~ excellent ~
Awesome work man!!! KUDOS! you n'th degreed them just as I do / would... I am surprised by the outcome... I am stuck on the Muse KZ caps. Use them in nearly everything (that I can get them to fit in...) I have / stock practically every value, in all voltage ranges...
Awesome work man!!! KUDOS! you n'th degreed them just as I do / would... I am surprised by the outcome... I am stuck on the Muse KZ caps. Use them in nearly everything (that I can get them to fit in...) I have / stock practically every value, in all voltage ranges...
That's interesting regarding the impedance. I know it represents a complex relationship I don't claim to fully understand, but I have used FM's alot which are low impedance and have not personally experienced what you're mentioning with my own equipment. I think of them as very clean, exciting or lively, not as sucking the life out of the sound. Maybe others will chime in on them or other low impedance caps, but they're general quite positively regarded from everything I've heard for myself and from others as sort of a "go-to" general purpose cap when good, competitive specs with high ripple are called for in various audio applications. They're a little bright (to me) which is why I don't use them very much. Love the Gold Tunes btw for power supply. From what I'm aware, LKG is the other heavy-hitter there for PS besides the LKS, along with KG within that series.
Last edited:
Okay guys, just reporting back on this KL recapped amp. I switched around the channels so that this KL capped amp was driving the center channel. Unfortunately, I think I am a bit disappointed with the outcome. The sound was not bright, per say, but it had kind of a clicky type sound in the upper mids and highs. It also was somewhat thinner than before. It sort of reminded me of the Bryston 7b3 monoblocks. They were thin sounding as well. I could definitely hear the lack of body in the center channel. The comments that say this KL cap are a bit cold are actually true in some of my tests. I also agree with HeavyNova's comments that these KL caps are somewhat sterile and unremarkable. It really depends, but they didn't work out in my system.
I think the KL caps could be similar to low impedance caps because they did somewhat suck the life out of the sound.
The sound also was not as realistic and did not have the impact that the FG Fine Gold caps had. It leaves me somewhat frustrated, because I had high hopes.
I'm ordering a bunch of KA caps to do this test project again. This is likely my last experimentation before I hit the big XPR amps. At least this will tell me what the KA really sounds like. If the KA don't work out, I may just end up doing KZ caps where I can put them and the FG caps for the rest.
I think the KL caps could be similar to low impedance caps because they did somewhat suck the life out of the sound.
The sound also was not as realistic and did not have the impact that the FG Fine Gold caps had. It leaves me somewhat frustrated, because I had high hopes.
I'm ordering a bunch of KA caps to do this test project again. This is likely my last experimentation before I hit the big XPR amps. At least this will tell me what the KA really sounds like. If the KA don't work out, I may just end up doing KZ caps where I can put them and the FG caps for the rest.
You may try the Rubycon RX30 series. I've used them in some recent recapping jobs and they're doing fine (mainly used for local decoupling in the power stage and as power supply caps for lower power circuits (opamp/preamp supply). Impact and "body" in the mids are there. Also they sound realistic. Very similar to the Elna RFS-series but without lacking dynamics.
I was curious about the RX30 as well.. I purchased some but haven't installed them in anything yet. Thanks for sharing feedback on them. Regarding the FM/ low impedance caps, I recently recapped an old Magnavox FD-1051 cd player using many (12+) of them. I've had either generally neutral or at least somewhat positive experiences using them in the past in different projects and circuits that sounded decent with them. I had them all over the boards and in different positions on this new project though and did not like them at all. It was unmistakably dull and lifeless afterwards compared to old, original stock caps. I swapped them all out for 2x voltage rated, same capacitance RFS (it's all I had) and it sounded much livelier, fuller and natural afterwards. Although I had some success with them in other projects, I don't think I will use low impedance caps any more. I've never liked UKL outside of power supply. Even then, they may/ may not impart a sterile, unremarkable sound signature depending where they are in the circuit. I have also had good outcomes with UKL, but again, they may not sound as good as some other alternatives if somehow involved in a signal path. You may try UKA. I like them alot. You also mentioned UKZ, and that you had positive experiences with them. That would be the "safe" bet for you here, but feel free to experiment. The UKA has competitive (more) ripple current at low (and usually high frequencies, while the UKZ does not, but has lower leakage. I'd think those are the biggest differences between them which may effect sound. Also, the UKZ does not sound sterile or lifeless just because it is lower leakage. It's still a Muse series cap designed for audio.
