Whether the make and materials of capacitors affect sound quality has a long and controversial history in audio. It has been debated often and unfortunately these debates have been contaminated by the two camps of the "Great Debate" in audio. Are you an objectivist or a subjectivist? Can people hear audible benefits from using different materials in wires and passive electronic parts even though performance differences cannot be detected by measuring instruments?
I don't intend to revisit this debate here – I'd prefer to stick with the questions the OP had about Pendragon speakers. If this thread becomes derailed by this debate, I won't participate any further. I think it's a waste of time, and worse, it's a disservice to new readers who want to learn something useful about audio.
But I will state my position about capacitors – one time. I am not an electrical or audio engineer. My education on these subjects was limited to one year of college physics which included basic DC and AC electronics. The rest of what I know about audio and speaker design is self-taught, learned mainly in the last 10 years as I became interested in DIY speaker building. Since then, my main interest in audio is just what makes a speaker sound good? That would include anything about a speaker, including capacitors.
I read the two "Great Capacitor Shoot Out' links and have some general comments. Most important is that this "shoot out' made the unfounded assumption that capacitors do sound different and they attempted to rate and rank the differences they heard. It has never been scientifically established that capacitors can actually do this. How can you rate the sound quality coming from using different capacitors unless it is first clearly demonstrated that people really can hear differences? So they fail on that very basic level.
A number of years ago, I attended a meeting of local DIY speaker builders in the Washington, Maryland, and Virginia area. A fair number of people traveled from outside the local area. About 40 or more people attended. Besides the usual show and tell of various DIY speakers people had designed and built, the major agenda that afternoon was to perform a blind listening test of whether a large group of people could hear differences in capacitors used in speaker crossovers.
I took part in that. It made use of two different small bookshelf speakers of excellent quality. These speakers had crossovers that contained two different capacitors as the primary high-pass filter caps for the tweeters. Each cap could be switched in or out by a remote control switch operated by the listener. The listener did not know which caps were in use, only that when he hit the remote, a red or green light (in a small box sitting on top of the speaker) would indicate that a switch had been made. Individuals and small groups could listen to various short musical passages and switch back and forth at will.
First, listeners were given a short version of the test to see if they thought they could hear any differences at all. Why go on if you don't hear any differences? If a listener thought he could hear differences, a longer test followed allowing enough repeats for the test to begin to acquire some statistical significance. As always, each listener answered this question after each pair of listening tests: Can you hear a difference between the green and red? I did the longer tests on two different speakers. Most others who were there did the same.
Overall about 40 listeners each responded to at least 12 (sometimes 15 or 20) repeats of the blind test. Among these roughly 500 or more test answers, the results showed that the group could hear differences in capacitors only 50% of the time, and 50% of the time they could not. They would have done no better or worse if they had been randomly guessing.
The capacitors being tested included cheap non-polar electrolyitic (NPE) caps, moderately priced metalized polypropylene (MPP) caps, and very expensive caps of exotic construction. I no longer remember exactly which caps they were.
These results had a profound effect, as you might guess. Argument and debates began right away. Many people there were not prepared to accept the unavoidable conclusion.
My own position swung widely during the afternoon. I was new at the time to DIY speaker building and I knew I had a lot to learn. That was the major reason why I was there. I was aware of the "Great Debate" but did not have enough experience to have my own position. During the listening test, I thought I could really hear differences between capacitors. I was thrilled because it meant that I had "good ears" even among these highly experienced speaker builders! I knew some others there that day were skeptical non-believers in enhanced sound from expensive caps, and I thought "here I am, a rookie, and I can hear these subtle differences".
When I heard the results, I knew immediately I was wrong. No one there that day could consistently tell the differences in sound due to the caps – myself included.
Similar versions of this listening test were repeated two other times later that year at other regional DIY speaker builder meets. Both got similar results.
One of the more respected DIY speaker builders there, who had been one of the primary organizers, discussed what the results meant to him as a DIY speaker builder. I remember his points very well:
- High priced, exotic, or “boutique” capacitors were clearly not worth buying.
- The cheapest NPE caps could work well with a large caveat. Many of them varied in their measured capacitance value by more than 10% from their printed value. In the experience of the organizers of the test, they found that you might have to buy as many as 10 NPE caps just to find a pair that were within 10% of the desired value.
- Inexpensive MPP caps (such as Dayton or Bennic, both of which are made in Taiwan) perform very well, nearly always are within spec as printed on the cap, and they never lose capacitance value over time. In fact, when measuring the caps that went into the test, these Taiwanese caps were often within 1% of the printed value despite their printed claim of ±5%.
So that’s where I come from. I strongly believe, based on the above facts, that high priced capacitors have no value in speaker crossovers.
You may choose to disagree – feel free. Everyone is entitled to his own opinions, but not his own facts. But if you wonder if you should spend an additional $750 for speaker crossovers that probably offer only exotic caps (yes, Mundorf silver/oil caps qualify as exotic), I would not recommend it. But it’s your money and your choice.
If anyone is inclined to question me further, please keep it out of this thread. Keep it civil – I’m basically a nice guy. As a final comment, I had mentioned earlier that I am not an electrical or audio engineer, but I am a scientist. I now earn a living designing and funding clinical trials of experimental anticancer drugs. I know quite a lot about blind trial design, hidden sources of testing bias, the placebo effect, and statistical significance and statistical bias. If you really want to argue with me on these subjects, I warn you in advance, if provoked enough I can rip anything you say into shreds.
