There are two threads offering non conclusive explanation on this topic, however, burn in as per this thread deals with electronics that start harsh and after a while the harshness stops, this after a while can vary greatly between devices for some this harshness never stops.
1. What causes this harshness ?
2. What changes could be occurring in these devices that the harshness stops ?
3. What components in these devices could be the greatest culprits ?
4. What is the frequency spectrum of this perceived harshness ? Some charts would help
5. What is the difference of harshness due to bad design and components specific harshness ?
1. What causes this harshness ?
2. What changes could be occurring in these devices that the harshness stops ?
3. What components in these devices could be the greatest culprits ?
4. What is the frequency spectrum of this perceived harshness ? Some charts would help
5. What is the difference of harshness due to bad design and components specific harshness ?
Last edited:
Do you mean electronics which is actually measurably harsh and then actually measurably not harsh, or just perceived to be such by one or a few listeners? If the latter, the explanation is more likely to be within the listeners than within the electronics. Two exceptions:
1. mechanical things like transducers
2. electrolytic caps used as coupling caps may need a few hours to settle down and reduce their leakage, especially if they had sat on a shelf for a few years after manufacture or if vintage gear is being used
1. mechanical things like transducers
2. electrolytic caps used as coupling caps may need a few hours to settle down and reduce their leakage, especially if they had sat on a shelf for a few years after manufacture or if vintage gear is being used
There is a word for trying the same thing over and over again while expecting different result. That word starts with a letter "i".
This is a controversial topic, I will add my observations and findings. Burn in per se does exist and an example is electrolytic capacitors final forming during initial operation with reduction in leakage being one measurable parameter. Film capacitors in conjunction with thermal cycling may mechanically stabilise with consequent minor change in characteristic.. These initial 'drifts' may or may not be subjectively significant.
So that's two possible mechanisms, in practice I find more. One observation that I have confirmed many times after construction of new amplifier stages or blanket resoldering of existing amplifiers is that of initial power up subjective harshness and outright distortion. After first noticing this I adopted a standard method of initial operation of amplifiers after major rework. This entailed fine bias trim and fine dc trim over two hours or so to establish equilibrium and without signal or load. Next is to connect known speakers and apply very low level audio and take a listen whilst periodically rechecking bias/dc. During this period many amps sound awful and nudging up in level makes the distortions even worse, until a big momentary peak occurs and then the sound changes a bit for the better. This process repeats as level is increased in stages until the first mild clip events occur and then there isn't much improvement change after that, and returning volume back down results in clean sound as it should be. This distortion harshness is magnetic/hysteresis in origin I suspect, maybe component leads taking up new magnetic orientations according to operational peak currents in individual component leads or something like that ?. With normal initial run up procedures (run test tones/loads or connect speakers and just crank it) this behaviour will go unnoticed and it occurs once only.
Dan.
Fixed formatting problem and removed subsequent posts.
Component leads these days are likely to be copper plated steel with a tin overplate, rather than the more expensive tin plated copper of the past.
Since copper is not ferromagnetic, and tin only paramagnetic, any non-linear magnetic behaviour would be restricted to steel cored component leads
However, I suspect that the magnetic flux density within the steel core may have to be high in order to produce a significant hysteresis loop. This may require higher current densities than are typically found in the component leads in an audio amplifier.
Yes, more so some manufacturers were copper others steel, and now we have smd with nickel endcaps.
Yeah these observations are regarding 'old school' amps with all manner of TH component provenance/materials.
There is some kind of 'initialising' going on if you listen out for it you might catch it.
Dan.
- Status
- Not open for further replies.