Last edited:
HeavyNova - I have found that low impedance caps excel in digital sections. I used some 1000uf Panasonic FR (ultra low impedance - 18 mOhms) for post regulator caps in digital power supplies on my LKS DAC. They were absolutely excellent. This is for anything digital, such as DAC VREF, SPDIF section, digital control circuits, HDMI video, etc).
Maybe that's why I had liked them so much. I had used them in lots of digital sections before with pretty decent results.
Posting an update. It turns out that my previous opinions of the KL caps were due to incorrect components in my HT processor. That has been fixed.
The KA caps were a failure. I suppose that if you really need a 105 degree / 2000 hour audio capacitor, I guess you could use them. But they sound like crap and there are much better choices. I had KA caps in both an Emotiva amp and a set of Yamaha HS8 studio monitors that I recapped at the same time.
The KA caps have many bad characteristics. They are weak on bass/midbass and push the upper mids/highs. This makes the KA cap sound very thin and without body. The upper mids and highs are very harsh and bright. High frequency sounds such as drum snares sound like clicks instead of the real instrument. They also seem compressed and without dynamics because the audio doesn't seem to play as loud as with other caps.
Basically, the KA are definitely not recommended unless you absolutely need a 105 degree cap.
Back on the Emotiva amps. I have two Emotiva XPA-DR2 amps. One is loaded with KA caps on the amp boards. The second is loaded with mostly KL caps and some VZ caps. The KL/VZ sounded hugely better, with more bass/midbass and much more realistic sounding. I guess the KL caps aren't actually that bad.
I'm moving on to pull out all those KA caps and just do a KW install. Next step will to see how the KL stack up against the KW (which I have always thought sounded amazing).
The KA caps were a failure. I suppose that if you really need a 105 degree / 2000 hour audio capacitor, I guess you could use them. But they sound like crap and there are much better choices. I had KA caps in both an Emotiva amp and a set of Yamaha HS8 studio monitors that I recapped at the same time.
The KA caps have many bad characteristics. They are weak on bass/midbass and push the upper mids/highs. This makes the KA cap sound very thin and without body. The upper mids and highs are very harsh and bright. High frequency sounds such as drum snares sound like clicks instead of the real instrument. They also seem compressed and without dynamics because the audio doesn't seem to play as loud as with other caps.
Basically, the KA are definitely not recommended unless you absolutely need a 105 degree cap.
Back on the Emotiva amps. I have two Emotiva XPA-DR2 amps. One is loaded with KA caps on the amp boards. The second is loaded with mostly KL caps and some VZ caps. The KL/VZ sounded hugely better, with more bass/midbass and much more realistic sounding. I guess the KL caps aren't actually that bad.
I'm moving on to pull out all those KA caps and just do a KW install. Next step will to see how the KL stack up against the KW (which I have always thought sounded amazing).
That doesn't sound right at all. I used several UKA (among some of my other favorite caps including UKW) in a recent SP14 Aikido preamp build and the results of the project are the best I've ever heard. I would never describe these caps as weak, thin, harsh, bright, or without body. To me they are realisitic and natural (as far as electrolytics go). They've been one of my main "go-to" caps lately and I've yet to encounter a lackluster result with them. I wouldn't necessarily install them in speakers networks or crossovers though. Polypropylenes, PPS, polystyrenes, polycarbonates or even Kaisei bi-polars would probably sound better than a polarized electrolytic. You also mentioned you didn't like RFS, which are similar in their ripple spread to UKA and among my favorites as well for a natural sound compared to some other alternatives which often present more of an exciting or "larger than life", over-emphasized LF/HF ripple current spread that usually results in a cleaner, crisper, or more "HiFi-ish" sound by comparison, most especially when they have a competitive high frequency ripple current that is listed. Not everyone is going to like the same cap choices or sound signature to their music. UKW may sound slightly more robust and present a little more on top compared to the UKA. The only audio caps I've tried that I really didn't care for were the UFG Fine Golds. It may have been the circuits I used them in, but I thought they were mediocre in everything I put them in.
