¿ Capacitores rusos, Jantzen o Mundorf ?

A la hora de mejorar un preamplificador (GGP) de Trascendent las pruebas entre capacitores arrojaron un solo ganador.

 

Interesante comparativa en la que para el conocido y reconocido Troels Gravesen la elección fue Mundorf

http://www.troelsgravesen.dk/GG.htm

Nice layout and easy wiring.

The potmeter following the kit was tried and immediately discarded and
replaced by the above 24 step attenuator.
Furthermore I had a 4 way ELMA selector on the shelf.
To the left the shorting circuit (delay) - a must have.

This attenuator is a gift!

Bruce says:
The attenuator starts with a maximum attenuation of -68 dB and then has 4 steps to -54 dB. 
Each additional step is another -3 dB.  Other attenuators reach -54 dB in the first click. 
The Transcendent Attenuator therefore provides much better low level control.. 
This is essential for use with high efficiency speakers.

- couldn't agree more!
       

I miss the ARC SP16L remote badly, but this line stage is so good I can live with it.
Next step would be buying a silk supermalloy transformer attenuator. Maybe some day.

 

Grounded Grid Cap Test:

April 2008: Making a test on the sound quality of capacitors is a difficult task. In any scientific work we would as precisely as possible describe the conditions under which the experiment was performed because our work doesn't hold any credibility unless others are able to reproduce - and possibly confirm - our results.
Ranking the quality of capacitors is as far away from science as trying to describe the taste of a rye bread. It's highly subjective. Our aural memory is short, and the time it takes to change capacitors is long enough to forget how the previous capacitor was performing.
I tried a range of caps over a relatively short period of time, just to get a basic feeling of what I was in for and then stayed with one cap for a couple of weeks where I would listen frequently. Then, after some serious listening, I would plug in a new cap.

At a certain point I made a switchboard allowing A-B testing of up to six different caps and much to my surprise, this wasn't the way to go. Being able to zap between caps in a matter of seconds is highly confusing. After a short period of time you end up not being able to hear a damned thing!

The problem with testing things like this is that we know what is going to happen and we have all sorts of expectations about what to hear. We can't help preparing our mind for the event and we're in for all sorts of self-delusions.
The best example I have experienced is an (in-)famous NordOst sales person who demonstrated a 2000 $ power chord. Now, the power chord came in after a 10 meter, 5 $ extension chord, feeding an integrated amp with a manually operated (non-stepped) attenuator and we were all carefully prepared for what was going to happen. And we could all - I think - hear how good it was. I mean, 0.5 dB less attenuation than before and we must hear things we didn't hear from the standard power chord. I can't prove this is what was going on, but the demonstration was highly problematic. I think the psychological preparation was much more significant in this event. We can all sharpen our senses - when we want to - or, when we're told to do so. And it takes courage to raise your arm in a group of twenty unknown people and say: Bullocks! The Emperor's New Clothes.....

The complete chain of components going into hifi system is enormous. From the phono cartridge with its phono wires, the cables including cable plugs to the RIAA with its internal wires and coupling caps, the cables from the RIAA to the line stage with its coupling caps, wires again to the power amps - with its coupling caps - and the cables to the speaker crossover - with its coupling caps - before the signal finally reaches the speaker drivers. From phono cartridge to speakers I have 6-7 coupling caps, all adding to the sound of the total system (Garrard 401/Linn Ittok LV II/Dyna DX20, Transcendent RIAA, Transcendent line stage, Audio Mirror 6AS7 PSE 20 wpc monoblocks, JA8008/TW034 DTQWT). And this is only a fraction of all the things that - more or less - adds to sound colouration. All caps in the current chain of components are high-grade silver or silver/gold caps, cables are silver plated cobber in teflon wrapping and I claim I hear differences when I change this single cap in the Grounded Grid line stage. I'm also sure that if anyone made me do a blind test on 5 different caps I'd probably be in deep trouble. That's how difficult I think it is.

I'm not going through the caps seen below and give a detailed description on each single cap. The silver/1% gold is the best followed by the Jantzen Superior Z-cap and the Russian 0.22 uF alu/teflon caps. The latter two I find almost the same. The 1 uF Russian alu/teflon giant was a disappointment, lacking in resolution despite a decent tonal balance. The brown Russian "cookie" isn't bad at all - if you listen to classical acoustic guitar and nothing else - and maybe harpsichord. Smooth and sweat but not able to keep apart complex orchestral works. Kind of surprising... Resolution is one thing - but one cap can display good resolution as long as you don't feed it complex musical signals. I'm not an expert in the transmission of electrical signals, but I see (hear) that one important parameter is the ability to transfer complex signals, i.e. electron load, amplitude and different frequences at the same time. Some skilled in the art of designing caps may tell us why this is so.

