[Niteshade]

Warm up is the process where the device as a whole reaches its average operating temperature. Typical warm up time is around a hour. It takes time for heat to spread from components which generate it to ones that do not.

When a components heats up, there are changes in how it conducts electricity. These changes disappear once the device cools back down. Nearly all electronic components change properties after they heat up. There can be audible changes after an amp/preamp/etc.. heats up.

[jtwrace]

Warm up is the process where the device as a whole reaches its average operating temperature. Typical warm up time is around a hour. It takes time for heat to spread from components which generate it to ones that do not.

When a components heats up, there are changes in how it conducts electricity. These changes disappear once the device cools back down. Nearly all electronic components change properties after they heat up. There can be audible changes after an amp/preamp/etc.. heats up.

In simple terms it's called Heat Soak.

[Letitroll98]

Warm up is the process where the device as a whole reaches its average operating temperature. Typical warm up time is around a hour. It takes time for heat to spread from components which generate it to ones that do not.

When a components heats up, there are changes in how it conducts electricity. These changes disappear once the device cools back down. Nearly all electronic components change properties after they heat up. There can be audible changes after an amp/preamp/etc.. heats up.

So I take it you are saying there is no "Scientific Basis of Break In" and it's all warm up in the first hour and misguided perceptions after that?

[Niteshade]

I never entered into the break-in subject. My response was specific to the warm up period of electronics.

I think what you want to know is whether temperature cycling can change a component's properties over a period of time. Yes it does and it typically leads to replacing the effected component due to material fatigue in extreme cases.

The best way to proceed into a good discussion regarding how something can change over time or temporarily (aka: warm up time) is to ask questions about specific components. That is because the answers will be different for  resistors, capacitors, etc..

[Ethan Winer]

Ethan, you are just throwing numbers around with the change in distortion vs.the operating temperature of a SS amp

If you believe those numbers are not realistic, let's see your actual numbers as measured on a competent SS amplifier.

the figures you cited represent a 100% change from .002 to .001.

Yes, but who cares if neither level of distortion is audible? I'm sure I already made that point.

Lets suppose the SS amp starts at 0.1% THD and falls to 0.05% I would willing to bet this would be audible.

Maybe, maybe not. 0.1 percent puts the distortion 60 dB below the music. That's awfully soft. Plus, aren't you now "throwing numbers around" too? Are you guessing, or have you actually measured those numbers? You say "willing to bet" but are you really? If you're willing to visit me in person we can easily do a test like that using any source you'd like, to see at what levels you can hear distortion come and go.

As an example of a competent but not state of the art power amp, let's use the Crown PowerBase 2 amp I bought back in the early 1990s. This is a serious professional amplifier made by a serious company that knows what they're doing. This amp puts out more than 350 watts per channel into 4 Ohms with less than 0.05 percent THD and IMD. The specs provided are pretty detailed, as is common for pro quality gear, and there is not one word about anything changing over time as the device warms up. There is a note with the specs that I found interesting:

At Crown our published specifications are guaranteed for three years. Further, because our "in-house" specs are more stringent than our published specs, every Crown amplifier will exceed its published specs.

If this amp changes quality over time in an audible way, I'd be very surprised. I've certainly never noticed that.

Plus in addition to this simplistic view of distortion the amplifiers IM distortion behavior when it is below its normal operating temperature has not been considered, nor has its behavior with regards to transient signals been examined.

Not been considered by whom? The average audiophile? I'm sure all of this and much more is considered by competent amp designers!

SS amplifiers are frequently not entirely stable before reaching their normal operating temperature.

I'm too lazy to type more quotes from the Crown manual's specs, but there's an entire long paragraph about output stability, lack of turn-on transients, protection against thermal runaway that Steve mentioned, and so forth. If you have any hard evidence to support your claim that competent SS amplifiers are "not stable" for [who knows how long] I'd love to see it. Not theorizing, or speculation on what types of distortion have or have not been measured, but actual hard data sowing an audible change over time longer than maybe 3 seconds.

Look Scotty, even if it can be shown that competent SS amplifiers change audibly over the first 1-30 minutes of operation - and that's never been shown to my satisfaction - the fact remains that the main reason some people perceive a change in sound is due to their own changing perception. That's the core issue here, and it's the reason discussions like this appear again and again and again. We can hypothesize all day long about what changes occur inside an amplifier as its temperature rises. But if those changes are never actually audible, it's just an exercise in mental masturbation.

--Ethan

[Steve]

the fact remains that the main reason some people perceive a change in sound is due to their own changing perception.

Ok, some sounds reasonable. Differences in design, masking effects from components, room, expectation bias against perceiving differences, etc can account for why some perceive no change.