Right now I'm in the process of recapping a Cambridge Audio DacMagic Plus and deciding between Nichicon UES Muse and/or UVZ. I've purchased identical sets of both and deciding which series to use (or perhaps a combination of both).
Right now I'm in the process of recapping a Cambridge Audio DacMagic Plus and deciding between Nichicon UES Muse and/or UVZ. I've purchased identical sets of both and deciding which series to use (or perhaps a combination of both).
Last edited:
It could be that your systems are not transparent enough to reveal the flaws of the KA. Speakers, interconnects, power cables, etc. etc.
NOTE: all these caps were burned in 500 hours (3 weeks) before I judged their sound.
NOTE: all these caps were burned in 500 hours (3 weeks) before I judged their sound.
I did some longer listening last night on the Emotiva KL capped amp. I don't think these are the right answer either. Even though they are more natural and fuller sounding than the KA, they are a bit uninteresting and flat sounding. Somewhat lackluster. They are not as engaging and don't have the sweetness or excitement of better caps. I started to lose interest while listening.
So, next step is a shootout between KW and VZ.
So, next step is a shootout between KW and VZ.
I don't think so. I take this stuff way too seriously and have personally optimized every single piece of equipment in my audio chain with several capacitor revisions until arriving at my preferred, final versions. I suspect you may like a different sound signature than I do. After all, I love the Silmic. You may like caps with more current energy than the UKA can deliver. Like I said, UKW is a little more robust than UKA looking at its LF/HF ripple current, and as such, should sound a little fuller and present a little more on top. I'm still uninitiated with the UVZ and what that sounds like. As annoying as some of this can be at times, part of the reward of this hobby is doing your own experimentations and coming to your own conclusions on your preferred sound signature.
Been busy, but finished my burn in and evaluations.
The VZ caps were a fail, in my opinion. They do have a natural analog type sound in the overall midrange. However, they come across very dry sounding and they have really bad highs. High frequencies are loose and messy and somewhat distorted, in my opinion. I guess they are okay for cheap cap, but they have somewhat of a listening fatigue and they just did not engage me. I suppose if you need a general purpose 105 degree caps they will do, but they are not the best audio cap by far.
The KW caps are absolutely excellent. Very sweet in comparison. High resolution and excellent bass. Very natural and realistic sounding, but excellent and strong high frequency extension.
So I think it comes down to using the Muse KZ caps as much as possible, then using KW if the physical size of KZ is just too large.
The VZ caps were a fail, in my opinion. They do have a natural analog type sound in the overall midrange. However, they come across very dry sounding and they have really bad highs. High frequencies are loose and messy and somewhat distorted, in my opinion. I guess they are okay for cheap cap, but they have somewhat of a listening fatigue and they just did not engage me. I suppose if you need a general purpose 105 degree caps they will do, but they are not the best audio cap by far.
The KW caps are absolutely excellent. Very sweet in comparison. High resolution and excellent bass. Very natural and realistic sounding, but excellent and strong high frequency extension.
So I think it comes down to using the Muse KZ caps as much as possible, then using KW if the physical size of KZ is just too large.
"Beware of Audiophoolery"
Dear ainami, your original question: "When not to use low leakage KL?" is quite good, but from there, it seems the thread went a bit "down the rabbithole".
Perhaps "Audiophoolery" is not the right term to use here, as the caps you discuss are standard good quality stuff, and not overpriced esoteric components.
