Head-Fi.org › Forums › Equipment Forums › Sound Science › Science behind burn in
New Posts  All Forums:Forum Nav:

Science behind burn in - Page 2

post #16 of 58
wet electrolytics have very well known issues with the electrolyte/oxide dielectric "forming" process being somewhat reversable
without correct polarizing V the oxide partially "dissolves"

some time at operating V is required for Al electros to reform full strength oxide layer after long storage at 0 V

Hi-temp/Long life types with generous V derating are a good idea if Al electrolytic have to be used in the signal path

doped Si is also fairly piezo-resistive - packaging&assembly stress can cause op amp Vos shifts, high grade DC instrumentation should be stress relieved by slow thermal cycling of the assembled board

and I think moisture absorbtion could be a larger factor than most appreciate - with a dielectric constant of ~80 the fractional % water absorbtion of most plastic PCB substrate, part packages and some cap dielectrics causes measurable differences in dielectric loss - the heat of continuous on Class A circuits will cook out some of the equilibrium moisture

I still doubt that with all above taken together the response variation would often be audible in controlled testing

"acoustic memory" is just too plastic, combinined with accomodation and expectation effects leads me discount any "I heard it" "just listen" ancedotes - real ABX controlled, blinded tests would be needed to convince me any of the likely electronic burn in phenom rise to audible thresholds


the best mechanical "burn in" story I've heard was about a "hot" piano mod, the professional player waxed nostalgic over his recorded performance and then regretfuly pointed out he would be unable to repeat the performance - the modded piano action had gone from "fast and fluid" to loose and unplayable as the mod "wore in" over a few months of playing
post #17 of 58
I think headphones certainly experience a burn-in period. My DT 880's sound much better compared to an out-of-the-box pair that I've demoed at a store. I'd say about 6 days continuous playback was about how long it took (but then - the psychological side of burn-in time may be affecting my estimate here...).

As a mechanical engineering student, my guess to explain it is just that the various materials need to be oscillated several times before they "settle" into place. Having completely disassembled my DT880's - I can guess how they were assembled. From that, I think I can say that it's a pretty hasty process (because it's all friction-fit, with few actual fasteners). As a result, there's a bit of play initially - especially in the resting position of the plastics. With some vibration and pressure, they find their places...
I suppose a manufacturer would also put 'phones through a testing procedure (of some considerable time). But I doubt they'd concern themselves with the precise behavior of the components during the initial period of use. It's cheaper and easier for them to disregard it, and simply check if the end product performs they way they want after x-number of hours. (Which is all we want as consumers anyway, right?)

feel free to criticize me now...
post #18 of 58
Assuming you are listening to the components everyday, psychological burn in may well play part of the perceived change. If you only poke in for short periods during weekends while allowing burn in of components for weeks, that has little impact. They do change sonically over time.
post #19 of 58
I can say with 100% certainty that computer components do burn in. I have witnessed many instances where certain CPUs would overclock marginally better after server hours of run time. Silicon is a strange substance. Currently, my friend's PC which has been OCed for years with the exact same settings, can now no longer be down clocked or up clocked. It must remain at those settings or instability occurs.

What does that have to do with audio components? Not a single thing.

When it comes to electrolytics or speakers(they have actual moving parts) or CPUs(nano switches) They can and will exhibit changes over hours of use. Whether those changes are audible is a mater of perception. Fortunately for me, I can't perceive any changes what so ever.
post #20 of 58
Heya guys, I have a question...

Say someone buys a new source, amp, and headphone with 0 hours on all of them. He/she then connects it to the source and the headphone in the socket, places the CD in the tray and sets it to repeat. The owner plans to leave the components running for a week straight.

Now... what happens if the owner accidentally left volume pot on the amp is set to 0 (off)? Would any burn-in hours commence on the components? If so which components (or parts within the components) are experiencing burn-in hours?

Any input is much appreciated.
post #21 of 58
Quote:
Originally Posted by digger945 View Post
I've seen this on a lot of capacitor data sheets, this one especially...

Quoted from this Nichicon VX data sheet.

Maybe it's just this one capacitor from this one company, or maybe after 2000hrs it's just like it specs out just like it did initially.
Maybe Nichicon is "pro-burnin."
I think they mean it the other way around, that they can guarantee the specs up to 2000 h's of use at a temperature of 85 degrees C.
post #22 of 58
Quote:
Originally Posted by NelsonVandal View Post
I think they mean it the other way around, that they can guarantee the specs up to 2000 h's of use at a temperature of 85 degrees C.
Those specs are intended to show the cap's endurance. Ie., after 2000h, this is how well they will perform. In any case, it shows that capacitors' specifications do change with use.
post #23 of 58
I think it's funny that (almost) all burn-in reports are for the better.

It's also funny that it's often harsh sounding products that gain from burning in, and eventually sound smoother. I haven't seen that many reports of products going from smooth and soft to aggressive and bright.
post #24 of 58
I worked on a large repair floor repairing electronics once. All of the items we repaired on my floor were burned in for eight hours after repair. Products were calibrated before burn-in, then tested to see if they still met specs. Some of these items didn't always make it through burn-in. So, it was a service we offered to our clients as they didn't get products which will fail soon upon arrival.

