![]() ![]() |
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
I've often wondered why measurements on an amp show +/-.1dB frequency response but they may sound bright or warm in the bass, etc. Does this have something to do with high fequency distortion making it sound bright? And for bass, lack of damping or how it responds to a real reactive speaker load?
Any thoughts?
![]()
Follow Ups:
Some amps, particularly tube amps, will sound wildly different into speakers with wildly different impedence curves (and there *are* some pretty wild impedence curves out there!) If you get the electrostatic bug, you start dealing with substantial reactance, which can drive some amps to complete failure. Conventional speakers still have some reactance, which can affect the amp. I also think some phase errors are a result of impedence mismatches, but don't really have any principles to base it on - it's just a feeling at this point.Frequency response into a resistive load is a pretty meaningless spec -- no speaker is purely resistive. So are distortion measurements. The amp's output impedence combined with the speaker's impedence curve *can* indicate whether you are likely to have major problems.
.
![]()
Speaker interactions account for a lot of it. As even the best speakers are far from being flat in terms of frequency response. Add to that many other factors concerning amp design, such as propogation delay, slew rate, rise time, distortion characteristics, and on and on. Did I mention that the quality of internal components (caps, resisters, tubes if applicable, switches, and everthing else) has a lot to do with sonics as well. Bottom line is that a flat lab measured frequency response is nice, but hardly a measure of what is going to come out of your speakers.
![]()
They're "flat" doing frequency sweeps into static resistive loads. When driving motors (speaker drivers) of complex impedance, amps have a hard time maintaining that flat response. Amps also deviate from flat response when driving the complex signals of music, as opposed to test signals.I think a good test is taking a large, 4-ohm nominal impedance speaker that is a challenge to drive properly, and doing a frequency sweep mixed in with white noise near its power rating. This can almost guarantee some ugly results from poorer amplifiers.
![]()
you kind of answered part of your question.lack of good damping,possibly caused by too low of a slew rate.too much feedback into the system,causing a softening in the bass and high treble.
jim buck
![]()
The measurements are done using a fixed set of conditions. In a real world application the conditions vary resulting in differing frequency response.
![]()
This post is made possible by the generous support of people like you and our sponsors: