Burning in hi-pass capacitors

[natediggidy]

Re: Burning in hi-pass capacitors

people sometimes forget about coil resistance and it's intended and calculated use in the passive crossover.

True, but if you're sharp, you adjust the inductor value to compensate for a lower resistance ;)

[natediggidy]

Re: Burning in hi-pass capacitors

If you really care about attempting to hear the differences in types, then you really need to verify that the caps under test are as close in value to each other as possible.

Agreed.

[ougryphon]

Re: Burning in hi-pass capacitors

I would also be really interested to see how the capacitance and DF of the caps change after burn in. That would probably give you a good clue as to what is happening and why it makes the caps sound better. A little harder to quantify would be the linearity of the cap, but you need $10,000 in equipment to really do a proper DC bias and frequency sweep.

[Herman Trivilino]

Re: Burning in hi-pass capacitors

For the purposes of my test, why do I need to measure my caps before I build the crossovers?

Let's say you have a tolerance on the caps of 1%. That means that a 8.2 µF cap could have a capacitance between 8.1 µF and 8.3 µF. So, you might be comparing, for example, a Sonicap with a capacitance of 8.1 µF to a to a ClarityCap with a capacitance of 8.3 µF. Any difference you might hear could be due to this small difference in capacitance, a difference in the drivers themselves, or a difference in the brands. You won't know unless you can accurately measure the capacitance.

That's why you need the meter.

What's more, though, maybe that original crossover had a cap with a tolerance of 10% (typical for a cheap cap). It may have had a capacitance of only 7 µF, even though it's stamped 8.2 µF. You took it out, replaced it with an expensive cap having a capacitance of 8.2 µF. It sounded different to you. Better, you thought to yourself. If there's a difference in the way it sounds due to an expensive upgrade, it's got to be an improvement.

Let's say you're right and in double-blind tests you can prove that it sounds better to you. That improvement, though, is not due to the design of the capacitor. You could have acheived the same effect by adding a 1 µF capacitor in parallel with the original 7 µF, giving you a total of 8 µF. Same improvement in sound, but it costs a few cents (or even dollars) compared to the $75 that the fancy cap costs.

[[email protected]]

Re: Burning in hi-pass capacitors

yeah you have to measure to be sure. i bought a meter from mcm for about $50.
i look forward to your results.

[Taterworks]

Re: Burning in hi-pass capacitors

DBT on caps were done a number of times by a lot of different people. If done correctly, none can spot the difference between the capacitors, burned in or fresh. You are wasting your time but hey do it for the laugh.

Too many lurking variables. Things like the acoustic environment and the skill level of the listeners (and Harman, with their focus on acoustic science, knows that listeners can be trained to listen for specific things) are unknowns in assembling any panel of listeners, even if the listeners seem (or claim) to be of audiophile 'pedigree'. I participated in ABX testing of multiple capacitors in multiple locations in a loudspeaker crossover and with my younger ears I was able each time to recognize the capacitor that was "x", though the differences were so subtle that I thought it was almost more of a gut feeling, and I would have fallen into the 'hogwash' camp if I hadn't correctly picked every single X capacitor in every single test. If someone isn't aware of what differences to listen for, or is mentally prejudiced (which would cause them to doubt their ears) they could very easily miss the difference.

If someone is not a 'cap-hearer', then they are free of a significant financial burden in completing their loudspeaker projects, because cheap capacitors will get them as far as they need to go. If someone believes they are a 'cap-hearer', the only upshot for them is they get to claim bragging rights (which I think are dumb; I only regard it as a dim novelty that I think I can correctly hear differences in capacitors), but it's also a burden because parts cost in the crossover goes way up in order to maximize performance within the limits of that designer's audibility.

