Peter
Compared to some of the people on this fora, I even have trouble calling myself a novice. Infant is more likely to hit the nail..
But I am still serious about my question. Reason I ask, is that my way of installing a cap, letting the unit play for a week or two, makes comparisons pretty difficult, as sonic memory is a very funny thing.
I alwayes seem to end up with a change that will be great for some recordings, but not all - so thats why I would like to see if there is a better way.. I mean, do you pre-condition them before trying?
Compared to some of the people on this fora, I even have trouble calling myself a novice. Infant is more likely to hit the nail..
But I am still serious about my question. Reason I ask, is that my way of installing a cap, letting the unit play for a week or two, makes comparisons pretty difficult, as sonic memory is a very funny thing.
I alwayes seem to end up with a change that will be great for some recordings, but not all - so thats why I would like to see if there is a better way.. I mean, do you pre-condition them before trying?
I never really observed any terrible sound from fresh caps (like for instance BG N or STD). I usually don't precondition them before comparison either. While they may improve after some period, the initial sonic signature stays, and if anything, it gets only refined with time.
I have some BG and Rubycon caps that I'm regularly swapping in my DAC and after at least 200 hours of operation (the DAC is always on), the differences between those two caps are still the same.
I have some BG and Rubycon caps that I'm regularly swapping in my DAC and after at least 200 hours of operation (the DAC is always on), the differences between those two caps are still the same.
OK - my findings are, as mentioned before, somewhat different. Most caps start off with a somewhat metallic sound on especially classical music/violins and then graduately become better until all of a sudden there is a big change and they fall into place. Have had that with BG's in both standard, VK and N versions. The Cerafine caps are especially terrible sounding to me in the beginning, but it is also a pretty cheap cap so that may be the reason why.
Only cap I have come across that seems to change very little is CDE. It also happens to be one of my favorite caps (however not in all places..), big bold and beatifull but not in any un-natural way. However they are, at least in this part of the world, hard to come by in some sizes..
Only cap I have come across that seems to change very little is CDE. It also happens to be one of my favorite caps (however not in all places..), big bold and beatifull but not in any un-natural way. However they are, at least in this part of the world, hard to come by in some sizes..
aarsoe said:
Agreed with all you wrote, same experiences here but maybe in a lesser extent. All caps sound better after a while with BG being the absolute winner in terms of the time they need to sound optimal.
But what are CDE ?
Edit: after 5 seconds with Google I found its the good old Cornell Dubilier brand. Never tried the newer types.
Peter Daniel said:
Peter,
What current were you drawing from the power supply and did you place a 0.1 film cap across the 100u Blackgate N?
Interestingly, over on the Aspen forum at ‘The Audio Circle’ they have found that their power amps sound better with small PSU caps rather than the traditional 50,000u or so. I hope so as my pre-amp has 4 x 4700u standard electrolytics per rail which would be expensive to replace with Blackgates!
Brian
When possible, I try to avoid additional bypass caps. The single electrolytic (of reasonable quality) usually sounds better. While the 0.1u cap may spice things a bit, when you listen more carefully, you'll notice that it's mostly additional flavour, robbing your music of any natural character.
I'm still with 100u BG N as only smoothing cap. So far it provides the best sound I got so far from a chip amp. I compared it to 1000u BG N, the character is similar, but bigger caps somehow take the liveliness out of music, making the sound less real. I also tried adding 1000u BG N and STD caps at rectifiers, as additional bypass, but again, it took something out of musc that makes it less real.
There was a similar effect when using batteries. A battery can be compared to a large cap, and it's effect on the sound is similar: it's less direct, lacking this elusive "element" of the presentation that makes you wonder how it's possible that the recorded music can sound this way
So for now, it's just 100u BG N, the only caps after the rectifiers and 400VA toroid. Try it, you won't be dissapointed.
Regarding the noise with such small capacitance; there is slight buzz in the speakers (94 dB efficiency), but only when you listen very closely. At normal listening position it's not detectable.
