The usual question: cap "upgrades" (real or imagined)

adamct · 10938

0 Members and 1 Guest are viewing this topic.

Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
I recently (well, relatively recently) picked up a Smack kit from someone who never got around to building his. I will build it in stock form first, but knowing me, I will want to roll in some film caps fairly soon after I have it up and running.

I see from the parts list that the Smack has the following caps:

2x 0.18uf 450V metallized polypropylene caps
2x 10uf 250V metallized polypropylene caps (Solen)
3x 220uf 250V electrolytic caps
2x 10,000uf 6.3V electrolytic caps
1x 10,000uf 10V electrolytic cap

Which of these are potential "upgrade" (i.e., replacement) or bypass (using a film cap with 1% of the capacitance) candidates?

For the 0.18uf / 450V caps, I haven't been able to find any replacements. Only 0.15uf, 0.2uf or 2.2uf. Don't know if any of those would work.

For the 10uf / 250V caps, I've found the following candidates:



Offline corndog71

  • Sr. Member
  • ****
    • Posts: 593
Reply #1 on: May 16, 2013, 09:55:56 AM
While I'm not familiar with the Smack I can make a reasoned assumption that the best place for upgrades would be the .18uF and 10uF caps.

I therefore recommend Clarity ESA series in 0.1uF/630V and 10uF/250V. 

If you want to really supercharge it then consider the 220uF/700V Clarity TC caps but others might say it's not worth it.  Or you could just bypass the current ones with 0.1uF/250V film caps.  (2.2uF seems kinda big for a bypass in my opinion.)

The world was made for those not cursed with self-awareness.

Rob


Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #2 on: May 16, 2013, 10:05:54 AM
Yeah, I saw the 220uf/700V ClarityCaps, but at $90/each, and given that I would need 3 of them, that just isn't going to happen. I thought the rule of thumb was that you would bypass a high capacitance cap with one that has 1% of it's capacitance? That's how I came up with 2.2uf.

As for the .18uf, am I wrong in thinking that a .2uf would be a better choice than 0.1uf? Or at least 0.15uf?




Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5751
Reply #3 on: May 16, 2013, 10:38:26 AM
2x 0.18uf 450V metallized polypropylene caps - these are in the shunt regulator stability bypass RC network, and may have some effect. It will be very small though! They must have good high-frequency behavior, as in VHF radio frequencies - so forget mylar. In polypropylene, look for film and foil, or anything labelled "snubber". AC line rated ceramics may be a good choice.

2x 10uf 250V metallized polypropylene caps (Solen) - these are the parafeed output caps, and by far the most important sonically. Look in the power amp sub-forums to see what others have used in that function. Whatever you try, remember that breakin is especially audible with these caps, it will take usually 50-200 hours of actual, loud, music before you really know what they sound like.

3x 220uf 250V electrolytic caps - these are in the high voltage power supply, isolated from the audio by two C4Ss and a shunt regulator, and not going to be directly audible. There are two areas for possible improvement - you can look for caps of higher temperature rating (105 degrees C) and higher voltage (up to 450v) which will help the caps live longer, and you can bypass them in the hope that it will reduce switching noise at radio frequencies. For the latter, snubbers and AC line rated ceramics are good candidates.

*** for switching noise, consider replacing the UF4007 rectifiers with Cree silicon carbide Schottky diodes, 600v or greater; that will likely make more difference than bypassing the filter caps. ***

2x 10,000uf 6.3V electrolytic caps - these are "soft-start" time delay caps; their AC behavior is shunted by the LEDs so they are unlikely to have much sonic effect. Also, the location is pretty tight; the extended wires necessary to use any alternative would probably do more harm than any possible good.

1x 10,000uf 10V electrolytic cap - this is the heater power filter, not connected to any audio area. It already has Schottky rectifiers, so the least unlikely improvement would be a bypass for high frequencies. Low voltage ceramics in Z5U material would be a good candidate. I'd want 10uF at least to have any possible effect, and note that these caps lose a lot of their capacitance with DC voltage on them - use 16v or 25v parts, and you'll probably want the 22uF or greater nominal capacitance.

Paul Joppa


Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #4 on: May 16, 2013, 11:40:22 AM
Paul,

Thank you for the quick, and VERY detailed response. Would it be a fair summary to say the the most worthwhile changes would be (i) the 2x 10uf 250V metallized polypropylene caps and (ii) the substitution of the Cree silicon carbide Schottky diodes for the UF4007 rectifiers?

It sounds like changing anything else is unlikely to have a significant sonic impact.

Best regards,
Adam



Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack


Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #7 on: May 16, 2013, 03:34:02 PM
Thanks, Paul.



Offline earwaxxer

  • Hero Member
  • *****
    • Posts: 1336
Reply #8 on: May 16, 2013, 04:26:51 PM
Paul's overview and advice is, as usual, spot on and grounded in reality/science.

It can be fun to tweak caps, much like one rolls tubes. I would probably listen to it stock for a good period of time. Get to know the sound well. Then try to focus on what you would like to change about the sound. The changes in capacitor sounds are real, and definitely worthwhile. Bypassing can be a less expensive way to the same goal. For me, it seems more hit or miss. From my experience, good caps 'open up' the sound, and elevate the kit to the 'next level'. Kind of like the difference between a good popular affordable piece of kit, to one that is closer to boutique or SOTA.