But I think a little reality is needed in this thread:
Do capacitors have different electrical properties? Yes certainly.
Is it possible for these differences to influence the sound?
Yes, but the effect is much smaller than "popular opinon" would seem to suggest, and it depends entirely on which function a given capacitor has in the circuit.
So for any meaningful discussion of the qualities of various capacitors, you at least have to be specific about the function of the caps:
Is it power supply or decoupling caps? DC blocking caps? Input caps? Output caps? NFB cap?
Keep in mind: The perfect capacitor has ZERO leakage and ZERO ESR.
So there is no such thing as "too low ESR" or "too low leakage" for a capacitor. It's nonsense.
If you need some resistance in a circuit, use a resistor.
The closest we get to the perfect capacitor is quality MKT/MKP film, which is why these are used in quality gear where possible.
Electrolytic capacitors are "all terrible", but yet, they are still quite good enough for power supplies and decoupling purposes.
IMHO, for power supplies, and when using quality industrial capacitors, there is generally no measureable or discernable differences to be found.
Swapping out the decoupling capacitors in your amplifier, and then trying to hear a difference, is a scenario filled with expectancy bias, and in addition, the human hearing (and brain) is just not able to discern and memorize minute differences over time. You are moving into the realm of fantasy here.
If you enjoy doing it anyways, great, but let's not fool ourselves 🙂
You should try and build an amplifier (many good designs to be found here), and compare to your Emotiva, then there will actually be some differences to hear!
-----
My answer to your original question:
Since a perfect capacitor has zero leakage, a lower-leakage cap is always better, as long as the other properties are no worse.
So if the Nichicon KL is just as good as eg. FG on all other electrical parameters, but just has lower-leakage, it is a better capacitor - for all purposes.
The datasheets only gives fairly basic specs, so to accurately compare these two, you would have to characterize them on a good LCR meter for all audio frequencies.
--
Personally, I prefer Panasonic FC and FR as the standard capacitors for power supply and decoupling:
They are longer life and higher temp spec than Nichicon KL and FG, and usually has lower ESR in my measurements.
Dear ainami, your original question: "When not to use low leakage KL?" is quite good, but from there, it seems the thread went a bit "down the rabbithole".
Perhaps "Audiophoolery" is not the right term to use here, as the caps you discuss are standard good quality stuff, and not overpriced esoteric components.
But I think a little reality is needed in this thread:
Do capacitors have different electrical properties? Yes certainly.
Is it possible for these differences to influence the sound?
Yes, but the effect is much smaller than "popular opinon" would seem to suggest, and it depends entirely on which function a given capacitor has in the circuit.
So for any meaningful discussion of the qualities of various capacitors, you at least have to be specific about the function of the caps:
Is it power supply or decoupling caps? DC blocking caps? Input caps? Output caps? NFB cap?
Keep in mind: The perfect capacitor has ZERO leakage and ZERO ESR.
So there is no such thing as "too low ESR" or "too low leakage" for a capacitor. It's nonsense.
If you need some resistance in a circuit, use a resistor.
The closest we get to the perfect capacitor is quality MKT/MKP film, which is why these are used in quality gear where possible.
Electrolytic capacitors are "all terrible", but yet, they are still quite good enough for power supplies and decoupling purposes.
IMHO, for power supplies, and when using quality industrial capacitors, there is generally no measureable or discernable differences to be found.
Swapping out the decoupling capacitors in your amplifier, and then trying to hear a difference, is a scenario filled with expectancy bias, and in addition, the human hearing (and brain) is just not able to discern and memorize minute differences over time. You are moving into the realm of fantasy here.
If you enjoy doing it anyways, great, but let's not fool ourselves 🙂
You should try and build an amplifier (many good designs to be found here), and compare to your Emotiva, then there will actually be some differences to hear!
-----
My answer to your original question:
Since a perfect capacitor has zero leakage, a lower-leakage cap is always better, as long as the other properties are no worse.
