 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.149.197.165
In Reply to: Re: Playing the devil's advocate posted by Ted Smith on June 19, 2006 at 16:28:52:
I am not familiar with this particular model or design but my initial reaction is one of extreme skepticism. It seems to me it is full of pitfalls and drawbacks not the least of which is that it responds to ambient sounds which are not indicative of the cone's position. The correct design goal is to correct for the difference between where the electrical signal driving the cone/voice coil combination would put it and where it actually is at any given instant and that is all. Anything which introduces extraneous variables will inevitably compromise performance probably to a degree where it often degrades, not improves performance. Unless I saw convincing evidence to the contrary, I would choose a non feedback model instead. Did you say this design was offered by Velodyne? As I recall, their usual design incorporated accelerometers as sensors. Ideally, a feedback element would look at instantaneous position directly but the next best sensor is a velocimeter and following that an accelerometer. This is because the signal from a true position indicator requires no further correction than proportioning while a velocimeter requrires an integrator and an accelerometer requires double integraton. This can introduce further errors in determining position compromising the accuracy of the correction signal. With a name like "Velodyne" you'd think they'd use velocimeters wouldn't you?
Follow Ups:
HowdyNope, I didn't mean to imply it was Velodyne. Since it was a prototype I don't want to color people's preconceptions about the line (even if I remembered the name correctly :)
I agree for feed back you need the best most direct measurement of the resultant, but I think it's arguable whether measuring the resultant wave in the air, the cone position, or using an accelerometer (and integrating twice) would give the most accurate answer.
I also am not a feedback fan, whether in speakers or amps. I like better the idea of designing as linear of a system as possible.
-Ted
"I also am not a feedback fan, whether in speakers or amps. I like better the idea of designing as linear of a system as possible."While you may be in the majority of high end audiophiles who have been convinced that negative feedback=bad, the overwhelming majority of professional audio engineers wouldn't consider specifying or installing an amplifier which didn't use negative feedback. Beyond that in the wider world of electrical engineering, negative feedback (servo control) is considered a fundimental principle and tool for virtually all control systems. The modern world would not be possible without it. The basic tool for many modern control systems is the PLC (programmable logic controller) and its principal control element is the PID loop (proportional, integral, differential.) The applications have been simplified by the hardware but the mathematical analysis is still a very tough one and is usually presented as a senior electrical engineering course or a grad course. Even for people working in the field, I'd bet they rarely get solved for real world problems. PLCs facillitate rapid adjustments to servo variables by simple software changes. Improperly adjusted, systems can hunt or even go into spontaneous oscillation. Negative feedback amplifiers and servo speaker designs are subject to the same flaws if the design is poorly executed and that may often be the case.
HowdyWell most of the pros I know don't care about whether the amp has negative feed back or not, they care mostly about robustness then specs. and price point.
I'm not sure it was obvious, but when I said "I also am not a feedback fan, whether in speakers or amps. I like better the idea of designing as linear of a system as possible." I was refering to audio uses, not general control theory.
I've implemented my share of control systems both feedback and (in one case) feed forward. I understand (fairly) well what negative feedback can do and what it can't. In audio some designers tend to use it to (try) to cover a multitude of sins.
Anyway I'm happy with my no (overall) negative feedback amps.
-Ted
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
