There is no best sounding capacitor Jesper. Since your aware of each caps pros & cons you can combine them like soem in PS & some in decoupling to achieve a sound that you like.
Re KZ only go for 50v version cause 25v version of the same value sounds dreadful. Btw he best sounding KZ that I found are the 22 & 33uf 50v version. In general my findings are 50v versions mostly sounds better then lower voltage ones.
Cheers
Re KZ only go for 50v version cause 25v version of the same value sounds dreadful. Btw he best sounding KZ that I found are the 22 & 33uf 50v version. In general my findings are 50v versions mostly sounds better then lower voltage ones.
Cheers
I have found that this is the case with KZ when they are undersized for that specific position or if the main power supply is not large enough. Try bumping up the size of the KZ caps if you get this. Or even increase the size of the power supply post-regulator caps if it's in a low voltage audio circuit.
is your judgement made after an well over 100 hours run in ??
otherwise it has no sense to judge an capacitor
.
@sumotan: Hi .. & thanks for your feedback. Although I am aware that there likely isn't any "one good" capacitor looking at the Denafrips Pontus - and remembering earlier designs with many of just one capacitor type - I was thinking/hoping that maybe a good overall capacitor could be found - and that one of you might have found this ONE (as I haven't myself).
Interesting also that the 22uF & 33uF 50V KZ versions are the ones you consider the best ... I reckon this is for decoupling as they do not provide much capacity for stabilizing a PSU voltage?
@ainami: Thank you also for your feedback ... Well, actually the capacitors I am using are the 1000uF/50V KZs combined with Nichicon KG and GU - altogether about 54000uF for the input stage per channel and 72000uF for the output stage per channel.
My personal thought about why the sound is a little screamy is that the 1000uF/50V KZs are rather large so I guess that their HF impedance probably is higher than the capacitors I used previously. And since the input stage has an appr. 11 MHz - 3dB point and the output stage a 2.2 MHz - 3dB point I reckon this could influence the stability of the circuits. I will see if/when I can find time to measure this.
I in general agree with the notion of combining capacitors, however, in the long term I find it to be pretty hard on the PCBs soldering on and off so many times. Also, from a systematic point of view, I find it to be suboptimal to do it in this way as the circuitry has to warm up again, the solder joints and/or the new capacitor has to be run-in - and this may make hands-on comparisons "tricky". So to me at least it would be nice to find either ONE capacitor (e.g. one for digital circuitry, one for analog circuitry) that was the "Best", or a method to allow combination of suitable capacitors in a simple way.
With regards to a method of combining capacitors I have been thinking of maybe using something like the Digikey analog discovery II and its impedance measuring feature to combine the appropriate sizes of e.g. KZ, Cerafine & KG/GU to arrive at a combination that yields a straight impedance curve from "DC to infinity" ... if this is at all relevant ... Well, thinking aloud here ...
Anyway, I will be off on my Christmas vacation now so if there are any comments to this I won't be reading them until ~ 28 December. Merry Christmas to you ...
Cheers,
Jesper
Interesting also that the 22uF & 33uF 50V KZ versions are the ones you consider the best ... I reckon this is for decoupling as they do not provide much capacity for stabilizing a PSU voltage?
@ainami: Thank you also for your feedback ... Well, actually the capacitors I am using are the 1000uF/50V KZs combined with Nichicon KG and GU - altogether about 54000uF for the input stage per channel and 72000uF for the output stage per channel.
My personal thought about why the sound is a little screamy is that the 1000uF/50V KZs are rather large so I guess that their HF impedance probably is higher than the capacitors I used previously. And since the input stage has an appr. 11 MHz - 3dB point and the output stage a 2.2 MHz - 3dB point I reckon this could influence the stability of the circuits. I will see if/when I can find time to measure this.
