Electronics burn in - Can we finally put the myth to rest?? - Page 3

Same here.

Most household power wiring around here is daisy chained using the crappy vampire teeth "push in" terminations on the outlets in the boxes around a room. Anyone want to bet on the quality of that connection 1, 5, 10 years later?

Not to mention the fact that in a lot of decent audio equipment the power supplies are actually the most costly subassemblies. Which are specifically designed to deal with issues on the mains.
Edited by cptomes - 8/18/12 at 9:05am

Each component has the same budget, roughly. My speakers have MIT Shotgun cables, the CD player has MIT S3's and so on. The turntable still has the factory connections from 1980 or so, as the table is somewhat collectible and I don't feel like digging into the linear tracking setup to replace them unless I have to. Thanks, and yes, the cable definately completes the aesthetic I was going after. Bob does not believe cabling matters at all, but after I was given permission to put his name on these amps, I wanted to do something worthy of the name, and honor some of his legacy, and they also had to look like they were made by the old Carver corp.
Edited by nooshinjohn - 8/18/12 at 9:12am

ah.. i'm done arguing replying. experiments are easy enough to do. do them. deduce your own conclusion. people have varying level of sensitivity. some don't mind if the bass is a bit fat, some don't mind if the midrange is a bit shallow. bass is just bass. midrange is just midrange. an amp is just an amp and THD is just a number.

now i'm gonna go enjoy my holiday.

I often wonder why the companies that make products that "require" long "burn in" times-don't go ahead and "burn them in" before the customer gets the products???

Why make your customer-who has spent a lot of money on the "special" product-suffer through bad sound for a long period of time?

Why not let them enjoy all the benifits that their money has purchased right away?

I bet they would sell more products if they sounded good right away-rather than sounding bad for weeks or months.

Just sayin'

---

Because it gets them past the return period.

To the OP, performance characteristics of capacitors and transistors are known to change measurably over the time frames we're talking about (10s to 100s of hours) especially when heated.

Ref, for example:

http://www.johansondielectrics.com/technical-notes/general/ceramic-capacitor-aging-made-simple.html
http://www.dfrsoft.com/DfRSoft%20-%20Transistor%20Aging.pdf

I don't know enough about the design of modern audio gear to speculate if or how this would effect the subjective performance of the device. I would guess an amplifier circuit would be the most likely place to measure a change.

"Because it gets them past the return period."

:-) and allows the mind to recallibrate to the new sound.

the whole burn-in myth probably started by several people buying some gear and not liking it at first, but then heard that it seemed to sound better over time. the change was occuring in their head (and it is a real change) not in the gear. then just got the explanation reversed. of course the snake oil guys capitalize on this.

"To the OP, performance characteristics of capacitors and transistors are known to change measurably over the time frames we're talking about (10s to 100s of hours) especially when heated."

yeah, but if it were audible, the components would be "run in" like ivan suggets, so that customers don't have to spend hundreds of hours listening to "crappy" sound.

Almost everything can be measured, but whether or not it will make a difference in your system.......... probably not

What if I were to propose a measurable scenario that demonstrated an improvement with a different power cord? You "shouldn't" have radiated emissions off a power supply, but bridge rectifiers be some noisy beasts and improved shielding could be one way to getting that crap out of a really low noise audio input stage. Or you could go to better input cables, or you could design a better input stage - pick your poison.



I'm pretty sure that's not what the audiophile is thinking though.

This: http://gizmodo.com/363154/audiophile-deathmatch-monster-cables-vs-a-coat-hanger

OP, ask your self this... is it burn in, or is it Fatigue? Are they the same thing?

And while we're at it, may as well measure our yardsticks and measuring cups to make sure they didn't somehow change size over time.

Calibrating TVs - always leave on for about 4 hours before even starting. Elsewise, by the time you get to the end of the calibration, the beginning values have all changed. Hence, wait 4 hours before starting.
Don't they say that Plasma TVs need an initial 200 hour Burn In?
When calibrating the same TV over periods of time, one can definitely see the change in the meter readings of the RGB. My TV, when new, had Blue Rising over time. Now that it's older, it's Red that's now rising. So values do continue to change as the electronics heat up, cool down, working into a damp area, then after, dry. All types of conditions are going to affect the electronics. (And how about those 10V caps they put on 12V boards?)
Edited by p5browne - 8/20/12 at 8:50pm

^^^

sorry, poor analogy... think about "why" those panels are exhibiting changes... hint: it's not the electronic bits like caps, etc.

The phosphor coating in plasmas is the only thing that changes.

Ten internets for this post!

Anyone that builds their own gear knows that step one is to build a proper power supply to provide a clean signal to the rest of the rig. Anything a properly-built power supply cannot rectify needs to be rectified through power conditioning, and that would have to be quite severe.

My advice for those that want fancy looking power cables is to buy some nice looking terminals, some braided wire sleeve, some heat shrink, and go to town. My advice for speaker cables is only slightly more advanced. Neither will burn in. The electrons will happily travel from the very first second.

As for the OP's question, I can't add any measurements, unfortunately. My measuring tools are old-school and low-tech. I always break in speakers because I feel that the surround is stiffer out of the factory, and needs to be worked in a bit. To me, the bigger drivers do seem to sound different after the surround softens up, but I've never measured.

I also like to burn in new tubes for a number of hours before I give them any critical listening. I believe that there can be some seem to change in dynamics after being warmed the first time for a handful of hours, more notably for me with some of the NOS Russian and Yugoslavian produced tubes I've tried. But again I've never measured. And my ears aren't acute enough to detect any changes after that initial heating cycle (other than when they're not warmed properly). These changes are subtle enough that they could be placebo.