I've been listening a lot to vocals and piano for these tests. To my ears the most difficult sounds. The Steinway piano on Keith Jarrett's The Köln Concert was one of my favourite tracks. I've had this record for more than two decades. It contains both quiet and loud complex passages and covers a wide frequency spectrum. Any system will be put to the test from this recording - not least the phono cartridge and loudspeakers.

I have ordered 1 uF Mundorf MCAP Supreme Silver/Gold caps for comparison and more to come when these have been tested.
After that the PSU caps will be replaced by Black Gate electrolytics and later the 3 x 22 uF/350 V de-coupling caps on the tube board will be replaced by some film MKPs, not sure which yet.
Also the internal signal wiring will be changed to silver plated cobber/teflon wrapped/shielded wire bypassing the tube board. No need to run the signal from the phono sockets to the board and then from the board to the input selector.

20-05-2008: To make a long story short: These Mundorfs will stay. They leave all caps tested here behind. The prototype silver/gold is good, really good - but the Mundorfs are better. The overall tonal balance from these two caps are much the same but the Mundorf excel in the level of transparency - and I'm a sucker for transparency. If a cap is slightly bright I can tune my speakers accordingly.

#1 cap test: Jantzen Audio Superior Z-caps

#2 cap test: Vintage Russian alu/PIO caps

#3 cap test: Prototype silver/1% gold
(didn't want to cut leads to proper length - yet ;-)

#4: Gigantic Russian alu/teflon foil caps. 1 uF/200V

#5: Russian alu/teflon foil caps. 0.22 uF/600V

#6: Mundorf Supreme, silver with 1% gold/PP film

More cap test, Feb-09

Left to right: Russian PIO, K40Y-9, 220 nF/1000V, Russian PIO (green), K42Y-2, 220 nF/500V, Russian Silver Mica: 178 nF/350 volts.

From SOVCOM I bought some more caps, just to know first hand what these Russian vintage caps can do. Initial plans were to test them in my 300B amp, but the GG preamp is a convenient amp for testing caps. The three caps seen above were all tested against the Mundorf Silver/Gold cabs and the PIO caps really aren't bad at all. Surprisingly good, actually. The Mundorfs may excell in overall resolution, but I might as well pick one of these two PIOs for smooth and balanced presentation. The Silver Micas are some chunky beasts and to cut a long story short, I might as well leave these caps in the GG for good. These cabs can be stripped from the metal case (magnetic) and put into some wooden containers filled with wax. I got two of these caps for 17 US $. Two 0.22 uF Mundorfs are ~82 US $.

The Final Mods

Black Gate caps in PSU section. 47 uF film de-coupling caps in amp section.
Silver coated copper wires in teflon sleeves. Mute switch.

 

I strongly believe it doesn't get much better than this. If you think you may have a better line stage, drop me a mail. The impact on sound from the latest mods is arguable, but possibly a slightly more easy-to-listen-to presentation - and it wasn't too expensive and for the benefit of the doubt.... As always the basic circuitry and coupling cap(s) is what has the greatest impact on sound.

Feb-09: I really shouldn't have made the statement above, because I've recently tried a TRAM w.o.t. line stage and driving the modded Audio Mirror mono-blocks, this amp beats the GG in the midrange. The TRAM from diy-hifi-supply/Hong Kong is no longer available, so I'll have to clone this line stage. I thought I was in for some years of peace on the line stage front, but apparently not....

 

The final-final mods

 

It's always a good thing to go back and see (hear) if the changes we make really are worth while. So, I re-installed all the original components - except the coupling caps - and to my ears, it didn't sound any different from all the Black Gate stuff. I've tried expensive caps for power supplies in other amps and always found the benefits to be small, if noticeable at all. The coupling caps are important, no doubt about it. This is where I would spend the money. Two good caps is all the GG needs - to my ears.
The expectation we build up when buying something, getting it home, installing it, setting it up again, etc., is a major psychological pressure. We so much want something to happen - and to hear our money well spent. And we're so good at fooling ourselves that we sometimes can hear whatever we want. Swapping back is an excellent exercise, although we rarely do - because it's a lot of work. So, above the GG today: The excellent attenuator from Transcendent and the two Mundorf silver/golds.