If you parallel them it changes ESR by a huge amount to the circuit - which can have way more impact than value variance. And contrary to popular belief lower resistance isn’t always better (why would we have resistors if it was?).
Modding my ARC Ref Anniversary
- Thread starter ayreman
- Start date
If you parallel them it changes ESR by a huge amount to the circuit - which can have way more impact than value variance. And contrary to popular belief lower resistance isn’t always better (why would we have resistors if it was?).
My personal experience is very different. Increasing uF with state-of-the-art caps ALWAYS brings sonic dividends. That's exactly why the Mundorf electrolytic caps I put into the PS unit are 680uF instead of 470 uF in stock. I've increased the capacitance there by about a third and went from Nichicon (originally) to Mundorf now. I love the results! I've been using this strategy for almost 7 years in my system and have no doubt whatsoever that this is the way to go. Furthermore, I'm not the only one who follows this route. Look at GS-Maxi. He did exactly the same thing and couldn't be happier! It's a popular myth... that manufacturer "carefully selects" components and voices same ideally. What a load of bollocks! They go for the cheapest parts to maximize their profits. That is the sad truth that most audiophiles prefer to sweep under the rug and pretend it's not there.
Attachments
I'm glad your approach has worked out for you. I would point out that you have no evidence that the values of your caps are accurate without measurements. For all you know, they may both be far closer in value than they appear, or even more discordant than they seem. But continuing along the lines that you suggest works for you without really knowing the actual values of the components you use, I would ask: if 680uF worked better than 470uF, why not use 800uF, or 1000uF? Random selection of values and trial and error approaches seem capricious to me, but if that works for you, carry-on.
Aaahhh, I've hurt delicate feelings of the industry-affiliated. I'm glad!
It's bloody time someone did...marty, this one is very easy to answer: 680uF is the max capacitance Mundorf makes in 450VDC catagory. If 800uF was available, I would definitely go for it. Alas, I'm out of luck here. But there's good news elsewhere. In my DAC there's a linear PS (see attached photos). Each of those is 10000uF 25VDC. Well, guess what: I'm replacing them with Mundorf 47000uF 25VDC - dream come true!
Attachments
Although all engineering activities are framed by cost, I disagree with such view. Capacitors are complex entities, not just charge store, and simply changing/increasing capacitors can surely change the sound of equipment, but does not forcefully result in better sound quality.
I respect DIY activities and know that individual subjective opinions on modifications are beyond discussion - we should accept them a single data point. But what is the purpose of insulting people who think otherwise than you?
BTW, the subject of bypassing and changing capacitor values has been discussed in audio technical forums - every time we change them we are changing resonances, either in frequency and Q factor, and intrinsic noise. Manufacturer use this fact to tune the sound of their equipment.
DIY and modifications are a way of enjoying this hobby. They are not worst or better than just using standard equipment. They are simply a different way.
^^^
Question, why don’t I use the most expensive caps I can?
Answer, because I don’t think they sound good and there’s typically little if any and/or the opposite that they have any objective value (besides the $ count). No one can explain why any cap should sound better in power positions. There is no objective reason for that whatsoever. You can explain all day why it’s a better capacitor, but not how that translates to better sound.
Besides, some “better” caps can’t even be used in certain positions because the circuit has other needs (like Micro is referring to about Q).
I’m not opposed to DIY changes but when you decided to start replacing any diode you see without knowing what the parameters are... and then start insulting everyone like you’re some sorta savior coming in... if you burn up a piece of gear or need to sell it someday but can’t, well, don’t expect any sympathy.
It doesn't take a rocket scientist to replace and solder components to a board. What it takes is a willingness to experiment and a good deal of courage. I applaud ayreman for that. But having done that for years, I think micro's follow-up post is kind of where I net out as well. Component changing is not always a beneficial change for the better. The sword can cut both ways. One of the things that really impressed me is an ad for Wilson where Daryl (or Dave) is sitting in the listening chair and changing brands and values of individual capacitors in real time (the caps are in their lap using clips to 25' ft of wire going to the crossover circuit boards in the speaker) as they finalize and voice their crossover. That has my deepest audiophile respect as its the method a scientist like myself would prefer to use. But I applaud ayreman's efforts although one can imagine that once modded, it would be tough to accept that the result could be anything subjectively less than great after the surgery, since additional changes would be quite cumbersome and unpleasant at that point. But does ayreman get bonus points as an enthusiastic hobbyist? You bet! After all, the only person he has to please is himself, so why not go for it if that's his desire? That's what the hobby is all about.
