Hi,
& want to use sanyo oscons for decoupling in digital power supply.Their impedance is small in wide frequency range.Does't they need parallel non electrolytics capacitors(like black gates)? & don't want to use wima because small impedances of both capacitors can be a result of high frequency oscillation and this change sound.Or & need just connect one cap with 1ohm resistor ?
Does anybody have experience with these caps?
Thanks
& want to use sanyo oscons for decoupling in digital power supply.Their impedance is small in wide frequency range.Does't they need parallel non electrolytics capacitors(like black gates)? & don't want to use wima because small impedances of both capacitors can be a result of high frequency oscillation and this change sound.Or & need just connect one cap with 1ohm resistor ?
Does anybody have experience with these caps?
Thanks
Here is optimal solution for digital chip decoupling.
Place on each chip 3 caps: small NPO ceramic (1000-2000pF is OK) + small X7R ceramic (0.1uF or so), AS CLOSE AS POSSIBLE to Vcc -Gnd pins. Both are smd - preffered, but small through-holes also works fine. Third capacitor - small tantalum 10-20uF x2...3 times voltage rated (or oscon, for your taste). Also closed to ceramics. Trace | Vcc pin -> NPO -> X7R -> tantalum or oscon | ends with ferrite bead (smd bead is best). From bead starts trace to voltage regulator, this trace can be relatively long. All cap's grount legs tied together to digital ground plane in shortest way.
Avoid film capacitors in digital circuits, most of them are wounded and you will get unpredictable ringing/noise.
cheers
Vitaly
Place on each chip 3 caps: small NPO ceramic (1000-2000pF is OK) + small X7R ceramic (0.1uF or so), AS CLOSE AS POSSIBLE to Vcc -Gnd pins. Both are smd - preffered, but small through-holes also works fine. Third capacitor - small tantalum 10-20uF x2...3 times voltage rated (or oscon, for your taste). Also closed to ceramics. Trace | Vcc pin -> NPO -> X7R -> tantalum or oscon | ends with ferrite bead (smd bead is best). From bead starts trace to voltage regulator, this trace can be relatively long. All cap's grount legs tied together to digital ground plane in shortest way.
Avoid film capacitors in digital circuits, most of them are wounded and you will get unpredictable ringing/noise.
cheers
Vitaly
hi folks,
looks seems out of topic, i plan to use sanyo-oscon, many people said is good in digital circuit and good as its gets.
the more i know aka i dont even know, has any mass product (or high-ender) use this caps in their circuit or for now only the DIYer know the recipe.
happy weekend,
datoek jacques
looks seems out of topic, i plan to use sanyo-oscon, many people said is good in digital circuit and good as its gets.
the more i know aka i dont even know, has any mass product (or high-ender) use this caps in their circuit or for now only the DIYer know the recipe.
happy weekend,
datoek jacques
DIYers are free from target-price limitations, so they csn use exotic and expensive parts. This is like going to the end. Manufacturer don't goes to the end, he use cost-effective solutions. In digital circuits, ordinary cheap capacitors are doing decoupling job well enough. "Extremely well" costs much more and unhearable for typical consumer. Second reason - relatively low power and speed requirements for digital chips in audio devices.
On second side, you can find a lots of oscons at modern computer boards. Newest power-hungry, high-speed digital chips really needs an extremelly good decoupling, and just don't work with ordinary 'lytics.
cheers
Vitaly
On second side, you can find a lots of oscons at modern computer boards. Newest power-hungry, high-speed digital chips really needs an extremelly good decoupling, and just don't work with ordinary 'lytics.
cheers
Vitaly
Hi,
keep the highest frequency decoupling as close to the supply pins as possible and keep the returns to ground very short. Use the lowest inductance (i.e. ceramic) closest to the pins.
In other words I agree with Vid.
Oscon seem expensive when all you want to recover is 0s & 1s and avoid the glitches, but maybe they're the bees knees.
keep the highest frequency decoupling as close to the supply pins as possible and keep the returns to ground very short. Use the lowest inductance (i.e. ceramic) closest to the pins.
In other words I agree with Vid.
Oscon seem expensive when all you want to recover is 0s & 1s and avoid the glitches, but maybe they're the bees knees.