Would you be kind enough to provide evidence that only one specification, harmonic distortion figure, can support your claim that your example of an amplifier is superior to others? Thanks Ethan.

(In the next paragraph, the higher the negative feedback signal, the higher the negative feedback ratio, the closer the feedback signal amplitude is to the actual input signal amplitude (except it is 180 degrees out of phase at midband. High feedback ratio may be for one volt input, .9 volts or higher may be fed back from the output to the input. This means more stages are necessary for a given gain with feedback applied, and more effects from warmup which affecting multiple parameters.)
 
Suppose we use extremely high negative global feedback signal (also called an extremely high feedback ratio) for Ethan's example of .001% harmonic distortion. We now apply an input jack signal at time point "A". Extremely high feedback signal "B" typically arrives back at the input jack 6-12 microseconds (us) late with respect to time point "A". This alters the waveform and sound since human hearing is sensitive to 5us changes, Jneutron sources say 2us. (known for decades, and high frequencies affected more than lows). These parameters are also affected by warmup as well.

Next, very high feedback signal "B" reappears at the amplifier output again, (only slightly lower amplitude because of high feedback ratio) and on the input jack a second time, this time 12-24us delayed from original time point "A", so the waveform is altered even more. Of course waveform distortion affects the high frequencies far more than the low frequencies and is not measured by a distortion analyzer.

So although extremely high negative global feedback appears very good on the surface, we simply traded harmonic distortion figures for another form of distortion. Now if we lower the feedback ratio, this waveform distortion lowers while harmonic distortion rises. So we are simply lowering one form of distortion while raising another.

We can hypothesize all day long about what changes occur inside an amplifier as its temperature rises. But if those changes are never actually audible, it's just an exercise in mental masturbation.

The changes in warm up parameters of components are scientific and well understood for many years. (Sources, late 60s college, and the RCA Radiotron Designers Handbook, 26 engineers, 4th edition, addresses the subject in 1952.) No one appreciates your negative mis-characterization that we are simply "hypothesizing. 

----------

In general to all, the danger of pushing "audio perception" testing to the for front above measurements is that it is an inexact science, and companies can simply pick and choose which tests to use for marketing purposes. It is also well known that audio testing is easily manipulated to a conclusion of no sonic difference (especially by those associated with big business/inexpensive companies for marketing purposes/public opinion manipulation).

This allows big business and/or low cost manufacturers to falsely claim "all amplifiers/preamplifiers sound the same" which of course is absurd due to different designs, different parts/quality affecting the sound etc.
 
In the end it amounts to marketing tatics reigning at the expense of science. 

I hope this post meets the guidlines since it is standard main stream science, just as my previous posts are. I really don't see the need to post again since lots of evidence has been presented concerning warmup and sonic changes, and most of the material has been covered.

Cheers.

[KnowTalent]

Does anyone here believe that a capacitor can take 400 hrs to "break in"?

[*Scotty*]

I have had electrolytic caps used for coupling caps to block the DC on the output of a single ended class A buffer take a long time to break in and fully form. The voltage biasing them was 2volts and the ac signal was less than that. Four hundred hours is about two weeks time and I am certain it took at least that long in my case for the cap to stop changing. I would say the time for break in to occur would depend on the cap construction and applied voltage.  I have no problem accepting that in some cases people may be able to hear changes in a caps impedance curve after voltage has been applied for some period of time.
Scotty

[j beede]

I have had electrolytic caps used for coupling caps to block the DC on the output of a single ended class A buffer take a long time to break in and fully form. The voltage biasing them was 2volts and the ac signal was less than that. Four hundred hours is about two weeks time and I am certain it took at least that long in my case for the cap to stop changing. I would say the time for break in to occur would depend on the cap construction and applied voltage.  I have no problem accepting that in some cases people may be able to hear changes in a caps impedance curve after voltage has been applied for some period of time.
Scotty

Troubling. I have heard of designers who select caps by ear--physically swapping them in and out of a circuit in real time (e.g. Shindo, D. Wilson). Perhaps they remove those caps from burn in boards, powering the board down just long enough to remove the "formed" caps and then inserting them into the circuit being tested before the caps "re-form"? Sounds painful.

In your case, if the caps are left discharged/shorted after "forming" for a long time do the curves return to the "before" curve?

I wonder what a small change in the impedance curve of an ac coupling capacitor sounds like. I suspect that the difference between an electrolytic and a non-polarized cap might be larger. Has anyone done this particular experiment?

[*Scotty*]

I agree it sounds like a pain in the butt. I tested three electrolytic caps a BlackGate,Ruycon ZL and a Panasonic FM.There was a clear progression towards a cleaner window,with the FM cap being the clearest view of the performance.
The differences in sound between the caps were not subtle.
Scotty

[jsaliga]

I agree it sounds like a pain in the butt. I tested three electrolytic caps a BlackGate,Ruycon ZL and a Panasonic FM.There was a clear progression towards a cleaner window,with the FM cap being the clearest view of the performance.
The differences in sound between the caps were not subtle.
Scotty

Interesting.  Subtle or not there's a number of flaws with your methodology.

A subjective perceptual test in which the size of the subject pool is one is not statistically valid, and the test itself lacks the proper controls.  So your test really doesn't prove anything meaningful.  Put a control group in, and increase the size of the subject pool, stick to a scientific protocol, obtain the same results, and then you may be onto something. 

This is where these discussions always wind up, with neither side budging.  I mean c'mon folks, it isn't like this is the first time that this topic has been brought up in the past 20 or so years.

In response to Ethan, the reason that no one has been able to point to scientific research that follows the proper protocols with the required controls is because there isn't any.  I don't know why that's the case.  Perhaps the questions that are of considerable interest to audiophools are not substantial enough to merit the attention of independent researchers.  Or perhaps the answer to these questions really doesn't benefit society at large, and with limited resources there are more important questions that need to be answered.  And perhaps it is because audio manufacturers are reluctant to help fund independent studies...because while they could win they stand to loose, and loose huge.

I don't know that audible break-in does or does not occur.  I believe that in some devices, such as speakers and phono cartridges, it does...and in other devices, such as amplifiers, it does not.  Earlier this year I spent several thousand dollars on a new speaker system, and initially I was very unhappy with their sound.  I thought the bass response was too boomy and the highs were a little too bright.  I did not have a lot of placement flexibility but the vendor did work with me to try to tune the ports and the compression driver.  I was having a tough time with it.  The vendor offered to take the speakers back if I wasn't happy, but I really wanted to give this more time and a fair chance to work.  Eventually the sound of the speakers began to improve on their own.  I don't recall how many hours were on the drivers at that time but if I had to guess I would say it was somewhere around 150 hours.  Did the performance of the speakers change because of break-in?  I think it did but I really don't know that for sure.  I do know that I am now very happy with my speakers and I think they sound great.  But for all I know what took place over those 150 hours was that the speakers didn't really break in, the listener did, and all that really happened was that I became acclimated to their sound.

--Jerome

[Steve]

Interesting.  Subtle or not there's a number of flaws with your methodology.

A subjective perceptual test in which the size of the subject pool is one is not statistically valid, and the test itself lacks the proper controls.  So your test really doesn't prove anything meaningful.  Put a control group in, and increase the size of the subject pool, stick to a scientific protocol, obtain the same results, and then you may be onto something. 

This is where these discussions always wind up, with neither side budging.  I mean c'mon folks, it isn't like this is the first time that this topic has been brought up in the past 20 or so years.

In response to Ethan, the reason that no one has been able to point to scientific research that follows the proper protocols with the required controls is because there isn't any.  I don't know why that's the case.  Perhaps the questions that are of considerable interest to audiophools are not substantial enough to merit the attention of independent researchers.  Or perhaps the answer to these questions really doesn't benefit society at large, and with limited resources there are more important questions that need to be answered.  And perhaps it is because audio manufacturers are reluctant to help fund independent studies...because while they could win they stand to loose, and loose huge.

I don't know that audible break-in does or does not occur.  I believe that in some devices, such as speakers and phono cartridges, it does...and in other devices, such as amplifiers, it does not.  Earlier this year I spent several thousand dollars on a new speaker system, and initially I was very unhappy with their sound.  I thought the bass response was too boomy and the highs were a little too bright.  I did not have a lot of placement flexibility but the vendor did work with me to try to tune the ports and the compression driver.  I was having a tough time with it.  The vendor offered to take the speakers back if I wasn't happy, but I really wanted to give this more time and a fair chance to work.  Eventually the sound of the speakers began to improve on their own.  I don't recall how many hours were on the drivers at that time but if I had to guess I would say it was somewhere around 150 hours.  Did the performance of the speakers change because of break-in?  I think it did but I really don't know that for sure.  I do know that I am now very happy with my speakers and I think they sound great.  But for all I know what took place over those 150 hours was that the speakers didn't really break in, the listener did, and all that really happened was that I became acclimated to their sound.

--Jerome

What is really sad is that the scientific community itself doesn't appear to agree on proper protocol and controls by all the arguing going on over the years. And you will hear every excuse in the book including how they are the real scientists etc. Check the association of authors to sponsors, employers, associations (and those sponsoring them), etc who perform the "scientific" tests. It seems to always be about swaying public opinion/marketing.