I in general agree with the notion of combining capacitors, however, in the long term I find it to be pretty hard on the PCBs soldering on and off so many times. Also, from a systematic point of view, I find it to be suboptimal to do it in this way as the circuitry has to warm up again, the solder joints and/or the new capacitor has to be run-in - and this may make hands-on comparisons "tricky". So to me at least it would be nice to find either ONE capacitor (e.g. one for digital circuitry, one for analog circuitry) that was the "Best", or a method to allow combination of suitable capacitors in a simple way.
With regards to a method of combining capacitors I have been thinking of maybe using something like the Digikey analog discovery II and its impedance measuring feature to combine the appropriate sizes of e.g. KZ, Cerafine & KG/GU to arrive at a combination that yields a straight impedance curve from "DC to infinity"
... if this is at all relevant ... Well, thinking aloud here ... Anyway, I will be off on my Christmas vacation now so if there are any comments to this I won't be reading them until ~ 28 December. Merry Christmas to you ...
Cheers,
Jesper
I personnaly use the KZ 22 uF but in 100V iirc in the last stages of an amp, very good.
But in a voltage PSU... uh.. will not do that -personal tastes- but in a CRC PSU after the resistor, first C being a slower Long Life type... Lelon for instance ! Or why not an ELna Tonorex or even Silmic I don't like... CRC permits some fun !
No all in one cap as Sumo is saying. A forgiving one is the Panasonic FR, but even though there are some place I never use it...
But in a voltage PSU... uh.. will not do that -personal tastes- but in a CRC PSU after the resistor, first C being a slower Long Life type... Lelon for instance ! Or why not an ELna Tonorex or even Silmic I don't like... CRC permits some fun !
No all in one cap as Sumo is saying. A forgiving one is the Panasonic FR, but even though there are some place I never use it...
Sorry...all write about the sound of the capacitors but I don't understand if electrolytics del thread are good only where the signal passes or also on the power supplies.
I usually use Panasonic FC that are found all over the place and have a range that includes a lot of values and voltages, but I am also testing the electrolytics by Wurth electronics.
I usually use Panasonic FC that are found all over the place and have a range that includes a lot of values and voltages, but I am also testing the electrolytics by Wurth electronics.
Last edited:
The only high end gear I seen filled with full red cerafines silmics. Black gates ..are old marantz.
cd 94ii and I heard a few other old cd players have them .
At the very most the Japanese high end have nippon chemicon (mostly)made for kenwood onkyo and so forth . The most expensive cd player by luxman d10x is just std nippon chemicon with maybe a few well placed boutique parts (like most high end equipment.) Cost like 20000 dollars .
I had my own share of these boutique parts thousands of dollars ,and I absolutely careful with boutique parts because i cannot reliably guess the outcome . It just might turn into a piece of junk after many months of use and realise oh oh ..rubbish . They can become dark . Bleached , dull. First impressions is unreliable . Boutique caps are just too colourful and can waste alot of time figuring out which I don't want to waste . And high end nichicons are the worst , bleached .dark . dull. Ringing. All types of funny issues after burn in . And I am very familiar all the parts sound ..but risk failure .open up the box . Solder and wait again ..? Too much time .nothing better to do . Just a hobby. Wasted years btw.got pawned by standard off the rack equipment all day long .
I trust and respect Nelson Pass who sold and designed way more amplifier than most of us here .I use mainly some silmics , elnas ,black gates if I can afford
and hardly other boutique parts . But mostly good standard parts . The biggest difference to.sound is topology. And decent parts selection. And I EVENTUALLY made some very decent projects that respected golden ears approved. On a winning streak here.grin
Just my 2cents worth
cd 94ii and I heard a few other old cd players have them .
At the very most the Japanese high end have nippon chemicon (mostly)made for kenwood onkyo and so forth . The most expensive cd player by luxman d10x is just std nippon chemicon with maybe a few well placed boutique parts (like most high end equipment.) Cost like 20000 dollars .
