heres an odd one

[soundguruman]

"the human ear is not that perfect a measurement device."

Oh, I beg to differ, the human ear is the ONLY perfect measurement device.
All your meters and scopes, are just an attempt to imitate, a mathematical representation of what the real ear actually hears.
And there has never been a meter, which matches the perfection of the human ear, and never will be.

You would love to be able to measure what sounds good and what does not. But to a greater extent, you can't.

[daz]

Sorry Steve, but as i said previously i must have tossed those caps because i can't find them. I think i was worried i'd forget about how they sounded and use them again. Anyways, they were dull and muddy, not bright. So if anything it might be the ones i like that have the specs you suggest are "good". Doesn't matter to me, as the ears tell me whats good.

As to the notion someone hears what they want to, that certainly can happen, but 2 things to note....1-when i person does this long enough they begin to look for and get good at determining if it's placebo effect or not, and even question themselves and A/B the tweak numerous times till the difference becomes obvious. And 2-if someone IS aware about that and has become good at recognizing these things, the proof that they aren't fooling themselves is that they will try many things where they hear no difference whether or not they expected to. That would speak to the validity of the instances where they DO hear a difference. I'm not a tech like many of you, but i do have ears that have been educated by 40 years of playing and tweaking. And during the last 5-7 years or so of endlessly experimenting with amp design, i have noticed my ears develop more that in the previous 30 something ! Serously....i can't believe how sensitive i have become to it and i now understand all the tone issues i went thru during my gigging years and I've become super in tune with the tiniest tonal details and can tell what they will do in a mix, be it good or bad. It's nuts. I should have been a mad scientist. (well, i'm not a scientist anyways)

[soundguruman]

Arrgh - you just had to make me get out my $5000 meter! Would you believe it? It doesn't even go boing! But, it does give me this:
[ATTACH=CONFIG]21880[/ATTACH]

The two green lines show the change in frequency response using a +/-20% tolerance capacitor.
The two blue lines (yes, there are two) show a perfect 0.68uf and a 0.68uF cap with a really bad ESR of 15 ohms.
I think this demonstrates that the differences are
(1) Easily measured
(2) Primarily due to tolerance
(3) Not due to ESR

Edit: BTW, I measured the ESR of a 1uf in my parts bin as 2.9 ohms

There's an update, to make it go: "boing."

[Chuck H]

Anyone price OD caps lately? Sprague is really getting up their own a$$. Atoms and ODs were my go to products for a couple of legitimate reasons. But I just won't pay those prices anymore. Panasonic and Nichicon make equal (and sometimes superior) products at one third the cost. As to the subject at hand...

I did once try polypropylene and polyester caps respectively in two otherwise identical (relatively speaking) amps. Now, it was rightly pointed out that I wasn't stringent in testing actual capacitance for individual parts. But IMHE most modern parts are very, very close to spec. I did hear a difference that wasn't the sort of thing an EQ does. One was a bit bolder and more forward sounding. Not to be obtuse. So lets just say it may have had better top and bottom clarity. The other was more middy. But not just in EQ. There seemed to be more complexity in the mids. Again, not to be obtuse. Lets say there seemed to be more peaks and valleys in the mid voicing. This is with two amps that I could actually play side by side. Not a change to a single amp. These differences were subtle but significant. I attribute the differences to a couple of factors discussed above. Mojo NOT being one of them. I only relate this because it's still implied in this thread that film caps of different construction shouldn't sound significantly different. But they do. But I don't think changing a single cap with the same value but different construction would make a change you could notice.

[soundguruman]

"The hifi tweako school of audio criticism has long held that micro-differences in sound quality exist that can only be heard by special individuals with superhuman abilities of hearing discrimination. Even more, that if instruments can't find a difference heard by these special "Golden Ears", then it follows that the *instruments* are wrong."

Yes there is Chinese violin, made of compressed sawdust.
And there is Stradivarius Violin also.
Both have four strings and are tuned just the same.
They sound absolutely identical, right?

To some people, they 'might' sound exactly the same. But to a musician, with trained ears, they do not sound exactly the same.
Do you dispute this statement also?

[R.G.]

Oh, I beg to differ, the human ear is the ONLY perfect measurement device.

Talk to an audiologist or a hearing researcher about that. Then come back and let's discuss the issue. Ever hear the terms "audible illusion", "critical bandwidth", "Fletcher-Munsen", "audible masking", and many others?