Sometimes it surprises me when I read reviews about a particular popular and high performing budget component, but they dont mention at all anything about the crossover components. Its kind of like not realizing that the zip cord one is using for speaker wire may not be doing justice to the speaker at the other end.

Eric
Emotiva XPA-2, Magnepan MMG (mod), Quickie (mod), JRiver, Wyrd4sound uLink, Schiit Gungnir, JPS Digital power cord, MIT power cord, JPS Labs ultraconductor wire throughout, HSU sub. powered by Crown.


Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #9 on: May 17, 2013, 01:58:57 AM
In principle, you're right of course. But it also assumes (a) a lot of time, (b) boatloads of money and (c) some way of anticipating how a certain cap will change the sound.

As a practical matter, I don't have the time to try out many different caps and break each set in, I can't really afford more than one or two sets of caps and have no desire to pay for a set of caps I ultimately won't use, and I have no way of saying to myself, for example: "I would like to tame the treble a bit and give the mids more weight" and then knowing which caps would be the best for achieving that. Reviews of caps are (a) limited in number and the kinds of caps reviewed, (b) largely perfunctory and lacking in any sort of objective details, (c) from using caps in all kinds of equipment that I don't have where the caps are likely to perform differently, and (d) highly subjective and written by people who may be experts, but are also likely to know as little as I do.

So as a practical matter, I'll buy some caps that people generally seem to find good (and that I have used in my Crack to excellent effect), drop them in, and be done with it. I'm doing it for fun more than necessarily because I believe it will radically improve the sound. Bottlehead seems to do a pretty good job of rationally choosing components that give the most bang for the buck to create kits that sound excellent in stock form and outperform the competition.

Anyway, first I have to build it!  :D



Offline earwaxxer

  • Hero Member
  • *****
    • Posts: 1336
Reply #10 on: May 17, 2013, 03:57:28 AM
.....So as a practical matter, I'll buy some caps that people generally seem to find good (and that I have used in my Crack to excellent effect), drop them in, and be done with it.

I wish I could do that!.... I think I need therapy!

Eric
Emotiva XPA-2, Magnepan MMG (mod), Quickie (mod), JRiver, Wyrd4sound uLink, Schiit Gungnir, JPS Digital power cord, MIT power cord, JPS Labs ultraconductor wire throughout, HSU sub. powered by Crown.


Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #11 on: May 17, 2013, 04:19:30 AM
LOL! That actually made me laugh out loud...  ;D ;)



Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #12 on: August 26, 2013, 05:21:45 AM
3x 220uf 250V electrolytic caps - these are in the high voltage power supply, isolated from the audio by two C4Ss and a shunt regulator, and not going to be directly audible. There are two areas for possible improvement - you can look for caps of higher temperature rating (105 degrees C) and higher voltage (up to 450v) which will help the caps live longer, and you can bypass them in the hope that it will reduce switching noise at radio frequencies. For the latter, snubbers and AC line rated ceramics are good candidates.

*** for switching noise, consider replacing the UF4007 rectifiers with Cree silicon carbide Schottky diodes, 600v or greater; that will likely make more difference than bypassing the filter caps. ***

I recently replaced all 3 220uf 250V electrolytics in the power supply of my Speedball'd Crack with 200uf metalized polypropylene film caps. I can't explain it, but I feel like the change is distinctly audible, much bigger than I ever would have expected. Frankly, I was really sort of shocked. Bass seems much stronger and resolution (if possible) seemed to get even better than it was before.  Best of all, the cost was surprisingly low.

Since I already have one of the Schottky diodes you recommend above, along with replacement film caps for the parafeed output caps, I'd like to perform the same mod to my Smack. I would have to mount the caps externally and run fly leads to them. I would be using 200uf film caps rated at 450VDC. Two questions:

1. Putting cost aside (again, the cost is fairly minimal), is there any reason not to do this? Is there a chance the fly leads will cause problems? Is there any reason why films caps are a bad idea here?

2. Are 200uf film caps sufficient, or should I use 250uf caps instead? The price difference is negligible. Or do the caps really need to be exactly 220uf?

Best regards,
Adam



Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5751
Reply #13 on: August 26, 2013, 07:01:32 AM
Crack, being a cathode follower, has the power supply in the output signal current loop and should be much more sensitive to improving the PSU capacitors.

For Smack or Mainline, the shunt regulator reduces the effect of power supply - but that also means you can get away with less capacitance. So no worries about the smaller value.

Experience tells me that someone will immediately ask "how low can I go?"  I have no answer, since there will probably be audible effects before you reach a purely theoretical limit. Try it and find out! It won't hurt anything except your wallet.  :^)

Paul Joppa


Offline adamct

  • Sr. Member
  • ****
    • Posts: 755
  • Maxxximum CAPacity Crack
Reply #14 on: August 26, 2013, 07:05:46 AM
Thanks, Paul. Since I have 200uf caps I like, I'll stick with those and won't try to go lower. And since I don't think there are too many people building Smacks anymore, you should be safe from others as well!

Best,
Adam