I had my own share of these boutique parts thousands of dollars ,and I absolutely careful with boutique parts because i cannot reliably guess the outcome . It just might turn into a piece of junk after many months of use and realise oh oh ..rubbish . They can become dark . Bleached , dull. First impressions is unreliable . Boutique caps are just too colourful and can waste alot of time figuring out which I don't want to waste . And high end nichicons are the worst , bleached .dark . dull. Ringing. All types of funny issues after burn in . And I am very familiar all the parts sound ..but risk failure .open up the box . Solder and wait again ..? Too much time .nothing better to do . Just a hobby. Wasted years btw.got pawned by standard off the rack equipment all day long .
I trust and respect Nelson Pass who sold and designed way more amplifier than most of us here .I use mainly some silmics , elnas ,black gates if I can afford
and hardly other boutique parts . But mostly good standard parts . The biggest difference to.sound is topology. And decent parts selection. And I EVENTUALLY made some very decent projects that respected golden ears approved. On a winning streak here.grin
Just my 2cents worth
I personally do not trust Wuerth.... might be just be a prejudice, but the origin of that company lies in screws, nuts & bolts. They still provide annual calendars with nude postergirls to workshops. Despite the sexy red sleeve, doesn't convince me at all... I mean, look at their datasheets ��
I would not reject my car tyres due to calenders with posters girl seen in every garage
Wuerth obviously bought a lot of electronic parts producers. Their portfolio is expensive but quality is first class imho.
The same can be said about their support in case you are an industrial customer.
Their coils are in the same ballpark as coilcraft, pulse, TDK etc...
For the hobbyist there are cheaper alternatives.
Wuerth obviously bought a lot of electronic parts producers. Their portfolio is expensive but quality is first class imho.
The same can be said about their support in case you are an industrial customer.
Their coils are in the same ballpark as coilcraft, pulse, TDK etc...
For the hobbyist there are cheaper alternatives.
Last edited:
> Wuerth.... might be just be a prejudice, but the origin of that company lies in screws, nuts & bolts. They still provide annual calendars with nude postergirls to workshops
I see what you mean (don't look). Not true "nude" but maybe my porn filter is engaged. However Ridgid has been doing this for a very long time.
I see what you mean (don't look). Not true "nude" but maybe my porn filter is engaged. However Ridgid has been doing this for a very long time.
Attachments
If its audio porn you want heres the balanced direct coupled 8 x 1704K 24/192 dac stage discrete jfet with 90ma biased fet output stage, nude Vishays, Blackgate N and NX Back to back for power supply, Relcap teflon bias bypass, MIT polystyrene caps with OPA 189 DC servos. The light blue trim pots are Bulk foil trimpots. Jung super regulators as preregulators before the shunts, hexfet bridges. discrete fet shunts regulators. R core transformers 240,000 uf CLC supply. Line stage and phono are identical in build components - now I have no money for software or records.......... .
Attachments
Last edited:
The dac, line amp, and phono and shunt regulators are all Erno Borbely’s discrete jfet design. The raw power supply boards are DiyAudio power amp power supply boards that got used for line level electronics. I sourced the parts, built, and tested the boards. The chassis, mechanical design and assembly and wiring were done to a very high standard by Les Bordelon (LB Audio). They are all one of a kind diy. Here is the balanced line amp with remote controlled Bent audio reed relay stepped resistor volume control boards, the dac power supply is a separate chassis shown here, the preamp and phono power supply chassis was in post 1693 pictures. Theres over 100 of 3 inch high TO 220 heatsinks between the 3 units and their power supplies. Each runs high bias class A with 30 watts per channel dissipated. Direct coupled, all discrete Toshiba / Hitachi jfet/fet. over 80 % of the 2 chassis for each is power supply.
Attachments
Last edited:
I hope that Siemens MKV was not build before 1983, cos till 83 they were very toxic!
According to sources I found, MKV was all PCB free.
https://www.chemsuisse.ch/de/?optio....download&catid=97&id=80&Itemid=1000000000000
https://www.chemsuisse.ch/de/?optio....download&catid=97&id=80&Itemid=1000000000000