Auricap Film capacitors

Product: Auricap - Premium Audio Capacitor - 0.47uF +-10% 450V EGA 7956-35
Manufacturer: Audience - USA
Price: 7.79 USD/Euro each
Reviewed: Giorgio Pozzoli
Reviewed: September, 2001

Capacitors sound is an old problem in audio. Anyway, probably because of the delicate situation, I have not seen hardly any review or comparison test published on the topic. Normally what's published is just nothing more than a press note.

So when I was asked to review some samples of Auricap capacitors by Audience, I was really interested and thrilled. On one end the task appeared to be not so simple and straightforward and on the other end Auricaps are used by several well known OEMs in the audio business.

First I had to choose a set of comparable capacitors in my junk box. To make it not easy at all for our DUT, I selected a couple of other high quality polypropilene capacitors and also a NOS paper in oil capactor.

Then I had to set up a comparison tool. For the sake of simplicity I decided to mount a selector and the capacitors under test on a Experimentor insertion board, with input and output pins so that it was possible to insert the capacitor in my current rig in series to the signal between the CD player and the pre-amp.

Not a typical use? Yes, but as I just have to compare different capacitors, all of them would have been absolutely in the very same situation, and therefore this was as good a solution as any other. Also, it required a short time to be available and any change took seconds.

Description

The capacitor arrived in a small box, packed in a plastic bag containing mounting instructions.

In fact, as the most refined of us require special cares, even these capacitors have a peculiar attitude, as it seems. The capacitor are axial models with flexible, insulated, multistrand leads coming out of each side. One lead is black and the other one red. If the capacitor is used in the signal path the signal must enter through the black lead and get out through the red one.
If the capacitor is used as power supply bypass unit, the black lead should be connected to ground and the red one to the + or - voltage. Finally, if the capacitor is used in loudspeakers crossovers, the red lead should be toward the + terminal and the black one towards the - one.

Sceptical about it? I was too, so I tested the capacitors in both directions, and I must admit that they seemed to sound best in the suggested configuration.

A very high quality presentation indeed. Even the body of the capacitor, yellow, with green side surfaces, is very attractive. This has nothing to do with the sound, but the look is always important...

The sound

Which are the nasties that affect capacitors sound, at such an extent that any designer tries to get rid of them wherever possible?

In my opinion there are two or three basic sonic problems.

The first is colouration. The value of the capacitance of any capacitor is not constant with frequency; moreover there is a resistive component and, what's worse, an inductive component, because of leads and physical dimensions of the body, both variable with frequency. This means that in a real circuit a real capacitor will have a behaviour which only roughly approximates its model, and where you should expect a perfectly flat frequency response, there are very subtle ondulations or phase anomalies that can be perceived by the ear.
Colouration consists in the fact that the reproduction seems to favour some frequency, typically in the medium frequencies range, so that the musical message seems biased towards this frequencies, and often becomes dull and repetitive.

Another problem is transparency. A very transparent system tends to disappear: you hear the flow of music direct from the playing instrument. If a problem occours, it is like listening to the music through a curtain: the message misses detail, precision, depth.
This veil has - IMHO - also the effect of making the image flatter, dead, feeling like you're listening to a reproduction instead of live music. At the same time the high frequency energy seems to be reflected and dispersed in an innatural glare that saturates mid frequencies making them dazzling and confused.
From the technical point of view I am tented to track it to a loss in the high frequency range, but as a matter of fact if you listen to each component of the sound it's high frequencies appear not attenuated at all; instead looking at the entire picture you feel the lack of real life detail, of air between the instruments.

The last problem is dynamics. There might be a lot of different causes. For example, the mechanical structure of a capacitor is rather simple: often it is just a piece of metallized film rolled up. The film is subject to varying internal electromagnetic forces due to the charging and discharging of the capacitor. If the capacitor structure is not rigid enough, these varying forces can modify the geometry of the capacitor, and as a consequence its capacitance!
This obviously can cause an alteration of the signal, which is modulated by this variable capacitance. Typically the problem is caused by the geometrical structure of the capacitor and by its internal resistance and inductance, that cause problems to the fast rise and fall of the voltage in the capacitor. The tipical effect of lacking dynamics is a lifeless sound, with no impact, dull, apparently very "slow".

Note that the problems in practice arise when a single type is consistently used throughout a design, as the effects of each one in this case tend to add up, even though each capacitor contributes for a small amount. In any case the differences between two single capacitors are often subtle and rather difficult to detect, especially in case of high quality components.

In our case the sound of Auricap capacitors seems to me very similar to the paper in oil capacitor I used as reference. The balance is perfect, the sound is perfectly neutral, uncoloured. Not even an hint of colorations like the one that one can find in industry standard components; for example a Wima MKP10 - 250V, a very good component, seems a little coloured in comparison with Auricaps.

I only find an extremely thin hint of the veil that distinguish paper in oil from lower class capacitors. It is as though a very thin curtain has been introduced in the recording system. I still have to find a film capacitor that is able to fully compete with paper in oil for what regards transparency and the cost of a paper in oil is normally more than the double of the Auricaps, so this is really a pleasant surprise. Not for Audience, anyway, that presents on their web test results showing that Auricaps measures better than Jensen paper in oil capacitors of the normal (aluminum) series. Unfortunately I had no Jensen capacitor available to make a direct sound comparison.

Compared with the other very high quality film capacitor, the type I have selected to use in tube designs where I do not want to include paper in oil, the differences are - again - rather subtle, but the Auricap is a little more detailed in low level signals, more precise, and even seems to have a little more dynamics.

Conclusions?

Shortly after the test I modified my power amp substituting all film coupling capacitors with paper in oil ones. I obtained a beautiful, lustful sound, with a lot of detail and a huge amount of depth. Really attractive.
But... "strange, the voice has a little less body then before. The bass is round and full, but not as much as it was beforehand. The depth is exceptional, I would nearly say excessive, it is a little as though we were listening in the distance, and not in the real soundstage. Ah, but the sound is really lustful...".
So when I discussed the test results with John McDonald and Richard Smith and they gave me some more technical information, I was ready to accept their suggestion for one more comparison test, the "perfect" one.
Which is the obvious perfect alternative to a coupling capacitor? A straight wire, obviously. Unfortunately the substitution is not possible in most of the rigs (no one inserts an unnecessary coupling capacitor in a circuit, unless perhaps to gain absolute protection from DC in case of malfunctioning of the upstream stage), but is perfectly possible in the configuration I used for the test.
And the test was really revealing. The Auricaps are rather difficult to distinguish from a direct connection. The paper in oil caps instead can be easily distinguished: as a matter of fact they outline a much more detailed, precise, deep and realistic soundstage, but such a soundstage is simply not present in the original signal! It is the result of a very peculiar form of colouration, that is evident in comparing the caps with the direct connection, and that makes the medium and lower frequencies thinner and lighter than normal.
Now, is it correct to use components that trade in reproduced sound accuracy for enjoyability, or not? Well, this is a matter of personal philosophy: as such everyone has his own answer to the question, and every answer is questionable. So I'll keep off.
But if you want absolute transparency without the paper in oil coloration, then you must give Auricaps a try.

Conclusions

The very positive evaluation above has been confirmed by some more testing performed by including a few Auricaps in a new design.
Auricap are very neutral and smooth capacitors; they are also really fast, but most of all they are exceptionally uncoloured and transparent. They are presented in an attractive build, sport the best measures in class, are available in a rather large number of values and are rather cheap if compared with the other audio-grade capacitors.
What can I say? Recommended!