Regarding the bass; it's surprisingly good. It's more detailed and tight. While more capacitance produce sort of "fat" bass, less capacitance provide same depth of bass, but it's faster, tighter and again, more real.
As you noticed I'm using word "real" quite often. Well, that's what you get, when capacitance becomes this low
Here's the pic of my test rig. I'm not using any series resistor at the input, as the resolution improves without it.
I'm still with 100u BG N as only smoothing cap. So far it provides the best sound I got so far from a chip amp. I compared it to 1000u BG N, the character is similar, but bigger caps somehow take the liveliness out of music, making the sound less real. I also tried adding 1000u BG N and STD caps at rectifiers, as additional bypass, but again, it took something out of musc that makes it less real.
There was a similar effect when using batteries. A battery can be compared to a large cap, and it's effect on the sound is similar: it's less direct, lacking this elusive "element" of the presentation that makes you wonder how it's possible that the recorded music can sound this way
So for now, it's just 100u BG N, the only caps after the rectifiers and 400VA toroid. Try it, you won't be dissapointed.
Regarding the noise with such small capacitance; there is slight buzz in the speakers (94 dB efficiency), but only when you listen very closely. At normal listening position it's not detectable.
Regarding the bass; it's surprisingly good. It's more detailed and tight. While more capacitance produce sort of "fat" bass, less capacitance provide same depth of bass, but it's faster, tighter and again, more real.
As you noticed I'm using word "real" quite often. Well, that's what you get, when capacitance becomes this low
Here's the pic of my test rig. I'm not using any series resistor at the input, as the resolution improves without it.
Attachments
And this is PS. The effect I'm describing here is possible only with AC supply, batteries won't work here, as they act as big cap. In the picture you see also the snubber circuit (with a switch) and 1000u caps I tried in addition to 100u on amps board. Adding those caps is a trade off. It may sound better with some material, and definitely degrading with other. I'm thinking about adding a switch which can connect those caps in case I'm into more "formal" type of presentation
Attachments
With regards to preamp (or my experience with DAC), lowering capacitance almost always brings improvement. For instance, in my LM380 preamp I had originally 6,800u and 330u caps. I had no problem switching them to 1000u and 47u with substantial gains in sonic reproduction.
For testing, I suggest installing sockets, and switching caps untill the results will satisfy you.
For testing, I suggest installing sockets, and switching caps untill the results will satisfy you.
I tried it briefly, and again it seems to work in a similar way. It takes away a bit of this "noise" between the notes, but at the same, the sound is drier with less air and microdetail. One might say less delicacy.
For instance, in Here's to Ben on track 'In a Wee Small Hour', without the snubber Jacinta's voice echo decay has very obvius resonance pattern which slowly dies out. With the snubber, this decay is still pronounced in a similar way, but it's more like one continuos sound, without the resonance pattern being obvious anymore.
For instance, in Here's to Ben on track 'In a Wee Small Hour', without the snubber Jacinta's voice echo decay has very obvius resonance pattern which slowly dies out. With the snubber, this decay is still pronounced in a similar way, but it's more like one continuos sound, without the resonance pattern being obvious anymore.
I totally agree with you Peter.
Though my knowledge on electronics is
limited, I've built a number of projects.
One pet project of mine was a DAC which
I built 5 years ago. With that DAC, I've learned
alot about electronics especially about power
supplies & capacitors. After playing with so many
brands of capacitors, black gates std are the only
type that I would use. I've also discovered that
using too much capacitance is not only overkill
at the same time it kills the tempo of the music.
The music becomes tight & stale meaning lifeless
& instruments do not expand & decay naturally.
Though my knowledge on electronics is
limited, I've built a number of projects.
One pet project of mine was a DAC which
I built 5 years ago. With that DAC, I've learned
alot about electronics especially about power
supplies & capacitors. After playing with so many
brands of capacitors, black gates std are the only
type that I would use. I've also discovered that
using too much capacitance is not only overkill
at the same time it kills the tempo of the music.
The music becomes tight & stale meaning lifeless
& instruments do not expand & decay naturally.
Hmm, Hmm...