You'd like what I do. I've spent years playing with different capacitors. I've swapped out so many caps... I mostly focus on PSU caps and signal coupling in electronics. Capacitors for speakers do not change the sound as much (but are significant) when you're using decent ones. But anyways if something sounds promising at all I often time with it, two weeks being pretty typical. You really know the difference when you've grown accustom to what you think you'll hear next. The ABing quickly is valuable but until you've started with slow changes you won't be any good at telling the difference in quick swaps.
1. What on earth makes you think "I start replacing any diode I see without knowing what the parameters are"?! How could such a preposterous idea enter your mind?!???!!! OF COURSE I know what the parameters are! I have the original schematic and the parts list right in front of me and I've been using the most qualified engineer (PhD) in this country to do the mods for me. Are you blind? Can't you see the workmanship in the photos????
2. Me? Insulting everyone?? Didn't you call me "a sad truth"? Verrrry polite and friendly statement!
3. Why in heaven's name would I EVER want to sell my RefAnn??!! You've got to be kidding me!!!
Last edited:
1. What on earth makes you think "I start replacing any diode I see without knowing what the parameters are"?! How could such a preposterous idea enter your mind?!???!!! OF COURSE I know what the parameters are! I have the original schematic and the parts list right in front of me and I've been using the most qualified engineer (PhD) in this country to do the mods for me. Are you blind? Can't you see the workmanship in the photos????
2. Me? Insulting everyone?? Didn't you call me "a sad truth"? Verrrry polite and friendly statement!
3. Why in heaven's name would I EVER want to sell my RefAnn??!! You've got to be kidding me!!!
So is this Phd engineer modeling all of the circuits in LTspice with the same profiles for the parts as the manufacturer? Maybe it's fine, but I'm speculative. I've had better profiles than found in patents before and tried to help a manufacturer and the "PhD" was too overqualified for them to learn about the instability inherent in their amplifiers... Sometimes the experience of the person making the circuit simply can't be overridden. When it comes to tubes this can be more true often because they're problematic often for a variety of reasons.
And why is a PhD engineer interested in doing this?
The quality of the physical work has nothing to do with the credentials. There's plenty of high school kids that can solder like that.
I did call you a sad truth, because you are insulting people by claiming subjective things are inherently better and others are near scamming you but you're exposing them... "shaking up the industry".
I'm curious what would make you believe that component mfr.s would label caps with incorrect values, especially audio caps where tolerances are commonly tighter?
They come from the factory with 20% variance per manufacture’s own data sheet. This has been the norm since the beginning of the electrolytic. Film caps however are usually 5 or 10% (but can be much lower).
The value of electrolytics is rarely critical to small a percent (if ever). And besides most know that they are often towards the shy side.
Agree with most of what you've said, but respectfully disagree about speaker caps. I've found that changing brands can often be beneficial especially when it comes to polypropylene/polystyrene bypass caps for the top end. I also found that the best electrolytic caps were the highly revered Black Gates, which are now discontinued (although are still available and plentiful), which I always preferred to use in the preferred series/parallel configuration (using 4 caps total for any single desired value). But admittedly this was done many years ago when the best modern caps (Mundorf, Duelund, others) were not available. My modding days are long over, but I did enjoy it very much.
It's not a matter of labeling the caps incorrectly, but rather to identify the typical variance of the lot (typically in the hundreds or thousands for a manufacturing run). Put simply, for a given lot, caps are labeled with their mean value and standard deviation. It's up to the engineer to decide what value is appropriate for their specific application and to select or reject accordingly.
Many years ago, I visited John Dunlavy at his company in Colorado because I admired his crossover design (published openly) and questioned him about part of the circuit I didn't understand, He brought me into the factory and took me to the bench where the tech was assigned to measure the capacitance value precisely of a particular part of the circuit and then adjust that value according to a 1% tolerance since the published cap values were inadequate for his needs. He said that it was critical for the time alignment and superb impulse measurements his speakers were famous for. But that key sub-circuit's precision, he said, was not necessary in other parts of the circuit. Then, we went out to dinner, where he got so smashed that the next day he totally forgot what he showed me the previous day in the factory. He was a giant of an audio engineer who I admired very much because of his transparency, but man, that guy could drink like a fish.
This is a valid point as caps may have a 10-15% variance from their notional spec. You can use this to your advantage when you need a cap value seemingly unavailable off the shelf by having the supplier find ones which test at the extremes of their variance range. That happened to me when I recapped the tweeter xo boards of my Benesch ACT and Tony Gee of Humble Home HiFi cap test fame was most helpful.
Well I can say there are some caps with speakers I consider unusable, even compare to electrolytics (which aren't always bad by any means in them). And yes BlackGates... I wouldn't call them plentiful. I've slowly used up all my 4.7uf's and such
Never to be seen again.