Best electrolytic capacitors

psu caps are also in the signal path, they form zeros or virtual ac grounds....unavoidable, perhaps ac coupling caps is more doable...

poles and zeroes in case you are not aware are all over our amp circuits...

poles are responsible for lowering gains as frequency went up, while zeros are responsible for gains going down as frequency went down.....

we live with poles and zeroes in our everyday amps whether we like it or not...
 
poles are responsible for lowering gains as frequency went up, while zeros are responsible for gains going down as frequency went down.....
I wouldn't phrase it like this, firstly as this sort of thing really applies to first-order zeroes and poles. Also its perfectly possible to have a zero in a low-pass circuit and a pole in a high-pass circuit (RIAA filter springs to mind for a zero in a low-pass).


Perhaps a better description is that first order zeroes change the response slope in a concave way, poles in a convex way.



Once you move to second order poles there is Q to consider - a pole pair may be highly resonant making a narrow peak in the spectrum, a zero pair can be a deep notch.
 
Yes mundorf MLytic is pretty good, the other that I like is Nichicon Super Through the higest grade version. Played with Silmic ll many times & I don't really like them especially the highs un natural & shut inn. Nichicon FG also not bad.

I tried super through after some weeks of burnin. I indeed like them better than silmic Ii as well.

The Silmics give music a rhythm and nice macrodynamic where normal caps (incl. Elna for audio or even Nichicon MUSe) sound flat, lifeless, sterile. So when you get the Silmic in you think: Hey, we got music ! Rhytmn, color, space, nice....

But when you put the super through in, you find that there is a certain dirtyness and roughness in the mids and treble of the silmic, that the bass is a bit bloated, that the resultion and grainess cant match super through.

ST sounds just cleaner, more micro details, but as well very musical...its a bit like going from a Linn Lp12 to a 120kg Mass Record Player....quieter, calmer, blacker backround.

I will continue the comparison and will do some more shootouts like Mlytic va. Superthrough etc.

Other question: Has anyone tried to put a PP Dc-link in parallel to ST ? The application data sheet of the Lt4320 I am using here even recommend a small PP cap.... The paralleling of caps of different styles was a game 25 years ago, but I never liked the Gulasch coming out of this...but maybe this time ?
 
Last edited:
Disabled Account
Joined 2019
imo a fast esr cap, whatever the ideal diode bridge should be always avoided whatever it's diodes //film cap or film cap // to the first smoothing cap after the diodes bridge, not sure LT4320 changes anything on this...
best is still to work around the load with caps and find your equilibrium near the diode bridge with // or not lytics only. the smoothing cap should be always slower imho and from what I experienced on normal devices (i.e. not exotic sota designs or fast cap of smps) than the following cap towards the load. and if a smoothing cap then a reservoir cap, so splitted by a L or a R, putt the second cap with ESR or with improved ESR by // in that area in relation to the first area just behind the diodes.


of course it is basic but it works quite well often enough. That doesn't mean you don't want to try // caps if the smoothing and reservoir main cap is the same : I refurbished amps with 10% of the value with very good ESR one // to a slower one with such typology with success.
Ok try the Panasonic FC 1200/50V big diameter low height - there are two different size in the Pan catalog- and see if the musical of the ST has not an added tempo, impact on percussions and added clarity. My feeling, why not tried with the ST ! So here the ST should be around 10 000 uF more or less.
 
Last edited:
KG Super-throughs only come in snap-in style and Silmic II only come in wire leads, correct? How can you compare the two in situ on the PCB? I’m just asking for a clarification...

Thanks

ST come im all kind of comfiguration, I used normal snap in and soldered it into my PcB...where I can solder as well wire leads in, where is the problem ? Nevertheless my Silmic are 3300uF/50 V, so they have thick little wires only to be aoldered into the same snap in soldering wholes...

I compare them by simply build copies of the lt4320 and filtering bank which I can i can exchange within 10sec.
 
I tried super through after some weeks of burnin. I indeed like them better than silmic Ii as well.

The Silmics give music a rhythm and nice macrodynamic where normal caps (incl. Elna for audio or even Nichicon MUSe) sound flat, lifeless, sterile. So when you get the Silmic in you think: Hey, we got music ! Rhytmn, color, space, nice....

But when you put the super through in, you find that there is a certain dirtyness and roughness in the mids and treble of the silmic, that the bass is a bit bloated, that the resultion and grainess cant match super through.

ST sounds just cleaner, more micro details, but as well very musical...its a bit like going from a Linn Lp12 to a 120kg Mass Record Player....quieter, calmer, blacker backround.

I will continue the comparison and will do some more shootouts like Mlytic va. Superthrough etc.

Other question: Has anyone tried to put a PP Dc-link in parallel to ST ? The application data sheet of the Lt4320 I am using here even recommend a small PP cap.... The paralleling of caps of different styles was a game 25 years ago, but I never liked the Gulasch coming out of this...but maybe this time ?


What type of sound are you looking for ? Before going further I would recommend especially for SS preamp or power amp to try out the Shindengen D4SBN20 or D6SBN20 bridge diodes. Smooth & grainless . You'd be surprise.

Cheers
 
What type of sound are you looking for ? Before going further I would recommend especially for SS preamp or power amp to try out the Shindengen D4SBN20 or D6SBN20 bridge diodes. Smooth & grainless . You'd be surprise.

Cheers

I am looking for no sound. So, like a piece of wire ideally. Thx for the hint...I tried all kind of Diodes, at least 20 different types incl. germanium, Sic, Schottky, Hexfred, Bridge Configuration as well as Full Wave etc...So far, I like for the filament lt4320 best and germanium in Full Wave for the DAC. For HV, I use 872A to get the 1000V and for the driver I use RGN2004 mesh.

But this gets a bit offtopic...I am very glad having found the super through. I have a few challenger still in the house, but I think I will stay with them.
 
Last edited:
No sound Blitz ??? Too bad even different wire will affect sound. Once I discovered by accident when I build Mr Pass Boz preamp one was point to point they other build on pcb for a friend parts all the same on both, guess what, the point to point just sounded smoother & better, can't explain, all I thought off was is it the way the electron flowing on solid core wire behaves differently from electrons flowing on flat pcb traces ???

Cheers
 
Sure...i know...the lt4320 with caps on board is the only pcb around...amd I am building my own cables like 25 years now...

Yes, no sound would be my ideal, because I simply want no masking of details or use components to tune sound...but this seldomly can be achieved, I think we talk about the same. But if your system gives you a deep look into the recording in a realistic way, that it makes your heart beat faster...and if a component helps to retrieve new, musical information without doing a new trade-off...then I know I am on the right path...only my personal, completely subjective approach.