It should be noted that the capacitor is a microphone. It is comprised of plates formed by metal foils or metallized layers on a dielectric substrate that is essentially elastic. If the capacitor is very physically large (or especially if it contains any amount of air between its wrapped layers), very large changes in pressure in the surrounding environment (or lesser external pressure variations if there's entrained air in the capacitor) can change the distance between the capacitor plates such that the capacitance changes somewhat. Some (reference: Tony Gee) have noted that particular types of capacitors can undergo a capacitance change (and I will test to see whether this is measurable with my cheap Tenma meter) simply by squeezing them between the fingers. As the plates of a charged capacitor are moved closer together, the trapped charge decreases but the voltage at the terminals increases, and as the plates of the capacitor are moved farther apart, the trapped charge (capacitance) increases but the voltage at the terminals decreases, because the charge is stored as static electricity in the dielectric (or in the entrained air, which acts as a dielectric; an air-dielectric capacitor can be built but is very physically large). If the capacitance is changing with time, this can introduce time-dependent linear distortion in the crossover's transfer function. Condenser microphones work on this principle. So to say that a capacitor has no moving parts is essentially untrue, and that renders it capable of introducing distortion. What remains to be shown (my own experience and the fallible subjective experience of other 'cap-hearers' notwithstanding) is whether this distortion is great enough to be audible.

The role of my experience (and that of each individual who claims to be able to hear a difference) is to shape the direction taken in selecting components used in the crossover for a particular speaker design which is to be used for my own subjective enjoyment, since the end goal of audio reproduction is individual (subjective) enjoyment. I can hardly fault someone who chooses not to spend money on high-end capacitors (in fact, I never have in the past; I've always used Dayton MP caps) but neither can I fault someone for whom a large investment in crossover caps, inductors, and resistors holds value enough to provide return on that investment.

Note: One can easily observe that all different types of capacitors are out there. Some use a paper dielectric, which has a very large amount of entrained air just by nature of the paper material, and can undergo a very large capacitance shift when squeezed because the paper is compressible. Other caps like MP and foil/poly have entrained air between the layers unless some sort of resin is used in the winding, and the layers are subsequently pressed until the excess resin squeezes out. Some capacitors are oil-filled, to try and fill in the air spaces and damp the microphonic behavior. It's always interesting to see the lengths to which designers will go to produce a specific type of capacitor, or use a specific type of capacitor in a specific part of their crossover, and comment on the results. I don't have that kind of disposable income, so I don't think it's really all that productive for me to do extensive 'comping' of my crossovers. Maybe someday when I'm rich (hah).

[Tony G]

Re: Burning in hi-pass capacitors

It should be noted that the capacitor is a microphone. It is comprised of plates formed by metal foils or metallized layers on a dielectric substrate that is essentially elastic. If the capacitor is very physically large (or especially if it contains any amount of air between its wrapped layers), very large changes in pressure in the surrounding environment (or lesser external pressure variations if there's entrained air in the capacitor) can change the distance between the capacitor plates such that the capacitance changes somewhat. Some (reference: Tony Gee) have noted that particular types of capacitors can undergo a capacitance change (and I will test to see whether this is measurable with my cheap Tenma meter) simply by squeezing them between the fingers. As the plates of a charged capacitor are moved closer together, the trapped charge decreases but the voltage at the terminals increases, and as the plates of the capacitor are moved farther apart, the trapped charge (capacitance) increases but the voltage at the terminals decreases, because the charge is stored as static electricity in the dielectric (or in the entrained air, which acts as a dielectric; an air-dielectric capacitor can be built but is very physically large). If the capacitance is changing with time, this can introduce time-dependent linear distortion in the crossover's transfer function. Condenser microphones work on this principle. So to say that a capacitor has no moving parts is essentially untrue, and that renders it capable of introducing distortion. What remains to be shown (my own experience and the fallible subjective experience of other 'cap-hearers' notwithstanding) is whether this distortion is great enough to be audible.

You know...a very interesting and quantifiable test (might have been done?) would be to wire a typical 6 to 10" woofer into a typical speaker box. Then have a capacitor inside the box wired to an external capacitance logger. Play sweeps, sine waves, white noise etc at different volumes and see if the acoustic compression causes the capacitor value to shift a meaningful amount Could do this with different capacitors of different brand but match values, and same brand of different values. This is doable, with no bias, and can be done with very high accuracy and precision if someone has the right tools. If no one does it in time, I might have to one day when I bring some labview hardware home lol.

[Servicetech]

Re: Burning in hi-pass capacitors

Zaph says right on his website there is no difference. He seems to know what he's doing and his opinion is good enough for me.

