How I see, for many of you behavior of amp by the clipping isn't interesting...so little example: once I had measured amp Bryston 4SST for test in one magazine. This very well made amp with good parameters had one mistake: behavior by clipping. By low and mid frequencies was all OK, but at higher frequencies it was horrible: when by 10kHz output running only lightly " touch " the clipping, power consumption at mains side jump change from 4A to 8A, i.e. transverse current in amp vastly rise...
Similar behavior have many amps and it lead to potential unreliability...
Similar behavior have many amps and it lead to potential unreliability...
I have asked an expert on commercial amps for live performing bands. He said that the usage of limiters for every source had been mandatory. In fact the circuit of limiters was more of a top secret than make of amps or speakers. Limiters were largely responsible for the charcteristical sound of a rockband.
Thus we have just two design objectives that really matter:
1 avoid clipping by all means even at the expense of distortion caused by soft clipping or just limiting
2 the damping factor must be over a wide range preferably >20 kHz independent of frequency.
In fact the frequency dependency of damping is like a fingerprint of the topology.
It matters, same as with thd, not much how the figure of damping is.
An amp featuring a dynamic damping of say 10 which is constant from 0 to 20kHz will sound audibly better than an amp featuring 1000 with a frequency dependency.
Thus we have just two design objectives that really matter:
1 avoid clipping by all means even at the expense of distortion caused by soft clipping or just limiting
2 the damping factor must be over a wide range preferably >20 kHz independent of frequency.
In fact the frequency dependency of damping is like a fingerprint of the topology.
It matters, same as with thd, not much how the figure of damping is.
An amp featuring a dynamic damping of say 10 which is constant from 0 to 20kHz will sound audibly better than an amp featuring 1000 with a frequency dependency.
I disagree with your last assertion. It is definitely important to have very good DF at high frequencies (too often ignored), but having higher DF at lower frequencies is definitely not detrimental. There will be an audible difference between amplifiers that have DF of 10 and 1000 at low frequencies, for sure, but the former sounds different because it is inaccurate in its frequency response when driving a typical loudspeaker.
Note that some speakers "sound" better when they are driven with a lower DF, but this does not mean that the system is more accurate.
Cheers,
Bob
that could also be due to common mode conduction, made worse by saturating the output transistors
limiting can be done by compression without distortion (unless you count the "envelope distortion" as well).
damping factor suffers with rising frequency because of the lack of feedback effectiveness at higher frequencies. to eliminate thiscompletely would require an amp that has an open loop gain turnover frequency of 20khz, not exactly something easy to do. but because it's low frequency drivers that have the most potential to "misbehave" due to low damping factors, lower frequencies are where a low damping factor is most important. a well designed amp will have a reasonable (100 or more) damping factor below 2khz, where it matters most. since speakers are more apt to misbehave at lower frequencies, i disagree that an amp with a level damping factor of 10 will sound better than one that has a damping factor of 1000 from 20hz to 200hz, sloping to 100 at 2khz, and sloping to 10 at 20khz, since that curve covers encompasses the frequency ranges where drivers are most likely to cause the most back EMF. higher frequency drivers because of their smaller dimensions and stiffer suspensions produce far lower back EMFs than do lower frequency drivers, and so damping factor is not as important.
Bear in mind that back emf from drivers is not the only thing that DF is supposed to deal with. DF is just another name for the inverse of output impedance. It got its name because of its importance in damping LF loudspeakers, but its importance does not stop there.
A low DF will permit any load impedance variation to cause a frequency response aberration. Loudspeakers often go through significant impedance variations near their crossover frequencies. DF wants to be high at these frequencies.
Cheers,
Bob
mmh...Agree with your definition of Damping Factor in the sense that
it s the inverse of the output impedance in the manufacturers
datasheets.
I suspect that they use this definition because it s more marketable.
After all, it sound easier for the consumer to understand that an amp
with a HIGHER DF is supposed to be "superior".
Worse, using the output impedance expressed in ohm could be
even counter productive (always in a marketing perspective), as
the general public impression is that more is better, so it can
lead to some confusion..
Otherwise, the damping factor should rather be the impulse response
damping measurement, i.e , the eigenvalue of the natural pulsation
of the amp.
regards,
wahab
you're right about that. crossovers are often the last thing thought of when designing an amp, even when modeling one... in any well designed amp, there should be sufficient damping factor within the audio spectrum to deal with crossovers and their impedance variations. it helps if the amp has a high GBW product, since the damping factor is proportional to the feedback ratio at any given frequency.
i've been in the process of building a dummy load with a crossover and the electrical equivalents of "real" drivers. it's not as easy as it sounds... i got one finished a while back, but didn't have high enough wattage resistors for the "tweeters", and "let the smoke out" of them at about 50W(average power with music, not just high freqs or a sine wave). i underestimated the amount of high frequency energy in my source material. since i repair some large amps, i want to be able to "torture test" an amp with a dummy load that isn't just a resistor.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.