There's OS-CON and there's OS-CON
Just curious, while people often take care to specify the exact
type of Black Gate, Rubycon and whatever, I think I have never
seen anybody saying which type of OS-CON they refer to. There
are several types. The ones I know of are those available here
which are the SA/SC type and the special audio model SG. In
addition there are two SMD versions, SM and SV. All of these
are Sanyo, by the way.
Just curious, while people often take care to specify the exact
type of Black Gate, Rubycon and whatever, I think I have never
seen anybody saying which type of OS-CON they refer to. There
are several types. The ones I know of are those available here
which are the SA/SC type and the special audio model SG. In
addition there are two SMD versions, SM and SV. All of these
are Sanyo, by the way.
That is true regarding OsCons. The ones I was using is probably the lowest grade. Yet I noticed that other people use similar type (in pictures) so before coming to any particular conclusion all available options. But the ones I tried are definitely no good (to me😉
setmenu said:.
I am at present fiddling around with some very,very tiny SOT packaged regs, trying to find a way of best mounting them on the bottom of the bypass caps...
Oscons in fact..
The differing opinions in 'best approach' tend to make my electronics newb head spin
It is all fascinating none the less😀
At certain point, I got also confused with so many parts choices and decided to check everything myself. That's why I built this prototype, where by using special sockets, I can easily swap caps and resistors as well regulators for sonic evaluation. It was quite a revealing experience.
Please not that non polarised BG caps are still directional, and they definitely sound different depending on their orientation in a circuit.😉
Attachments
Please not that non polarised BG caps are still directional, and they definitely sound different depending on their orientation in a circuit.😉 [/B][/QUOTE]
---------------------------------------------------------------------
Which direction sounds best to you. I usually use the long lead for signal input.
And how does the direction of your ouptput resistors affect your conclusion?
---------------------------------------------------------------------
Which direction sounds best to you. I usually use the long lead for signal input.
And how does the direction of your ouptput resistors affect your conclusion?
My experience is it better to have a large ground plane for analog and Digital. It just make the design easer to implement and reduced digital switching noise and increase the analog performance. Also, it's help if the power traces or low impedance too.
This can be accomplished on a prototype with p2p wiring. 3M makes a copper foil tape with can be attached to the bottom or top side of the board. Of course you can place a piece of Mylar between bottom and the foil if you have p2p connections their. The copper tape is also helpfull when resolving emi problems.
So does any one know if Black Gate has a Web site, I have never seen one.
This can be accomplished on a prototype with p2p wiring. 3M makes a copper foil tape with can be attached to the bottom or top side of the board. Of course you can place a piece of Mylar between bottom and the foil if you have p2p connections their. The copper tape is also helpfull when resolving emi problems.
So does any one know if Black Gate has a Web site, I have never seen one.
fmak said:
Which direction sounds best to you. I usually use the long lead for signal input.
And how does the direction of your ouptput resistors affect your conclusion?
Short lead is marked with non polar print on a sleeve. I prefer short lead on the input (long lead to the output). As to the bypass positioning, I prefer short lead on a pos supply, long lead to the ground.
The short lead marks the start of the foil and this is in line with my other observations regarding caps orientation. Usually, the outer foil should be directed towards the output or connected to the ground.
As to the resistors orientation, it seems like there is a slight variation in different orientation, but so far I didn't pursue it more extensively. It is too inconsistent and time consuming that I let it go for now😉
Direction
Directionality is also true for some film caps, TRW use to mark horizontal to the end of the cap the end. That end would go to lowest end impedance part of the circuit, acting as a shield. You can check to see if the BG can is connected to one of the leaded, if so do the same.
Directionality is also true for some film caps, TRW use to mark horizontal to the end of the cap the end. That end would go to lowest end impedance part of the circuit, acting as a shield. You can check to see if the BG can is connected to one of the leaded, if so do the same.
Actually, a linkt to the Black Gate webside was once posted in
a thread discussing these caps. I cannot remember if there were
any actual datasheets, but there was at least some more or less
credible description of how they work.
Unfortunately, I cannot remember the URL and I couldn't come
up with any good search terms to find the thread, but by saying
that I remember there is such a thread, perhaps it can trigger
somebody elses memory to remember even which thread it was.
The thread might have been called something like "Black Gates,
myth or reality" or something similar. Anybody remember it?
It might have been Per-Anders starting it, or at least I know he
was involved.
a thread discussing these caps. I cannot remember if there were
any actual datasheets, but there was at least some more or less
credible description of how they work.
Unfortunately, I cannot remember the URL and I couldn't come
up with any good search terms to find the thread, but by saying
that I remember there is such a thread, perhaps it can trigger
somebody elses memory to remember even which thread it was.
The thread might have been called something like "Black Gates,
myth or reality" or something similar. Anybody remember it?