I wonder if what you're listening to is mains ripple breaking through to the music and giving it stuff it's not supposed to have.
I guess you're not using more than about 15 Watts per channel.
I'm using Watt hungry four way full range monitors and 200 Watt or 300 watt mono blocks, all custom built or changed beyond original recognition. I have super tweeters. I can here what difference changing any resistor in the chain anywhere makes.
My findings are that I can almost not get enough capacitance in power supplies. My preamp used to have 50,000 mic. It started with two x 4700 and grew until there was no further improvement. Until I stopped using a preamp (read volume control), which was better in many ways when I realised I could wire my CD player to my amps I now have fitted a source selector in the form of a Shallco switch which seems to not have degraded the sound. I maybe going to be trying a volume control one day as it's bit impractical really as it is.
Anyway. I find the BG's to be punchy and impressively dynamic. The standard range can be a bit coarse. They can sound grainy and unpleasant. But also quite good. Nichicon KZ are lame in terms of dynamics by comparrison but can be nicer to listen to. I find it does to some extent depend where you put them.
I run my capacitors in for three months before I do any comparrisons. No point otherwise. Maybe that's longer than needed but I like to be sure. I've found caps usually sound thin and tinny and bright and brash to start with. They become sweeter and fuller. Same for film caps, silver wire in teflon, vishays or caddocks, etc.
80,000 mic of cheap Elnas was on my stereo amp to get the tighest bass and sweetest treble, best imaging and most transient response. Now use the superb Nichicon GSIV with only 40,000 on my mono blocks. But that's only temporary, I have another 20,000 per side which typing this might propel me to get out of storage and run start running in.
I agree with your findings about paralleling caps can have unpleasant sounding results. I had BG FK 100@50V across 1000@50 KZ across cheap 10,000@63's by the output devices of our stereo 120W per channel integrated amp. The bass was fuller but a lot more but muddled. Treble also. Ended up with just the BG's across the 10,000, was a little bright but the best compramise as we all prefered it with the extra transient response. KZ's added too much bloom every where.
I would not use coupling cap's anywhere. They are all carp. Destroy the sound quality no matter what combination or how it's done. I've found. I've also found copper PCB track to be awful and hard wire all components with silver wire where necessary but usually by trying to join them all directly to each other.
As far as CD player go, I have also found that a large main reservoir to best. I have 5 transformers on it now. Two with healthy main res. The analogue stage being up to 30,000mic and no measurable mains ripple. And even with all that there, I found adding some oscons, 470@16V, afte the two discrete regulators, made the sound more clear and clean. I haven't played around much with that really. There is much improvement still to be had I think.
I'm interested to hear about these other types of capacitors. I've not tried any of them, ZA, etc, and new polymer ones that are better that oscons, what are they called? I've only tried a few Oscons, the old G series I think they were called, with copper leads. Now I notice an SP seies i the catalogues. But I've not seen SEP. What are they?
I logged on to find out about the Jelmax ceasing production thing. Interesting thread.
The BIG advantage the BG N has over the Oscon is the non-polarised nature of it and the large values and the higher voltages.
Can anyone tell me if there is a good outlet for Audio Oscons with a good range of values in the UK or who will ship to here.
Thanks.
I wonder if what you're listening to is mains ripple breaking through to the music and giving it stuff it's not supposed to have.
I guess you're not using more than about 15 Watts per channel.
I'm using Watt hungry four way full range monitors and 200 Watt or 300 watt mono blocks, all custom built or changed beyond original recognition. I have super tweeters. I can here what difference changing any resistor in the chain anywhere makes.
My findings are that I can almost not get enough capacitance in power supplies. My preamp used to have 50,000 mic. It started with two x 4700 and grew until there was no further improvement. Until I stopped using a preamp (read volume control), which was better in many ways when I realised I could wire my CD player to my amps
I now have fitted a source selector in the form of a Shallco switch which seems to not have degraded the sound. I maybe going to be trying a volume control one day as it's bit impractical really as it is.Anyway. I find the BG's to be punchy and impressively dynamic. The standard range can be a bit coarse. They can sound grainy and unpleasant. But also quite good. Nichicon KZ are lame in terms of dynamics by comparrison but can be nicer to listen to. I find it does to some extent depend where you put them.