All your meters and scopes, are just an attempt to imitate, a mathematical representation of what the real ear actually hears.

Actually, no. Meters, scopes, and other instruments are an attempt to get something down as a measurement. Remember Lord Kelvin?

"In physical science the first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it. I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of Science, whatever the matter may be." [PLA, vol. 1, "Electrical Units of Measurement", 1883-05-03]

And there has never been a meter, which matches the perfection of the human ear, and never will be.

Let's say that another way: there has never been a meter, which matches the IMperfectionS AND QUIRKS of the human ear well enough to read out what a human hears, as most of hearing happens in the brain, not in the ear. By definition, since the mind is inserting most of what we "hear", a lot of what we "hear" is an artifact of our minds. Note that this matches wild advertising claims exactly.

You would love to be able to measure what sounds good and what does not. But to a greater extent, you can't.

I agree with that completely. The problem is that "sounds good" is a phrase that has a different meaning to every person. You fundamentally cannot measure something that has a different meaning to every person. Case in point:

Yes there is Chinese violin, made of compressed sawdust.
And there is Stradivarius Violin also.
Both have four strings and are tuned just the same.
They sound absolutely identical, right?
To some people, they 'might' sound exactly the same. But to a musician, with trained ears, they do not sound exactly the same. Do you dispute this statement also?

Absolutely not. See my recent posting on capacitors. Details matter.
However, the issue you're obscuring with this particular red herring is not whether they sound the same, which they (measurably) would not, is which one sounds better (i.e., more "good").

The real issue is that although they sound different, different people like them differently. Someone who likes erhu (Chinese two-string violin; look it up) will probably think that an erhu of almost any quality sounds better than either one of these. And there are probably people who will think the Stradivarius is annoyingly bright.

Anecdote time. I have a friend with unusual hearing, from birth. He hears within normal bounds for human hearing up to about 3kHz, then hears substantially nothing at higher frequencies.

... until about 32-40kHz. He hears ultrasonic dog whistles. Ask him what HE thinks of the sound of a bow on the strings of a Stradivarius.

Mmmmm. Then there's strings. If I put cotton sewing thread on the Stradivarius, and gut strings on the sawdust violin, does that change how they sound? As background, the erhu historically used silk strings, but most are metal now. Does that change things?

The problem I'm trying to illustrate is that it doesn't make any sense to speak in absolutes about something that can't be measured. And you're arguing both sides of that - on the one hand that the human ear is absolutely perfect, and on the other that it's impossible to measure. Which side of the question do you really want to argue?

As an aside - do violins made from compressed sawdust exist, really? I did duck into several musical instrument stores in Hong Kong and China while I was there, and noted some very, very inexpensive violins. These were made from wood, not particle board. I suspect it was a good grade of plywood, but I don't remember any made of compressed sawdust.

[olddawg]

"The hifi tweako school of audio criticism has long held that micro-differences in sound quality exist that can only be heard by special individuals with superhuman abilities of hearing discrimination. Even more, that if instruments can't find a difference heard by these special "Golden Ears", then it follows that the *instruments* are wrong."

Yes there is Chinese violin, made of compressed sawdust.
And there is Stradivarius Violin also.
Both have four strings and are tuned just the same.
They sound absolutely identical, right?

To some people, they 'might' sound exactly the same. But to a musician, with trained ears, they do not sound exactly the same.
Do you dispute this statement also?

So... Let me get this straight. Capacitors are really violins and are made of either pressed wood or of Stradivarius composition? Makes sense to me...

[Steve Conner]

Oh, I beg to differ, the human ear is the ONLY perfect measurement device.
All your meters and scopes, are just an attempt to imitate, a mathematical representation of what the real ear actually hears.
And there has never been a meter, which matches the perfection of the human ear, and never will be.

You would love to be able to measure what sounds good and what does not. But to a greater extent, you can't.

This is just plain wrong on one level but right on a deeper level. The best golden ear in the universe can't match the performance of a good modern condenser mic, preamp and 24-bit converter. You can show in the lab that the hardware can discern tiny details of the sound that nobody could ever hear.

The design details of psychoacoustic codecs like MP3 and AAC actually mirror the limitations of the human ear, so we understand them quite well. You can test your own ears against the "spec" by ABX testing with a lossless audio file against lossy files of various bit rates. Mine give out somewhere between 192k and 256k MP3. If you already have a good computer audio system, this costs nothing, the software to do it is all free.