So if the Nichicon KL is just as good as eg. FG on all other electrical parameters, but just has lower-leakage, it is a better capacitor - for all purposes.
The datasheets only gives fairly basic specs, so to accurately compare these two, you would have to characterize them on a good LCR meter for all audio frequencies.
--
Personally, I prefer Panasonic FC and FR as the standard capacitors for power supply and decoupling:
They are longer life and higher temp spec than Nichicon KL and FG, and usually has lower ESR in my measurements.
Congratulations JVHB ! -- no not the "song" by Sir Cliff but you have just won the-
D.Self gold cup for practicality-down to earth -no nonsense prize .
Not that I disagree with you you make sense but it spoils whole threads and articles about it worldwide on websites and if somebody hears a difference ?--well as the audiologists say we all have different levels of hearing .
D.Self gold cup for practicality-down to earth -no nonsense prize .
Not that I disagree with you you make sense but it spoils whole threads and articles about it worldwide on websites and if somebody hears a difference ?--well as the audiologists say we all have different levels of hearing .
No doubt, so?
Scientific tests do the same to supposed Covid cures, lose weight fast diets, techniques to beat Vegas Casinos and so on.
"Hear hear" or "thinks" he hears?
Not the same but double blind testing can solve that doubt.
No doubt, and catastrophically degrading after childhood.
For the typical "audiologist" it´s already appalling, as any Phonoaudiologist (the licensed Medical kind) can easily test.
It's unfortunate that so many engineers on this forum believe that measurements tell everything and nothing else will affect the sound.
This is an interesting discussion and many good points are raised. For me, it's a yes and no proposition. It's true electrolytics are easily bested in certain respects, but I don't think their inclusion in gear always translates into poor performance. Not the best perhaps, but often times not too far off either. Caps in different parts of any circuit will have varying effect on sound. One interesting sound comparison is to find channel symmetry within circuit design to experiment with replacing different series on just one channel as the experimental, while keeping the control (factory) caps in place for comparison on the other side.
Here is where you can methodically choose caps in all positions for one channel that have different sound characteristics. I like to combine series of electrolytics on recaps, as long I keep series on any symmetrical pairs. Something as simple as hand matching any stereo paired electrolytics within 3% can make a difference in balance or sound, in addition to what series is used.
I've noticed lower ESR conveys cleanliness in an audio signal. I don't think all electrolytics of the same capacitance and voltage ratings are going to have the same sound signature by any stretch. Namely (to my ears anyway) lf/hf rated ripple current. Most all the industrial, power-designated power filter capacitors are going to have both very high, competitively rated lf/hf ripple (current) figures. What this directly translates to which isn't often mentioned, is the power, energy and impact of the music beyond the spec sheet. You won't find these figures listed most places where you research specs. I find/use them in my designs from Digikey.
It was hard to tease out, but once I started to pay close attention to all my recaps and cap specs with regards to their varying ripple, I started to see an association. I haven't missed a beat since and use it to my advantage to match up better suited electrolytics for various circuit positions, mostly based off ripple current-to-capacitance/voltage ratio that I haven't quite perfected yet. Increased voltage rating on any given capacitor is associated with increased ripple, where capacitance stays the same. Once you experiment around with paying special attention to your capacitor current, you start to see the association and effect on sound.
For instance, if you put a stereo pair of power filters in an amp, one being the highest ripple rated, designated power cap, versus a non-power rated general purpose cap let's say, with much lower available current at low and high frequencies on the other channel, what you'll notice is that the "better" power cap with its higher ripple is going to put forth much more energy to the sound both @ lf/hf. In effect, making the entire musical presentation much more forceful and accentuated. There can be so much ripple as to overemphasize sound characteristics, making the music sound larger than life basically which can be a good or bad thing (depending). One thing for sure is that any cap will sound more forceful and exciting (if given the chance), depending where it is in the circuit or whether it's even capable of conveying an audible effect or not. However, that's what lf/hf current the cap is putting out to any connecting circuitry. Whether you hear it or not in all cap positions all the time may always be debatable for some, but it's of course not even remotely possible to hear an audible difference every time with every cap, based off different circuit topologies and functions. You'll never know whether it does or not unless you try. There are simulations such as LTSpice which may be helpful for these types of theoretical approximations with regards to sound outcome, but I much prefer to just get my feet wet and jump in. A computer program may give you the appropriate levels, but without actually hearing it. Even if you could hear the results of your work, it would be a simulation and not your actual equipment making the sound. In the end, the actual sound is all that matters (to me anyway).
