Test: Alumen Z-Cap, Jantzen Audio
Copyright 2016 © Troels Gravesen



The list of capacitor brands for crossover networks is considerable and each manufacturer will praise its brand and promise sonic qualities delivering true fidelity. A couple of years ago I tried digging into two basic families of capacitors, the standard PP and what I call super-caps. The file is to be found here. The article does not include metallised polyester and electrolytic capacitors although both types are highly suitable for some functions in a crossover network, but not exactly for those placed in series with the driver. Here we need better, and most speakers these days will feature standard metallised polypropylene and a few high-end speaker even super-caps. Super-caps are double-lane metallised polypropylene, i.e. two capacitors in series, hence the excessive volume - and price - as two capacitors in series will display 0.5 x capacitance, which means it will be roughly 4 times as big for the same value.
Double-lane caps were originally designed for high voltage application but has found a significant place in high-end audio equipment, maybe not in quantity, but certainly in quality. Most common types are aluminum and silver-gold metallised polypropylene, the latter seriously expensive.
There are other ways to good caps like the VCap, copper foil and teflon (PTFE) dielectricum. Price is horrific if we talk several microfarads. Teflon is not an easy foil as it takes two layers to minimise arching due to tiny holes in the foil. Copper foil is pricy too - at least when it has to be as thin as needed for capacitor manufacturing. Aluminum is more manageable and we're talking a price range between standard aluminum and silver-gold double-lane capacitors. In fact quite a a lot cheaper than silver-gold.

I'm currently working on a 3-way construction (10+6+1) speaker with a point of crossover between mid and tweeter at 1600 Hz. Quite a low value and highly demanding on tweeter performance. Next I have my reference ATS-4 speakers, having a point of crossover between mid and tweeter at around 2.9 kHz. The new speaker was set up with Superior-Z caps and my ATS-4 has Silver/Gold caps for tweeters.
From Jantzen Audio I requested Alumen Z-Caps for both speakers to evaluate the performance against the two types of capacitors I've been using for years and know very, very well.

The super-caps are characterised by having an extraordinary level of transparency compared to standard PP capacitors. The difference is significant. Super-caps pass complex source material with a high degree of detail resolution without smearing detail, and this goes for midrange as for treble. Within the super-cap range there are flavours to pic depending on metal used for metallisation, alu, silver, and silver-gold (usually 1% gold). Silver-Gold may deliver a little more smooth treble performance, but differences are small and to my mind not justified by the price difference. Polypropylene film can be impregnated with oil, hence "oil" versions of the super-caps. I have capacitors from same brand and value plus/minus oil and I don't find any difference in performance for loudspeakers. Bit of snake-oil in this context I think.

Every capacitor has a memory. Electrons get stuck in the plastic or paper that separates the windings and may distort the shape of the signal passing. Lots of research has gone into minimising this effect by using different materials and winding techniques and we certainly have better components these days compared to only a few years ago.
The thing is that even super-caps can add a certain colouration to the sound - although miniscule. Super-caps can almost make music sound just a little bit better than life by adding a certain sheen/glair/brightness/radiance to particularly high frequency content. English is not my native tongue and different words that may be used to describe the issue, so take your pick. I wouldn't use the word lush. Lush is more related to 2nd harmonic distortion to my thinking.

My prototype 3-way having a point of crossover between mid and tweeter at 1.6 kHz is particularly well suited for demonstrating the effect. I found a very forward sounding saxophone on the Pawnshop recording and swapping between Sup-Z, Silver-Gold and Alumen-Z was quite a surprise. Suddenly the sheen was gone and the overall sound image appeared darker. "Darker" is mostly a good sign when you know frequency response is exactly the same and yet, the sound becomes darker without any smear of detail. You start listening to treble rather than tweeter or tweeter + capacitor or whatever adds its sonic signature. This sheen also sometimes adds to the perceived sense of detail and transparency. The Alumen-Z appear to present a more credible overtone structure to instruments and voices and most importantly, not diminishing the level of transparency. 

As can be seen below, my next move was to replace the Silver/Golds  in my ATS-4 with the Alumen-Z. The result here was more subtle, probably due to a higher point of crossover, but still noticeable. Listening through a wide range of recordings it struck me that the perceived sense of depth from the midrange was noticeable improved. I interpret this phenomenon as a matter of change of attention. If any driver or component adds some signature to the sound image it may catch our attention and distract us from other aspects of the presentation. The midrange transparency is there as before, because nothing has changed, but all of a sudden we notice its quality because we're not distracted by some other (mis-) information.
The Bosendorfer piano used by Keith Jarrett's in the Köln Concert is a recording I've been listening to for decades and believe it or not, but I remember the sometimes hard sounding transients from my system 10-20 years ago. How it really did sound we'll never know but the recent Alumen experiments added another level of appreciation. For my ATS-4 speakers, the Alumen-Z has come to stay until something better comes up. Copper foil/PTFE for the ATS-4 would cost me some 5000 USD. I think I'll hold my horses.

Alumen-Z capacitor range is made for crossovers, having a maximum voltage rating of 100 volts. This means the polypropylene film separating the aluminum foil windings can be very thin = reduced memory. They will probably be suitable for solid state amplifier DC coupling, but rarely for tube circuits.

ATS-4 tweeter crossover detail.

ATS-4 mid-tweeter crossover section.