Certain components, such as some ceramic disc capacitors, are known to "age" over time, gradually changing in property. This isn't a "burn-in" phenomenon, just gradual aging: a continual decay. Sometime such components can be included in builds that are initially over-spec, so that over time they age into a more appropriate range. This would be a very slow process - occurring over potentially many hundreds of hours. Not what I would call "burn-in", but I suppose one could argue that point.

Some of these same components also suffer from measurable changes due to operating temperature, so would measure different when first turned on versus when warmed up to "working" temperature, but again this is not a "burn-in" issue.

Others argue that it's possible for a freshly-soldered board (or component) to need a good heat cycle to "re-flow", which is said to improve the connections hence potentially change measurements. I'm curious how much heat would be required for such a change...

360 degrees F, give or take. In other words, enough to cook many of the components and cause all the joints to fail. This is but one more example of a theory that inherently makes no sense.

Regarding "Burn-In":
For our discussion; I think you have to look at this as 3 categories:


Speaker drivers (and other mechanical devices)
I believe that yes indeed; anything with mechanical movement can change over time with usage (and with different usage patterns).
In the case of speaker drivers, hopefully an improvement

Semiconductors
The following comments are from my experiences many years ago working in a semiconductor reliability laboratory, the short story:
We use to run accelerated life tests by cycling "excersizing" banks of semiconductors in an oven.
An accelerated life test would typically consist of 168 hours of operation cycling and/or under load at an ambient temperature ranging from 125degrees C to 150 degrees C (depending on the package type and rating)
Any changes we would experience; would typically show up as "infant mortality" failure early in the life test (usually a complete failure).
My job was to characterize a devices electrical properties before and after; the majority of the time these characteristics did not degrade (or shift in either direction), from before to after!
Again; when there were changes - the change was often failure, the device simply quit functioning.


Other electronic components, "Passive"; like resistors, capacitors, etc..
Yes these devices do degrade over time and are suspectible to "heat fatigue"
for example; resistors change values, and some capacitors dry out

My opinion the mechanical items will likely change with "burn-in"
If a designer is not conservative in his design, or specs cheaper passive components, then it is quite possible that the electronic portion could degrade with burn-in.

How's that for a definite maybe

You claiming Danley Sound Labs does this?

Good post. This part is interesting to me though because it's generally understood that transistors change over time, especially when stressed. What kinds of components were you testing? I would expect your results from digital chips, but I've seen tests where B changes pretty substantially (10%?) under similar conditions.

Still, 150C is pretty toasty in the real world. I'll agree with your definite maybe, but on analog stages only.

For capacitors at least It's actually the other way around. Soldering heats the component and it goes out of spec until it drifts back.

I was testing mostly digital logic chips (but occasionally would test switching transistors also).
In the case of digital logic >>We were verifying propagation/switching times, and logic levels - virtually no change

For the switching transistors (mostly TO3 cans) we were testing for junction leakage current before and after on the same samples, and then catastrophic tests to see if breakdown voltages changed between baked and unbaked.
Also no deterioration here either (just a few hard failures)

While not always audio components, I worked as an electrical engineer in the electronics validation lab at Delphi Automotive (read: largest supplier of OEM automotive parts in the world [use your Jeremy Clarkson voice]). We simulated 10 year and 100,000 mile operation (and frequently quite a bit more). We tested everything from key fobs to radios to engine controllers. We did shake/rattle testing where parts were literally put on a machine that shook them for hours. We did thermal shock testing where parts experienced wide temperature swings (it has been quite some time since I quit that job but temps ranged from well below freezing to nearly boiling) and many other lifetime simulation tests.

Unless there was a physical failure, any change in electronic functioning (measured output voltages and currents that had to fall within specifications and not just pass/fail testing) from the beginning of the testing to the ending of the testing was minimal. There was often some degradation over this long-term simulation testing. Just as often, however, changes were well within the error of measurement.

We did absolutely no burn in whatsoever. Parts were end of production line tested just like what I would hope you would see from any audio component you might purchase. When we got them, we plugged them in and went to town.

Delphi makes electronic components for GM, Ford, Chrysler, BMW, Honda, Toyota, VW, Lamborghini, Harley Davidson, Caterpillar, and dozens more. The quality of the components used varies because of customer design / price specs. We were using many of the same basic components and semiconductors from the exact same manufacturers as those who make audio components. From lowest quality parts to highest quality parts I never once saw a burn in issue in the 3 years I worked as a validation engineer.

Sorry, I quit the job about 10 years ago else I might have been able to look at millions of data points from thousands of parts more closely. I don't really know how much more data you'd like. While these parts were not high end audio equipment, most of the parts were honey badgers.

Maybe this was true 30+ years ago. But with today's electronics I'd call it a myth. $0.02
Edited by chilort - 8/21/12 at 9:27am

I will add that parts do act differently under different operating conditions (heat but not so much heat that we were headed toward a cascading failure). But when parts were brought back to nominal operating conditions they functioned the same electronically. With the exception of heating something to failure and other tests of that nature (which we did rarely do catastrophic testing for some customers because they wanted to know the gross limits of what they were getting from us) all parts returned to normal operating parameters. When operated within the design ranges the parts were not altered in some way or another.

chilort
+2, I agree!

(by chance were you working in Indiana during those years)

^^^

that's a real good post...

In my experience with electronics most companies perform a burn-in before shipping them. If it does make a difference the manufacture's (most not all) already burn them in. When it comes to Audio equipment I go by my own experience . In the pro audio world (mixers, effects, equalizers, etc.) most of the electronics consist of stuff that doesn't even need to be burned in, the only exception is with tube-amps but all the new ones are already burned-in at the factory.

Yes except the engineers don't believe in magic - burn-in simply refers to the act or turning equipment on and stress testing it, to see if a part is faulty.