Here is where you can methodically choose caps in all positions for one channel that have different sound characteristics. I like to combine series of electrolytics on recaps, as long I keep series on any symmetrical pairs. Something as simple as hand matching any stereo paired electrolytics within 3% can make a difference in balance or sound, in addition to what series is used.
I've noticed lower ESR conveys cleanliness in an audio signal. I don't think all electrolytics of the same capacitance and voltage ratings are going to have the same sound signature by any stretch. Namely (to my ears anyway) lf/hf rated ripple current. Most all the industrial, power-designated power filter capacitors are going to have both very high, competitively rated lf/hf ripple (current) figures. What this directly translates to which isn't often mentioned, is the power, energy and impact of the music beyond the spec sheet. You won't find these figures listed most places where you research specs. I find/use them in my designs from Digikey.
It was hard to tease out, but once I started to pay close attention to all my recaps and cap specs with regards to their varying ripple, I started to see an association. I haven't missed a beat since and use it to my advantage to match up better suited electrolytics for various circuit positions, mostly based off ripple current-to-capacitance/voltage ratio that I haven't quite perfected yet. Increased voltage rating on any given capacitor is associated with increased ripple, where capacitance stays the same. Once you experiment around with paying special attention to your capacitor current, you start to see the association and effect on sound.
For instance, if you put a stereo pair of power filters in an amp, one being the highest ripple rated, designated power cap, versus a non-power rated general purpose cap let's say, with much lower available current at low and high frequencies on the other channel, what you'll notice is that the "better" power cap with its higher ripple is going to put forth much more energy to the sound both @ lf/hf. In effect, making the entire musical presentation much more forceful and accentuated. There can be so much ripple as to overemphasize sound characteristics, making the music sound larger than life basically which can be a good or bad thing (depending). One thing for sure is that any cap will sound more forceful and exciting (if given the chance), depending where it is in the circuit or whether it's even capable of conveying an audible effect or not. However, that's what lf/hf current the cap is putting out to any connecting circuitry. Whether you hear it or not in all cap positions all the time may always be debatable for some, but it's of course not even remotely possible to hear an audible difference every time with every cap, based off different circuit topologies and functions. You'll never know whether it does or not unless you try. There are simulations such as LTSpice which may be helpful for these types of theoretical approximations with regards to sound outcome, but I much prefer to just get my feet wet and jump in. A computer program may give you the appropriate levels, but without actually hearing it. Even if you could hear the results of your work, it would be a simulation and not your actual equipment making the sound. In the end, the actual sound is all that matters (to me anyway).
It is, but there is a lot of ignorance in the world more important than this. It's their loss.
Last edited:
I agree regarding the Silmic RFS, there was some music where they actually filtered out some of the vocals completely in one instance when I had used them.
Some of the low loss Nichicon parts generated more 3rd order distortion than would be preferred, but still seem to sound very good in most positions used. I used them in digital power supplies, near the regulators.
They all act as filters, or effects generators, some more than others so you have to be careful I guess, unless you just don’t care that much about details and the sound. Everyone has different preferences after all.
Some of the low loss Nichicon parts generated more 3rd order distortion than would be preferred, but still seem to sound very good in most positions used. I used them in digital power supplies, near the regulators.
They all act as filters, or effects generators, some more than others so you have to be careful I guess, unless you just don’t care that much about details and the sound. Everyone has different preferences after all.