select capacitors of exactly the same value as the ones you replace.  The value of any given part is what it measures, not what is printed on it.  Changing part values randomly will, of course, change the sound as pole points are moved and phase gain differences occur in the two channels.

Or is it more like testing tires without a pressure gauge and wheel balancer?   :)


If you change caps to blackgates in a power supply filter application, does it really matter if the caps are slightly different, because of tolerances?  I understand how it would matter in a filter, but did not think it would really make a difference in a PS.
A manufacturer can match caps, but it is not really pratical for a hobbyist like me.  Gates are too expensive to allow me to buy a bunch, and then hand match them.  And since I mainly use gates for power filtering (either on the PS or local), I never really worried about it.

Randy

Lou

  • Jr. Member
  • Posts: 43
Black Gate caps
« Reply #10 on: 24 Mar 2003, 09:03 pm »
Hi Randy,

In power supplies, the issue tends to turn to ESR, and reserve capacity. ESR is effective series resistance to an AC signal at a given frequency. The lower the resistance, the lower the loss. ESR is about the power loss due to heating, which is due to electrostatic attraction/repulsion of polar molecules in the dielectric.

Thus, the lower the ESR, the more power and the more ideally power is transferred to the load which is where we want the power to be used. The load in this case not only includes the speaker system, but the amplification circuit as well. Ultimately lower ESR caps tend to live longer because there is less stress due to heat. I believe that, unless you are dealing with gross ESR, that the heat issue is the more important issue.

In an ideal power supply circuit, caps would merely filter DC. In reality, they are also a ready reserve for instantaneous peak demands, a resistive load, and they even have some inductance. The larger the capacitor, generally speaking, the higher the ESR, but the higher the reserve, and as they are larger the ESR is spread out over more surface, making it less an issue than higher ESR in the same value capacitor. Some avoid this by paralleling many smaller value capacitors together. I know some early Perreaux preamps did this, as well as many others. I question the value of doing this, but then I have not experimented with it, so I am no authority on this issue.

There can also be special issues with ultra wide bandwidth amps and their power supplies. While power supply resonance usually is not a problem, with ultra wide bandwidth amplifiers it can be.  Resonant frequency is typically much higher than the unity gain bandwidth of most amplifiers and doesn't tend to cause amplifier instability problems, but with ultra wide band amps aditional filters may be needed. And on it goes… Sorry, I may have gone off the deep end here. I did not mean to try and go into power supplies 101, though some may be interested.

Lou

Lou

  • Jr. Member
  • Posts: 43
Sorry Randy
« Reply #11 on: 24 Mar 2003, 09:09 pm »
When I went back and reread your question, I realized that I never really answered it. :-(

“If you change caps to blackgates in a power supply filter application, does it really matter if the caps are slightly different, because of tolerances?”

No, it really is not going to matter unless the differences are really gross, which they will not be. OTOH, I have found that Nichicon’s premium caps are pretty good in power supplies, and they cost a good bit less.

Lou

randytsuch

I gotta' question - Rubycon
« Reply #12 on: 25 Mar 2003, 12:57 am »
Hi Lou,
Thanks for both answers  :wink: .
Wondered if you ever compared gates to Nichicons for power supply filtering?  Right now, I just use gates in smaller values because the big ones (above 1000 uf) are TOO expensive for me.  If the Nichicons even get close to gate performance in the bigger sizes, then they would be a good bargain.

BTW, my original question was really aimed at AVA, because I was wondering how they would recommend a hobbiest test/sort caps.  I do appreciate you answers though.

Randy

CE

  • Jr. Member
  • Posts: 97
I gotta' question - Rubycon
« Reply #13 on: 28 Mar 2003, 11:32 am »
Randysuch, how did you arrive at the conclusion that  Gates capacitors always work better than Nichicons? Lou, how did you arrive at the conclusion that Auricaps ALWAYS work better then Audiocap.  Is one brand more out of spec than another?  If you can't do an electrical check on them, how do you know? If the caps are that the label says?  Lou, why do Panasonic capacitors not "sound'' as good as BlackGates?  Is it maybe because you associate Panasonic name with some consumer products, and BlackGates are advertised as some super duper audio magic parts?  Perception is a big thing in advertising.  After some years of AVA products, Pre amp Omega III, 2 Hafler DH500 Omega modded amps, Omega DAC,  I am convinced, Frank and crew know their audio electronics!!!!!!  I use SoundKing speaker wire 12ga, $29 100/ft roll from Parts Express,  to hook up my Bi-Amped  Legacy Audio, Focus speakers. If wire has sound, like capacitors, I guess SoundKing is way underpriced. Perception is about 75% of the effect you are expecting from some magic parts.

avahifi

I gotta' question - Rubycon
« Reply #14 on: 28 Mar 2003, 02:17 pm »
This is in response to Lou's post of March 24th.

Interesting that he should bring up resonant frequency of power suppies as this is one thing we have dealt with in our amplifier designs that we have not seen addressed much by others.

In the Fet Valve and OmegaStar amplifiers, every power supply and output node is critically damped right at the output devices themselves by an RC network to provide a Q of 0.5 for each node.  This provides absolute stability for our ultra wide band designs and essentially makes the audio circuits effectively power supply garbage independent.

