Joachim, remember the KEF 107 'reference speaker' with the swivel top?
My buddy thinks they are too hot at the top (I agree). I just opened one up, a xover network with enough parts to build a small cyclotron.
I think I'm going to try your 1 ohms trick
Jan
The series resistor quality also makes a difference.
Well, it is a process I have found to work. Of course there are ways to measure, but I am not about to elaborate on it. The general concept is when you have an ideal component selected and wish ti find a more economic or smaller solution, you measure reactance of the component and use the data as reference, then choose from other components and try different bypass till you get more ideal reactance. Tony mentions a bypass in his site which is a good start. This concept works for caps in speakers but not for line level signal devices because the current levels are different.
While you may feel very upright in saying the above, it didn't really answer the question. Delving further into non issues like money, doesn't help either. You have no idea what financial situation any of the respondents may have, so it should not be part of the topic and is completely unrelated.
So I ask. If you have a test scenario where nothing else changes but the brand of capacitor, would you hear a change?
If so, what is the reason for this? Poor or improper cap construction techniques resulting in parameter changes that have an impact on the response of the network in question?
Hi Joachim, I will be in Munich some time in July. If you will be available, I would like to have the opportunity to meet with you.
A bit like drinking toilet water, while entering heated discussions on the pros and cons of chateau d'Yquem, imo.
Who says I always drink Plonk?
Couple of weeks back, I was listening to £200,000 worth of HiFi and acoustic room treatment...You know, 4X 15" subwoofers, CD floating on an air bed, speaker cables about an inch thick...
It was better, but I still enjoy my own system.
Pallas, it is what it is. That is what I was trying to point out. Gee's ratings jumped out at me the first time I read down his list those 4 years back. Maybe today the correlation no longer holds. I havent re-done it. Now, I'd be more interested in ploting his ratings (albeit quite questionable as they are) vs cost.
So, what is your opinion on the AES paper done by the folks at Clarity Cap and a UK University research chap regarding cap construction and controlled listening trials? More pseudo-science perhaps?
For those who keep pointing out in this and similar threads the much greater effect driver performance and room interactions have on sound, you miss what I believe is the basic underlying assumption or premise, and that simply is: with ALL THINGS EQUAL and NO OTHER CHANGES in the SPEAKER, then what impact do caps of varying VDC and cost have on the sound?
Last edited:
I'm interested in how the esr conclusion was made , was it by measurement , controlled listening or was it conjecture , how was it determined an el cheapo electrolytic sounded better than a PP...
If it wasnt a controlled test as outlined in Linear audio , then all kind of brain farts could have happened....
Just saying
I have done a lot of recapping and selling of recap kits to customers over the years and the effect of ESR on the brightness of the speaker's sound is well known and obvious to those who have done it. The science is simple. If you remove 0.5 to 1.0 or more ohms of resistence in the tweeter series circuit, the tweeter's spl will accordingly increase with a perceived loudness increase by the listener. Sometime it is acceptable and sometime not. YMMV. But the effect remains.
Go back to post #61. The link to my paper is there. Within that paper is a link to the Clarity Cap research which eventually resulted in their "MR" rediculously expensive caps series.
I'd be surprised if CC's improved winding technology was any better than what Mundorf had been using all along.
This is true. Additionally, since the load on the cap is usually reactive, there will also be changes in frequency response of the network transfer function. How significant this will be depends on the network, the caps, the driver, and the position of the cap in the network.
Amazingly, it's all easily predictable and measurable. Straightforward engineering.
Joachim, remember the KEF 107 'reference speaker' with the swivel top?
My buddy thinks they are too hot at the top (I agree). I just opened one up, a xover network with enough parts to build a small cyclotron.
I think I'm going to try your 1 ohms trick
Jan
KEF's high end loudspeakers are well known for having ridiculous parts counts. They also contain no boutique parts from what I remember. Funny how their loudspeakers always tend to measure and review superbly.
The Clarity Cap research is here: ClarityCap Capacitor Information Page
They seem to detect some singing around 14kHz with lesser caps. I'm not sure the methodology is right though, because the stress on a capacitor depends on the load factor, and at 14kHz, there is very little because it's way above crossover.
Here's what I concluded on the specs, DF is damping factor which is another way of talking about ESR or equivalent series resistance:
http://www.diyaudio.com/forums/multi-way/215542-clarity-cap-mr-vs-mundorf-silver-oil.html#post3080839
They seem to detect some singing around 14kHz with lesser caps. I'm not sure the methodology is right though, because the stress on a capacitor depends on the load factor, and at 14kHz, there is very little because it's way above crossover.
Here's what I concluded on the specs, DF is damping factor which is another way of talking about ESR or equivalent series resistance:
http://www.diyaudio.com/forums/multi-way/215542-clarity-cap-mr-vs-mundorf-silver-oil.html#post3080839
Last edited:
The impedance change is academic and that was never my question , you made a statement , that after changing out from an electrolytic to PP you ruined the sound , my question to you was
How did you determine you ruined the sound , was it
1. Via measurements
2. Via controlled listening as per linear audio article
3. Was this pure conjecture ..
You more than anyone else knows listening test done without proper controls is playing with tricks our evil brain plays on , surely you did not listen without and worst putting out such subjectivism without proper scientific methodolgy ....
a.wayne I think SYs comment was suppose to be a simple reminder that a lot of the time a good designer will design the higher series resistance of a lytic into the crossover and that changing it to a PP will alter things for the worse.
This endless nitpicking gets boring very fast.
Exactly, so take SYs comment in the manner in which it was intended (to warn that simply swapping flims for lytics isn't always a good idea) and trust in the fact that he is no idiot. Regardless of that, even ignoring the subjective side of things, which is very easy to do because it is often meaningless, the technical merits of what he said hold.
This endless nitpicking gets boring very fast.
Exactly, so take SYs comment in the manner in which it was intended (to warn that simply swapping flims for lytics isn't always a good idea) and trust in the fact that he is no idiot. Regardless of that, even ignoring the subjective side of things, which is very easy to do because it is often meaningless, the technical merits of what he said hold.
My own testing pointed to frequency below 100Hz, but you could easily get different result based on how you do the tests. Having a setup as close to real conditions usually help.
I wouldn't argue with a low frequency effect, soongsc, thin metallic films are going to struggle with big currents.
I could add to 5th element's comments about capacitor ESR having a big effect. Half an ohm of resistance in a capacitor shunt not only affects amplitude, it can change phase between 10 and 20 degrees.
This affects lobing and power audibly. You can hear it. The speaker loses its sweet spot horizontally. But that is not an argument for boutique capacitors, just a design consideration.
I could add to 5th element's comments about capacitor ESR having a big effect. Half an ohm of resistance in a capacitor shunt not only affects amplitude, it can change phase between 10 and 20 degrees.
This affects lobing and power audibly. You can hear it. The speaker loses its sweet spot horizontally. But that is not an argument for boutique capacitors, just a design consideration.
Last edited:
If parts would be of sufficient quality meaning modern film types
of the same capacitance and typical ESR, in scientifically controlled
evaluation conditions , there would be no perceivable difference
between different brands, IMO.
ESR (equivalent series resistance) and capacitance can easily be
simulated if you wish to see what it means.
- Status
- Not open for further replies.