No, what he's referring to is standard practice for equipment that must not fail (e.g. military). All components and the resulting assembly suffer from what's called infant-mortality, whereby new equipment can suffer early failure. Once past a certain threshhold, equipment that didn't fail is likely to be highly reliable for a long (and usually predictable) time. Critical equipment will be taken out of service before it's time-expired. This doesn't only apply to electronic gear, things like jet-engines are treated similarly.
IMO burn-in/break-in is total nonsense! And if it's true (or to the extent it's true), it's a very bad thing!3. If the specs/performance change after burn-in (or several hours use) by the end user, the manufacturer's factory tests & specs mean nothing... It's a poorly designed & built product by an incompetent manufacturer! If the performance/specs are not stable, the manufacturer has no idea if the customer is getting a great product or a crappy product. There's always going to be some performance drift, aging, and some measurement variation. But the drift should be insignificant, and most importantly the product shouldn't drift out of spec (at least for a year or more).
Don't we have some type of acoustic memory, a frame of reference to what things are "supposed" to sound like, that may take some time to accommodate a change in that frame of reference?
DVDdoug, how much of the production cost (not selling price) of the device your company produce is due to QC (burn-in, factory calibration etc...)?
Can't it be the manufacturer is aware of changing specs due to ware and tare/burn-in and compensate for this? As conducers are not purely electronical but also mechanical (moving/vibrating parts), I expect they can ware. Bend something often enough and it's integrity will degrade (hair fractures). Compensating will likely have to be a compromise and aimed at an average if the performance keeps changing throughout the product's life. I wonder if this break-in or ware is linear or has a steep curve in the start and then levels out...
Wait.. doesn't that mean you essentially ship pre-damaged equipment?You light a match and say "this one works!" and put it back in the box.
Quote from: Nessuno on 18 April, 2013, 05:20:09 PMDVDdoug, how much of the production cost (not selling price) of the device your company produce is due to QC (burn-in, factory calibration etc...)?The last time I was doing "shake & bake" testing it took a 25 cent part up to about 2 dollars. This was with hundreds of parts in a rack.
If you've never worked in manufacturing, you might be surprised to learn that some "brand new" products off the production line have been repaired. I knew it was common in electronics, but I was surprised when someone who used to work for an autombile manufacturer told me she'd done "body work" on the production line!
If you are doing high volumes, say making iPhones, you've got to do better and get almost 100% quality without "fussing around" troubleshooting & repairing things.
One big difference between Packard and Rolls Royce Merlin engines in WW2 were that the RR engines were all hand fitted, which made field repairs far more time consuming. Packard scrapped that approach for their US production and replaced hand fitting with precisely made parts that just fit, so any field repairs could be performed with off-the-shelf parts that fit into any engine.
Quote from: MikeFord on 18 April, 2013, 07:58:42 PMDon't we have some type of acoustic memory, a frame of reference to what things are "supposed" to sound like, that may take some time to accommodate a change in that frame of reference?Yes and it lasts on the order of a few seconds. Anything beyond that point is prone to being influenced by factors that may not even be related to the actual experience of interest.This should not be news to anyone who has followed one of the many similar open discussions that exist on the forum.
both sound tuneful/musical.
Quote from: MagR on 20 April, 2013, 04:18:32 AMboth sound tuneful/musical.You need to understand that these words don't mean anything.
The guy Tyll from innerfidelity measured the effect of "break in".From what I've understood of his article, the changes are too small, and he didn't want to conclude anything .What I see, there are less changes in the frequency response with time, than there are differences between left & right.http://www.innerfidelity.com/content/measu...headphone-break
Another term was added to the placebophile lexicon.PRAT -> PRAPT
4. Why is it that the product always seems to improve with burn-in? Why doesn't it sometimes get worse?