Yes I can understand how with many amplfiers changing brands of power supply capacitors (which also changes resonant frequencies of the supply) can have major effects on the sound, but that simply is a red flag to us that the output circuits were not properly designed in the first place. And since the hobbiest level changes are essentially random, how do you know what is "right" without doing the circuit analysis?   I suspect its back to testing tires without a pressure gauge all over again.

Frank Van Alstine

Lou

  • Jr. Member
  • Posts: 43
I gotta' question - Rubycon
« Reply #15 on: 29 Mar 2003, 04:08 am »
Quote from: avahifi
This is in response to Lou's post of March 24th.

Yes I can understand how with many amplfiers changing brands of power supply capacitors (which also changes resonant frequencies of the supply) can have major effects on the sound, but that simply is a red flag to us that the output circuits were not properly designed in the first place. And since the hobbiest level changes are essentially random, how do you know what is "right" without doing the circuit analysis?   I suspect its back to testing tires without a pressure gauge all over again.

Frank Van Alstine


First and foremost, I want to state that I completely agree with you in as much as the better the design, the less the impact of mods, to a degree. I'll also give you that there is a random element, but the results are consistant between inumerable circuits. In your theory, one would necessarily make one circuit better, and another worse, as it is random, the mix SHOULD be about 50-50.

Thus if results are consistant, this would indicate that your model is wanting. That is, your scope is too narrow? If not, my results with say Panasonic FC caps would be all over the map, same with Nichicon Muse and Black Gates. As they are predictable and consistant, I think you have a way to go yet before you can prove your case, actually, as this completely contradicts your argument, I think your argument must be overly simplistic, or improperly applied to this application.

A circuit analysis can mean many things, are you using Spice, or what? As you have not defined your application of the term, I cannot say why I believe it is flawed, are you are considering Q alone.

eico1

I gotta' question - Rubycon
« Reply #16 on: 29 Mar 2003, 04:37 am »
Lou, what is arbitrary is that you are using your preferences to determine which brand of cap you like. Swaping caps in a theoretically well designed amp my will cause a consistant change to such an amp, one cap may be underdamped, the other overdamped. You just prefer the sound of one condition over another.


steve

Lou

  • Jr. Member
  • Posts: 43
I gotta' question - Rubycon
« Reply #17 on: 29 Mar 2003, 05:30 am »
Quote from: eico1
Lou, what is arbitrary is that you are using your preferences to determine which brand of cap you like. Swaping caps in a theoretically well designed amp my will cause a consistant change to such an amp, one cap may be underdamped, the other overdamped. You just prefer the sound of one condition over another.


steve


Hi Steve,

The argument falls apart because the result is constant. If we were discussing identical circuits, then I would be forced to agree. Further, how are you determining the dampening? Are you speaking to physical dampening, or are we still on Q? If Q, then I stand pat, if physical dampening, that is another issue, but we were speaking of Q. If Q were too high in one circuit, than X cap would work better than Y cap in that circuit, but when Q was low, the inverse should be true. In fact, Panasonic FC's yeald less bass in all circuits, and Nichicon Muse more, highs are better with Nichicons as well. I have not done sufficent testing with Black Gates, due to cost issues in larger power supply caps, to say for sure if they are better still. In other applications they are.

Thus, if I generally preferred an underdampened circuit, there would be a point where even this was too much of a good thing, and I would have had to resort to different caps to address this. As this has not been the case, the argument does not seem to have an application here.

Lou

CE

  • Jr. Member
  • Posts: 97
randytsuch
« Reply #18 on: 29 Mar 2003, 03:08 pm »
See, you may have learned something.  Your hit or miss swapping of parts, based on what?  Perhaps you need to approach it with some logic and reason, perhaps you have allowed yourself to fall prey to slick ads about the powers of magic caps. Would you say AVA with 35+ years of audio electronics design, knows a few things about components and it's effect on functionality? How bout' magic wire, used inside of amps and pre amps?  Some people think they can hear the sound of Teflon insulation versus plastic insulated wire?  More amazing, some can hear the sound of wall outlets!! And they can hear the difference between a Leviton and a Hubbell!!!!!!!!!!!!!!!!!!!!!!!.  Those kind of ears need to be donated to research labs, cus' that is beyond human abiltys, WOW.

Lou

  • Jr. Member
  • Posts: 43
Re: randytsuch
« Reply #19 on: 29 Mar 2003, 03:14 pm »
Quote from: CE
See, you may have learned something.  Your hit or miss swapping of parts, based on what?  Perhaps you need to approach it with some logic and reason, perhaps you have allowed yourself to fall prey to slick ads about the powers of magic caps. Would you say AVA with 35+ years of audio electronics design, knows a few things about components and it's effect on functionality? How bout' magic wire, used inside of amps and pre amps?  Some people think they can hear the sound of Teflon insulation versus plastic insulated wire?  More amazing, some can hear the sound of wall outlets!! And they can hear the difference between a Leviton and a Hubbell!!!!!!!!!!!!!!!!!!!!!!!.  Those kind of ears need to be donated to research labs, cus' that is beyond human abiltys, WOW.


Why not come back when you can actually contribute something? I have been working in the field for 25 years myself. If all you want to do is ridicule others, find an insult forum. You obviously know nothing of the field as you contribute only insult.