It might have been Per-Anders starting it, or at least I know he
was involved.
http://www.blackgate.jp/english.htm
A bulk of Jelmax Technical Reports can be downloaded from PartsConnexion’s site:
http://www.partsconnexion.com/resources.php
BG was even kind enough to send me some missing Reports that I was interested for… But there is always something left unsaid. At the other hand these caps sounds extremely good, used either like PS or signal carrying caps.
Pedja
A bulk of Jelmax Technical Reports can be downloaded from PartsConnexion’s site:
http://www.partsconnexion.com/resources.php
BG was even kind enough to send me some missing Reports that I was interested for… But there is always something left unsaid. At the other hand these caps sounds extremely good, used either like PS or signal carrying caps.
Pedja
Peter:
>The OsCons I was using have no grade marked on them, just the value and are sourced from Chinese seller.<
The SP grade has "SP" marked on the side, integrated into the whitish negative polarity band. I believe that the SG was likewise marked on the side, but I don't have any open samples at hand (I'd have to search), and can't confirm the SG markings with 100% certainty (the lettering and polarity band will be gold on the SGs).
These two are the only grades of Sanyo OS caps that use decent-quality copper in the leads, and according to Sanyo, the chemistry has also been tweaked somewhat for audio use (compared to the other Sanyo OS caps). Also according to Sanyo, they should be powered up for a pretty long time (weeks) after they have been subjected to the heat of a soldering iron.
The technical explanation given to me for the above was that the OS electrolyte should remain tar-like in texture for best operation, but partially turns gasseous under the heat of soldering. It takes time (with power applied) for the gases to be re-absorbed into the tar.
BTW, years ago I was far more focused on componentry than I am now. Today - I believe largely due to my improved skills in the areas of schematic, pcb and harness design - I find that I can use a much wider range of componentry and still get good sonic results. It is surprising what a clever schematic and matching board layout can do to correct a variety of sonic problems that you may have initially thought were caused by other factors (such as component choice).
I also agree with the Jim Wilson about the desireability of designer-dependent features like a ground plane (and don't use it only for shielding!), low-impedance power traces, and more. Or perhaps I should say "potential desireability", as an ineptly designed ground plane can cause more damage than good. Once again, a ground plane is a tool, and the results depend as much on the person using the tool as the tool itself.
hth, jonathan carr
>The OsCons I was using have no grade marked on them, just the value and are sourced from Chinese seller.<
The SP grade has "SP" marked on the side, integrated into the whitish negative polarity band. I believe that the SG was likewise marked on the side, but I don't have any open samples at hand (I'd have to search), and can't confirm the SG markings with 100% certainty (the lettering and polarity band will be gold on the SGs).
These two are the only grades of Sanyo OS caps that use decent-quality copper in the leads, and according to Sanyo, the chemistry has also been tweaked somewhat for audio use (compared to the other Sanyo OS caps). Also according to Sanyo, they should be powered up for a pretty long time (weeks) after they have been subjected to the heat of a soldering iron.
The technical explanation given to me for the above was that the OS electrolyte should remain tar-like in texture for best operation, but partially turns gasseous under the heat of soldering. It takes time (with power applied) for the gases to be re-absorbed into the tar.
BTW, years ago I was far more focused on componentry than I am now. Today - I believe largely due to my improved skills in the areas of schematic, pcb and harness design - I find that I can use a much wider range of componentry and still get good sonic results. It is surprising what a clever schematic and matching board layout can do to correct a variety of sonic problems that you may have initially thought were caused by other factors (such as component choice).
I also agree with the Jim Wilson about the desireability of designer-dependent features like a ground plane (and don't use it only for shielding!), low-impedance power traces, and more. Or perhaps I should say "potential desireability", as an ineptly designed ground plane can cause more damage than good. Once again, a ground plane is a tool, and the results depend as much on the person using the tool as the tool itself.
hth, jonathan carr
Thanks Jonathan,
that was interesting. For once we get a technical reason
from a manufacturer for burning in a component. There are
probably other such cases , but they seem to be rare.
that was interesting. For once we get a technical reason
from a manufacturer for burning in a component. There are
probably other such cases , but they seem to be rare.
jcarr said:
I noticed that. When looking through your older posts on AA, I perceive a similar attention to componentry (that I have now) , but your recent comments here, show less regard in that matter.
Yet to me, at this point, it still seems not comprehendable how in more complicated and advanced circuit, componentry plays a lesser role. To be clear, I don't play with complicated circuits. Most of my current boards contain only few components, and I can't avoid not having to pay attention to my choices if first rate performance is desired. Could it be that with more complicated designs, the amount of components and what's going on a PCB simply masks the effect of a single part and although such circuits prevail in certain aspects of sonic retrival, yet, they don't posses delicacy and intimacy that most simple circuits produce?