I run my capacitors in for three months before I do any comparrisons. No point otherwise. Maybe that's longer than needed but I like to be sure. I've found caps usually sound thin and tinny and bright and brash to start with. They become sweeter and fuller. Same for film caps, silver wire in teflon, vishays or caddocks, etc.
80,000 mic of cheap Elnas was on my stereo amp to get the tighest bass and sweetest treble, best imaging and most transient response. Now use the superb Nichicon GSIV with only 40,000 on my mono blocks. But that's only temporary, I have another 20,000 per side which typing this might propel me to get out of storage and run start running in.
I agree with your findings about paralleling caps can have unpleasant sounding results. I had BG FK 100@50V across 1000@50 KZ across cheap 10,000@63's by the output devices of our stereo 120W per channel integrated amp. The bass was fuller but a lot more but muddled. Treble also. Ended up with just the BG's across the 10,000, was a little bright but the best compramise as we all prefered it with the extra transient response. KZ's added too much bloom every where.
I would not use coupling cap's anywhere. They are all carp. Destroy the sound quality no matter what combination or how it's done. I've found. I've also found copper PCB track to be awful and hard wire all components with silver wire where necessary but usually by trying to join them all directly to each other.
As far as CD player go, I have also found that a large main reservoir to best. I have 5 transformers on it now. Two with healthy main res. The analogue stage being up to 30,000mic and no measurable mains ripple. And even with all that there, I found adding some oscons, 470@16V, afte the two discrete regulators, made the sound more clear and clean. I haven't played around much with that really. There is much improvement still to be had I think.
I'm interested to hear about these other types of capacitors. I've not tried any of them, ZA, etc, and new polymer ones that are better that oscons, what are they called? I've only tried a few Oscons, the old G series I think they were called, with copper leads. Now I notice an SP seies i the catalogues. But I've not seen SEP. What are they?
I logged on to find out about the Jelmax ceasing production thing. Interesting thread.
The BIG advantage the BG N has over the Oscon is the non-polarised nature of it and the large values and the higher voltages.
Can anyone tell me if there is a good outlet for Audio Oscons with a good range of values in the UK or who will ship to here.
Thanks.
CHS said:
Hi CHS,
No I don't have a pre amp either. I used a VR to decide what my 'happy' listening level was, then measured the resistance and soldered two Vishays at the amp end of each interconnect. The alps pot I was using sounded very muddled and thick and wooly and caused a great deal of hiss. I tried lower values of Pot, down to 10K, was about or just the same, tired paralleling resistors which increased the hiss. A much better sounding way is to use fixed series resistors and just use the pot as the shunt. Though I wouldn't do that on my reference sysem as even the shunt has an audible effect (deterioration) on sound quality. Many years aog I ran in vishay and metal oxide resistors for a few months, all 10K Ohms, in parallel or something like that. Then I fitted them to a rotary switch as an L-pad, 10K fixed element, 10K shunt. One set of vishay and one set of metal oxide. Did a listening test. Easy to tell which was the Vishay, the metal oxide was a bit foggy, grainy and a bit warmer due to the thicker muddled bass. The vishays gave plucked strings a very fast leading edge and made metal sound more, well, metal. I then used some metal oxides as the shunt with series vishays and compared with the pure vishays, again easy to tell the difference. Maybe 20% of the degradation of metal oxide in series, but worth not using them as shuts either, if money allows.
My problem now is to make a volume control that has no deleterious effect. I have purchased many vishays and two Shallco 48 way switches. I thought of one problem straight away. Each time I change volume I'll be using a relatively un-used and therefore not run in shunt resistor, and that will sound bighter and thinner. And it would be nice to have the volume control centralised which means electronics after it to stop the bass and treble attenuation effect. I guess it's about which compramise to choose.
Re the black gates, do you mean parallel a BG NX with a standard BG F to reduce the coarseness?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.