"In my opinion, electrolytics sound every bit as good as poly, mylar or metal film capacitors. There is a lot of hype over more expensive capacitors. You either have to believe me or be willing to spend nearly $600 for a pair of crossovers rather than $160."

http://www.zaphaudio.com/ZDT3.5.html about 1/4th the way down the page.

[Wolf]

Re: Burning in hi-pass capacitors

I disagree with the first sentence. 'Lytics do not sound as good.
There is a lot of hype, but that does not mean that they don't actually sound better on occasion. Oh- and very rarely would I spend $300 on a set of xovers anyway.
It's funny how people are willing to accept one man's interpretation on everything regarding distortion and other things.
Later,
Wolf

[PWR RYD]

Re: Burning in hi-pass capacitors

I generally try not to use electrolytics because their values change so much over time. Hey! Shouldn't THAT be called a capacitor that is "burned in"? ;)

[ougryphon]

Re: Burning in hi-pass capacitors

@Tater,
An interesting corollary is that capacitors are also speakers. The electric field between the plates is strong enough to cause the entire cap to contract when voltage is applied. Ceramic capacitors are particularly prone to this because they also exhibit a strong piezoelectric effect.
The shrinking and expanding of the capacitors also adds harmonic distortion because when the plates get closer together, the effective capacitance also increases.

[carlspeak]

Re: Burning in hi-pass capacitors

Zaph says right on his website there is no difference. He seems to know what he's doing and his opinion is good enough for me.

http://www.zaphaudio.com/ZDT3.5.html about 1/4th the way down the page.

Tony Gee has a different opinion

http://www.humblehomemadehifi.com/Cap.html

[OlderMongrel]

Re: Burning in hi-pass capacitors

The fact of the matter is that everything can, and has, been measured, except for the depth and breadth of human gullibility.

Audio measurements are like medical science. Highly effective in some respects, primitive in others. We don't have to worry about polio or malaria, but they can't cure the cold, or allergies.

"Imaging" is a real property of loudspeakers (several properties, actually). Imaging can't be measured. If you move your speakers and listening position around in the room, or toe the speakers in or out, or add or change room treatments, etc, the imaging will often change. Sometimes the changes are subtle or hard to characterize. Sometimes they seem quite obvious. But either way, you can't put numbers on the changes. I'm not sure if you could even put a mic (or two mics) in the listening position of a room, measure, and tell, from the measurement alone, if the speakers are in or out of phase. Even if you could, it would be a very crude start at measuring imaging. We can't measure and say, "On the test recording, speaker A presents instrument X at 40-42% left, whereas speaker B presents it at 38-47% left on notes up to A above middle C, increasing gradually to 32-49% for higher notes". This is not meant as an example of a valid imaging measurement. It's just to make the point that we don't have any imaging measurements at all.

This lack of quantifiability opens the door to things that go a bit beyond, or way beyond, that which can be known as "factual". If you can't put numbers on the effect of moving the speakers, and you can't put numbers on taping a baggie of shiny rocks to your cables, are those two methods equivalent in validity? To me, no. But until the relevant measurements are developed (I think they will be, but not in my lifetime), we will have audio snake oil.

Maybe it's just part of the fun of DIY, trying to figure out if you are making progress or just being neurotic or gullible.

John

[johnnyrichards]

Re: Burning in hi-pass capacitors

I remember reading soimething Dunlavey wrote where he stated that imaging is a result of two speakers being as identical as possible. Furthermore, he went on to say that designers spend too much time on FR and not enough matching stuff...

I tend to fall into that camp, and as such - imaging can most certainly be measured in so far as two speakers can be measured and compared. I doubt anyone would seriously argue that driver cost, capacitor cost, placement, amplifiers, preamps, cables, room treatments, and music selection will result in a killer stereo image from two different speakers... The more I think about it, the more I think Dunlavey is right and most of us are wrong.

I have no reason to doubt Mr. Dunlavey - every thing I have read about the man leads me to believe he knew more about designing speakers than I could ever hope.

[Andrew P]

Re: Burning in hi-pass capacitors

Tony Gee has a different opinion

http://www.humblehomemadehifi.com/Cap.html

Thanks for the laugh. I really enjoyed that.