Peter
You should try some of the SP series OSCon's, in my view.
Sonically these are very different from the standard parts. They are also, in my experience, better than the earlier SG parts, which I was less keen on.
I'm with Jonathan here, the SP series' sound is almost the complete antithesis of the comments you have made about Oscons.
Andy.
You should try some of the SP series OSCon's, in my view.
Sonically these are very different from the standard parts. They are also, in my experience, better than the earlier SG parts, which I was less keen on.
I'm with Jonathan here, the SP series' sound is almost the complete antithesis of the comments you have made about Oscons.
Andy.
Peter:
>your recent comments here, show less regard in that matter.<
No, I am still attuned to what components can do to the sound, but now I am also aware that other issues can play a much bigger role. And as I mentioned in my previous post, I observe that sometimes we may be blaming component choice for sonic issues that may have more to do with schematic or board design. What I spend a lot of effort now is to analyze what sort of operating environment a given active or passive component needs to provide acceptable performance, and then figure out how to create that environment. And as a personal challenge, I find this approach more satisfying than component tasting.
Now _if_ I should ever feel that my present, more design-oriented approach is beginning to peak out and major steps forward in terms of sonics become harder and harder, at that time I _may_ once again focus more acutely on component choice. OTOH, I could simply resolve to work and study harder on improving my design abilities. Only time will tell.
>Yet to me, at this point, it seems still not comprehendable how in more complicated and advanced circuits, componentry plays a lesser role.<
If you can provide the componentry with an operating environment where their flaws are less likely to be evident, their sonic impact will be mitigated as well. Should circumstances force you to recite the Ten Commandments convincingly while running the 100 meter dash in under 12 seconds, you'd need to be a decent athlete as well as a practiced orator. But if you could ask someone else to recite the Ten Commandments while you did the running (or vice versa), the total task wouldn't even be a challenge worth mentioning.
>Most my current boards contain only few components, and I can't avoid not having to pay attention to my choices, if first rate performance is desired.<
With circuitry of few components, the functions required of each component tend to be complex (and often ill-defined), and so each component needs to have a multitude of characteristics that are all very good. With more complex circuits, you can break the desired functions into smaller, better defined tasks, and you can then begin to use components that only need a few characteristics that are outstanding (and frequently obvious).
>Could it be that with more complicated designs, the amount of component and what's going on a PCB simply masks the effect of a single part and although such circuits prevail in certain aspects of sonic retrival, yet they don't posses delicacy and intimacy that most simple circuits produce?<
Depending on the audio system (including RF conditions, acoustics, and setup) and personal preference, of course anything is possible. But according to reports from professional reviewers, audio dealers and private audiophiles (as well as what I hear with my own ears), the latest Connoisseur 4-2 and 5 clearly outperform all of the older models when it comes to sonics as well as measured performance. The new designs have lower noise, faster overload recovery, better low-level detail, more complex timbres, more natural tonal balance, better soundstaging, greater intellectual and emotional involvement and so on. In fact, it would be very difficult for me to think of any sonic aspect of a older design like the 3.0 or 2.0 which I prefer over the 4-2 and 5. Circuitry-wise, the new designs are considerably more complex than the old ones, so the "simple is best" philosophy has not held up to at least _my_ personal scrutiny.
But again, you're not me. And the point should not to defend your own subjective choices and personal aesthetics (which you really don't need to justify to anyone but yourself), but to make great-sounding audio designs in the best manner that you can.
If your guiding star doesn't lead you naturally in the direction of complexity, complexity shouldn't be the path that you take. In the end, you should go with whatever approach is most rewarding for you.
regards, jonathan carr
>your recent comments here, show less regard in that matter.<
No, I am still attuned to what components can do to the sound, but now I am also aware that other issues can play a much bigger role. And as I mentioned in my previous post, I observe that sometimes we may be blaming component choice for sonic issues that may have more to do with schematic or board design. What I spend a lot of effort now is to analyze what sort of operating environment a given active or passive component needs to provide acceptable performance, and then figure out how to create that environment. And as a personal challenge, I find this approach more satisfying than component tasting.
Now _if_ I should ever feel that my present, more design-oriented approach is beginning to peak out and major steps forward in terms of sonics become harder and harder, at that time I _may_ once again focus more acutely on component choice. OTOH, I could simply resolve to work and study harder on improving my design abilities. Only time will tell.
>Yet to me, at this point, it seems still not comprehendable how in more complicated and advanced circuits, componentry plays a lesser role.<
If you can provide the componentry with an operating environment where their flaws are less likely to be evident, their sonic impact will be mitigated as well. Should circumstances force you to recite the Ten Commandments convincingly while running the 100 meter dash in under 12 seconds, you'd need to be a decent athlete as well as a practiced orator. But if you could ask someone else to recite the Ten Commandments while you did the running (or vice versa), the total task wouldn't even be a challenge worth mentioning.