I worked for a major audio manufacturer for a while as well. They burned in amplifiers for one hour with specs I don't want to give out. During that one hour, there were amplifiers which failed due to little things which spot testing, or "final" testing on a production line don't catch. One example I can offer had to do with sil-pad material. There were output transistors screwed to a heat sink. The small metal flakes from drilling, didn't all wash away quite well enough. The holes drilled were where the output transistors were mounted to a heat sink. In between the heat sink we had sil-pad to keep the devices from shorting to one another. Well, after burn-in for one hour, some of the filing were able to cut through the now softer sil-pad material, and in some cases short to complimentary output transistors.

In the case of the audio amps, burn-in was very good.

There is another phenomenon which takes place with surface mount capacitors. The heat used for reflowing and even hand solder tend to skew their values. The time it takes for the caps to come back to their actual values isn't instantaneous by any mean. I forget the name of this problem, but will look it up again if anyone cares.

Now, how about that electron migration?
post #25 of 58
I have often wondered whether reports of headphone burn in weren't related to breaking in the pads on the phones, thereby changing the seal and/or positioning of the phones on the ears. Certainly this could explain some reports of breakin while wearing the phones, or possibly if the phones are mounted on surfaces which depress the pads.

I recently got a set of Stax 007a's which showed a modest breaking of about 20 minutes. I can't say I noticed much after this. However I fiddled with the phones trying the mods, which have been discussed in the Stax thread, and so the phones did change their sound and this may have obscured further beaking changes. See http://www.head-fi.org/forums/f113/t...estion-381975/

With electrostatics you have issues of charges on the diaphragms and you can even get electret effects in which the diaphragms develope a charge in some cases contrary to the intended charge which could effect sound.

However I wouldn't discount that flexing diaphragms could breakin material by loosening it up.

However these effects may not be universal, i.e. they may be related to the particular materials used in phones.
post #26 of 58
As noted above, burn-in reveals faulty components. Generally burn-in with electronic components is used to break things. If a component is weak, IE a transistors internal wiring is not quit right, it is likely to break within the first 100 hours of use (give or take, taking cost of testing into account). After this they are exponentially less likely to fail. Same could be said for capacitors, inductors, and resistors. They might be several times less likely to fail then an active component, but quality control is not 100%... anywhere.

With that said. I'm not convinced burn in of a capacitor affects sound quality at all. I do believe that after I built my loudspeakers, and crossover. After sometime of use they didn't sound as harsh on the high end. This could have been strictly due to the tweeter breaking in, which I do believe effects sound. Or this might have been due to the capacitor used in my High pass filter, maybe a combination of both. Maybe nothing at all. But to me, it sounds better now.

So I'll continue to burn-in my new Grado Sr80s and tomorrow I will pick them up and probably be convinced that I notice some difference. Hands down the biggest difference will be the headphone amp I'm gearing up to build... I wish digikey would hurry up!

Oh yeah and this is my first post, you all have been helpful in my decision to guy the Grados, so thank you
post #27 of 58
Thread Starter 
After trying to AB test two amps and seeing how hard it is to discern a difference when switching with only a second or two delay, I can't imagine that someone could register an accurate differences 100 hour later. Pyschology is sooo important to how you perceive sound, components could sound more/less harsh if you are tired, in a good mood or after you listen to them for a while. I definitely don't have the ears to accurately quantify a sonic signature as being too harsh, too warm, too bright, so maybe its just me but I highly doubt someone being able to remember how something sounded 100 hours ago. I'm not saying it can't or isn't done, I'm just saying I doubt it. If you want to test for burn in, you would have to have two units, burn in one and then switch back and forth, that is the only realiable way.
post #28 of 58
I don't think flowing electrons "burn in" the circuit. I think it's the settling of the components from exposure to heat and cooling down.

Amps and sources usually get hot, and with no doubt, the heat changes the structure of the components on the circuit board. This structural change is probably permanent after a certain amount of hours of constant burn-in.

The real question is how the heck does this molecular change make the amp sound "better" or "more mature", etc. stuff Head-fiers make up. Perhaps when the molecular structure of the circuit hasn't settled down yet, things don't conduct currents as well!

Who knows! I haven't heard any clear-cut instances of burn-in, but I do it anyway just to be on the safe side. It's sort of like plasma TV burn-in - all it is is electricity running through the screen, but in the end, it has been said to reduce image burn-in in the long run.
post #29 of 58
So who believes in cable burn in? Just read a review of a Sennheiser cable where the reviewer believed in burn in, to me, he lost all credibility for that review.
post #30 of 58
Quote:
Originally Posted by Sovkiller View Post
<snip>
I would like to see those really golden ears that most of the times are behind this theory in a really double blind test trying two identical amps, to determined which is the burned in, and the one new...

OTOH mechanical, and electromechanical devices, are a completelly different case...
He'd have to be reeeeal quick, too. Because by the time he tries to figure out which is which, the other amp will be allready "burned in"

EDIT:mbd2884@ I personally find cable burn in sort of hard to believe in but somehow, I still wouldn't dismiss the possibility that a virgin cable will perform kinda funny the first time he (or she) meets electricity
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Science behind burn in