Hello
I found that using OSCONS to decouple the filter and DAC chips made a noticeable difference to the sound. Details are much better.
BTW, I am thinking to replace the ageing Cerafine audio output coupling caps in my Philips player with Panasonic FCs. Anyone could tell me are the Cerafines better than Panasonics FCs sonically as coulpling caps?
ckt
I found that using OSCONS to decouple the filter and DAC chips made a noticeable difference to the sound. Details are much better.
BTW, I am thinking to replace the ageing Cerafine audio output coupling caps in my Philips player with Panasonic FCs. Anyone could tell me are the Cerafines better than Panasonics FCs sonically as coulpling caps?
ckt
Re: Hello
As coupling or decoupling? As coupling, Cerafine is not good, I don't like it. As decoupling, Cerafine is better than FC IMO. On oscilloscope, FC is worse than most other hi-en caps. On oscilloscope Cerafine is better than BGF as decoupling in power supply, but I prefer the BG.
diyman said:
As coupling or decoupling? As coupling, Cerafine is not good, I don't like it. As decoupling, Cerafine is better than FC IMO. On oscilloscope, FC is worse than most other hi-en caps. On oscilloscope Cerafine is better than BGF as decoupling in power supply, but I prefer the BG.
Coupling Caps
I am referring to the final coupling cap between the opamp and the CD output. Would the BG NX be better than both the Cerafine and the FCs? I've heard that if I install the BGs, I have to play regularly because if I stop for a while, I would have to go through the whole process of running them in again, is that true??
ckt
I am referring to the final coupling cap between the opamp and the CD output. Would the BG NX be better than both the Cerafine and the FCs? I've heard that if I install the BGs, I have to play regularly because if I stop for a while, I would have to go through the whole process of running them in again, is that true??
ckt
Oh yes, BGNX is very good for coupling. Cerafine is bad IMO. Haven't tried Panasonic FC for coupling, but I don't think I need to try But I'll bet it is better than Cerafine for coupling.
BG needs burn-in. But once done, you don't have to go through the whole process again. Fortunately you will likely get good result at first install so you don't need to worry about the burn-in issue, unless you want to join a blind test competition
I don't know what makes some caps sound more musical. For DAC and digital, we need precision, so we don't want supply ripples (or jitter) at mega-hertz frequencies. OSCON and Cerafine is good. But in analog, less ripples sometimes means less details and sonic (may be a Q or ESR issue). By listening, my preference is BG, which is low ESR (NX is low Q), and for ripples issue I parallel them with small ceramics (decoupling) or MKP (coupling).
But I'll bet it is better than Cerafine for coupling.BG needs burn-in. But once done, you don't have to go through the whole process again. Fortunately you will likely get good result at first install so you don't need to worry about the burn-in issue, unless you want to join a blind test competition
I don't know what makes some caps sound more musical. For DAC and digital, we need precision, so we don't want supply ripples (or jitter) at mega-hertz frequencies. OSCON and Cerafine is good. But in analog, less ripples sometimes means less details and sonic (may be a Q or ESR issue). By listening, my preference is BG, which is low ESR (NX is low Q), and for ripples issue I parallel them with small ceramics (decoupling) or MKP (coupling).
NO! Do non use a small NPO and a X7R somewhat ceramic.
Definitely yes.
Just use a 1µ X7R 0603 Chip capacitor as close as possible to the VCC - GND pins. It's simply the best solution. Forget anything about the BlackGates, they're not able to decouple anything in the high frequency range! OSCONs are fine to some MHz but not more. The larger the capacitance in a given (smallest!) case the better.
For signal coupling purpose try to get rid of any cap in the signal path. If you want to try a cap a good MKP is better than polarized 'lytics. The OSCONs are only avialable in small values and a cap like the ICW is superior and not much more expensive. The only trade off is that a foil cap is much larger than a 'lytic.
Yes.
20-30dB improvement over a typical 1nF NP0 and 100nF X7R parallel decoupling. The impedance for frequencies above some to some 10th MHz (above the series resonant frequency) is only dependant of the parasitaric inductance of the cap; this is only dependant from the size.
For frequencies below the series resonant frequency the noise filtering depends on the capacitance only. And this is what you see, when using the higher cap value.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.