Cheers.

ps. Oooops. I am bowing out.

[turkey]

The reason I chose those two cable geometries is that they are in common usage in the audiophile community and people within that population would agree that they sound different from one another. What defines an abnormal geometry when it comes to audio ICs is something I can't answer. The pundits say that DC to light bandwidth is not necessary in a cable carrying audio band signals. I think the two examples I gave would fall into the realm of adequate for audio signal transmission.

I'd say that coax is normal for an unbalanced RCA-RCA patch cord. The two you describe have no shielding, and are probably what I would describe as abnormal.

Yes, I know you specified twisted pair for one of them. Unfortunately, twisted pair doesn't work with an unbalanced connection where you don't have a differential input stage. But I'm sure you already knew that.

I would think the chance to learn something new about yourself and your system would be compelling enough to get you try the experiment.

I really don't see that playing with wires is a good use of my time. It's just not a pressing issue IMO. Been there, done that, etc.

The last time I listened to wire was on a system with expensive Gallo speakers, Audio Research electronics, a very expensive turntable and cartridge (I didn't pay attention to the brand names because I'm not into LPs at all anymore), and a Meridan CD player. There were power cords that looked like firehoses, CD demagnetizers, Bybee gizmos, etc. All the goodies.

He said he had upwards of $40K in his system, and he works in the business, so I think we can assume that his system was "resolving."

I didn't hear any difference between OmegaMikro wire and some Zu Audio wire of more normal construction.

I've done controlled tests of wire on enough systems, and with enough different people to conclude that the big differences people report in in sighted tests simply don't exist.

The construction phase of the cables would take longer than the listening test session. You stick the cables in and you hear a difference or you don't. The differences between the two should easily be gross enough in nature that no SBT or DBT testing is necessary.

So one or the other of them has gross FR irregularities?

If you are not interested in the cable geometry experiment that's cool,frankly I suspect that if you do the break in test with couple a Shack patch cords you may not hear any difference. You may have a chance if you system resolution has

So Radio Shack patch cords are immune to "break-in?"

[turkey]

Well, when you compare crap to crap, what else would you expect?

Are they crap because they don't cost a lot? Or is it because of the brand name? Just exactly what are you claiming here?

Do "crap" cables not break-in? Because the point of the test was to see if wire breaks-ins.

[turkey]

Good point, face.  A few posters here who come from a certain owner's circle feel that the point of this hobby is to spend as little as possible and then to sneer at those who would rather make an investment in truly high-end gear.

Ah, so you're that Bill person that hates Frank? I wondered why you popped up suddenly and started replying to my posts.

You should probably give it a rest, because I've said that I didn't hear a real difference between my AVA gear and a Pioneer receiver. I hardly think that qualifies me as an AVA cheerleader.

I feel that one should buy a good set of speakers, and also get their listening room into the best shape possible or practical.

From there, I would rather buy music.

I'm not sneering at your spending $100K on an amplifier or whatever. I'm just saying that, when I have done controlled tests, nobody could tell the "truly high-end gear" from a $200 AVR.

Ditto for expensive wire versus what some people here call "crap."

Actually, if you want sneering, look at the people who seem to think my hearing or the products I choose are "crap."

[catastrofe]

Bruce Brisson of MIT has something to say about this in his recent Dagogo interview:

http://www.dagogo.com/View-Article.asp?hArticle=793

[turkey]

In general to all, the danger of pushing "audio perception" testing to the for front above measurements is that it is an inexact science, and companies can simply pick and choose which tests to use for marketing purposes. It is also well known that audio testing is easily manipulated to a conclusion of no sonic difference (especially by those associated with big business/inexpensive companies for marketing purposes/public opinion manipulation).

This allows big business and/or low cost manufacturers to falsely claim "all amplifiers/preamplifiers sound the same" which of course is absurd due to different designs, different parts/quality affecting the sound etc.

I don't think I have ever seen any manufacturer of audio equipment claim that their products sound the same as another maker's do.

[turkey]

Bruce Brisson of MIT has something to say about this in his recent Dagogo interview:

http://www.dagogo.com/View-Article.asp?hArticle=793

If his claim is true, it should be easy to measure. Too bad nobody ever presents any measurements to back up this claim.

One major flaw is that audio signals are AC, not DC. I guess 'ol Bruce missed that.

[lcrim]

This topic is now locked.

There is nothing like a friendly discussion going on here.  The need by some members to impose their own viewpoint on this topic and others is causing me to reevaluate whether this Circle serves a positive purpose.