>Most my current boards contain only few components, and I can't avoid not having to pay attention to my choices, if first rate performance is desired.<
With circuitry of few components, the functions required of each component tend to be complex (and often ill-defined), and so each component needs to have a multitude of characteristics that are all very good. With more complex circuits, you can break the desired functions into smaller, better defined tasks, and you can then begin to use components that only need a few characteristics that are outstanding (and frequently obvious).
>Could it be that with more complicated designs, the amount of component and what's going on a PCB simply masks the effect of a single part and although such circuits prevail in certain aspects of sonic retrival, yet they don't posses delicacy and intimacy that most simple circuits produce?<
Depending on the audio system (including RF conditions, acoustics, and setup) and personal preference, of course anything is possible. But according to reports from professional reviewers, audio dealers and private audiophiles (as well as what I hear with my own ears), the latest Connoisseur 4-2 and 5 clearly outperform all of the older models when it comes to sonics as well as measured performance. The new designs have lower noise, faster overload recovery, better low-level detail, more complex timbres, more natural tonal balance, better soundstaging, greater intellectual and emotional involvement and so on. In fact, it would be very difficult for me to think of any sonic aspect of a older design like the 3.0 or 2.0 which I prefer over the 4-2 and 5. Circuitry-wise, the new designs are considerably more complex than the old ones, so the "simple is best" philosophy has not held up to at least _my_ personal scrutiny.
But again, you're not me. And the point should not to defend your own subjective choices and personal aesthetics (which you really don't need to justify to anyone but yourself), but to make great-sounding audio designs in the best manner that you can.
If your guiding star doesn't lead you naturally in the direction of complexity, complexity shouldn't be the path that you take. In the end, you should go with whatever approach is most rewarding for you.
regards, jonathan carr
jcarr said:
Well said Jonathan, and of all people here, you are one of the few that can provide the most insight and inspiration when it comes to truly high end design.
Well, I will be doing all I can and as always, I will follow my intuition and personal 'feel'.
And I already, on few occasions, learned that the simplest is not always the best. But it can be pretty damn good.😉
Re: Peter
The ones I tried were SA and SC series from this source: http://hk.geocities.com/wkchenghk2000/orderformEng.htm
ALW said:
The ones I tried were SA and SC series from this source: http://hk.geocities.com/wkchenghk2000/orderformEng.htm
Peter Daniel said:
jewilson said:
WRT ground planes, totally agree. I think it's funny seeing
people evaluating various dig bypass caps such as ceramic
and low esr electros when they don't go directly or almost directly
to a ground plane.
Maybe that's why we have so many differing views on sound
of various bypass configurations.
WRT proto boards, also agree, I usually use a single sided
solid copper covered pcb and xmas tree off that. All the decouple
caps can get a good lo z gnd even though its still P-P wired.
WRT Blck Gate caps, some info.
http://www.blackgate.jp/ebg3.htm
http://www.octave-electronics.com/Parts/cap/bg_tech.shtml
I have to admit that most of this sounds like marketing
bullsh!t. However as ALW also found, these puppies definately
have a very transparent and clean sound unlike most other
electros.
Some (KYW) find them too mechanical and unmusical, this is
also a common thing. I have one friend who will not use Black
Gates, period.
In the end it's all subjective... choose your poison 🙂
I really like Jensen 4 pole electro's, seem to have
a good balance of transparency and playing "music", but have
only tried large HV ones for tube amps.
cheers,
Terry
Jonathan,
If you don't mind, may I ask you two brief follow ups? Was that
tech. explanation on burn in after soldering specific to the audio
grade OS-CONs or all models? Is your experience that it matters
to use the audio grade versions also for digital decoupling, for
analogue decoupling or just for coupling and similar?
Another reflection, probably not very relevant in practice but
nevertheless interesting. The design philosophy you describe
of making an overall more complex design to get "modules"
with more well-defined behaviour/environment for components,
has some similarities with good software engineering where
we tend to break things down (hierarchially or otherwise) into
well-defined modules that make it easier both to design,
debug and understand the program, even if the overall complexity
becomes much greater.
If you don't mind, may I ask you two brief follow ups? Was that
tech. explanation on burn in after soldering specific to the audio
grade OS-CONs or all models? Is your experience that it matters
to use the audio grade versions also for digital decoupling, for
analogue decoupling or just for coupling and similar?
Another reflection, probably not very relevant in practice but
nevertheless interesting. The design philosophy you describe
of making an overall more complex design to get "modules"
with more well-defined behaviour/environment for components,
has some similarities with good software engineering where
we tend to break things down (hierarchially or otherwise) into
well-defined modules that make it easier both to design,
debug and understand the program, even if the overall complexity
becomes much greater.
- Status
- Not open for further replies.