However, it is true that you can't "measure what sounds good". No amount of computer software can analyse a 24-bit wave file and tell you if it sounds good. The software can see all of these tiny details of the sound, but it can't decide which are relevant and which aren't. Only our brains (not our ears) can do that.

But my argument still holds, that nothing can be heard that couldn't in principle be measured. Even if you can't do the measurement in practice, it is still a powerful argument.

1-when i person does this long enough they begin to look for and get good at determining if it's placebo effect or not

That is one possibility, the other is that they get high on solder fumes and start hearing things that aren't there. That's when measurement (or even thinking about measurement) can be a good sanity check. I'm not specifically accusing Daz of getting high on solder fumes, in this case I believe that his caps do sound different.

[daz]

That is one possibility, the other is that they get high on solder fumes and start hearing things that aren't there. That's when measurement (or even thinking about measurement) can be a good sanity check. I'm not specifically accusing Daz of getting high on solder fumes, in this case I believe that his caps do sound different.

Well, look at #2 in my post. If you are even somewhat skeptical about my ability to hear things like this, how do you explain the things i don't hear? Why no placebo effect there? theres really only one answer to that ! Everything is relative. If someone tells you you are using too much treble at one of your gigs and accuses you of having hearing damage, that makes no sense to me. Why? Because i am listening to all the other instruments too and I'm going to adjust my tone accordingly. If i crank the treble i would notice MY tone is much brighter than thiers. Again, everything is relative. I'm not tech, and i have no doubt i'm looked at here as an idiot because i know so little about electronics on the technical side. But ears have nothing to do with that. For all you know i may have ears that would make the proverbial EJ ears look tone deaf. No offense at all, just sayin...

[bob p]

For those of us that deal with aging gear, the ability to check ESR and frequency response on old electrolytic caps should not be overlooked. Although just about any fully-outfitted testbench would have an LCR meter sitting on it, they are expensive, so most hobbyists don't consider this kind of purchase.

If anyone is interested in an LCR meter that offers a lot of bang for the buck, I''ve recently picked up a fairly inexpensive hand-held meter, a BK Precision model 879B. I really like it. I wish I had bought it a lot sooner. I probably would have continued to put off the purchase, but in this instance my hand was forced by needing an accurate L meter.

Now that I've got a good C meter, I find that I'm getting a LOT of use out of it, finally getting around to checking caps in all of my gear. Since I've started using the meter to check ESR and frequency respone on all of my electrolytic caps, I've found caps that were limiting the frequency response of a hifi preamp. The result is that now that I've got the right gear, I can use an intrument to identify the caps that are causing problems and selectively replace them, rather than resorting to the shotgun method of re-capping an entire circuit. This saves a lot of time.

IME a good LCR meter was worth every that I paid for it. The problem is that most of us put the purchase off for way too long because of the cost.



Model 879B, 40,000 Count Dual Display Handheld LCR Meters - B&K Precision

[R.G.]

That's a neat toy...

... argh... wait... NOOOOO!!! GALS* AGAIN~!!!

(* Gear Acquisition Lust Syndrome)

[trem]

Speaking of tools you guys have.....
How "important" is to have a Cap/ESR meter that will measure in-circuit with high voltage applied.?

[R.G.]

I'd have to let someone else reply.

I would never do that, as I'd always be suspicious that whatever else was connected to the part in circuit would pollute the measurement.

But there is some reasonable argument especially for electrolytics that the applied voltage may change things a bit, just like the temperature of the electrolyte changes the ESR by making it work better - right up to the temperature that starts degrading the electrolyte and decreasing its life span.

[bob p]

I guess that's another way to say, "not important."

[olddawg]

Speaking of tools you guys have.....
How "important" is to have a Cap/ESR meter that will measure in-circuit with high voltage applied.?

I worked in a shop that had a big ass expensive Sencore cap checker that would test caps at working voltage. You had to test the caps out of circuit of course. Techs were always blowing up caps with it. A couple even hurt themselves. I used the Enzo rule even then. If you are taking a cap out. Most of the time you just replace it. Your time is more valuable than the component. I can see the argument for big filter caps. Bit with those you can usually see the ripple or other artifacts in circuit with a scope. I used to use an ESR meter. I got tired of callbacks. If one cap of a particular age and type is bad. Chances are the rest will be going south shortly.