Negative Feedback in Audio Amplifiers

AFAIK there are 4 wire systems which run 2 sense wires to the speakers to compensate for cable losses.

No emoticon, so I assume that he's serious. In which case, even though I know nothing about what he's talking about, I do know that he is full of shit. Youtube ... gimme a break!!! Bob

the

me

of

e of

ce

re

or

lt

in

and

that

ing

But

al

gh

yeh, wire resistance "adds" to voice coil dc resistance, whcih effect parameters

Maybe

many years ago I looked a lot at different speaker feedback, negative impedance etc.

The negative impedance needs to be matched to the speaker and cabling so it is most realistic for speakers with build in amplifiers, and then there's the problem of the resistance varying with temperature

there have been speakers with dual coils where one was used for feed back, afair it had problems with varying coupling between the coils

The most interesting one and probably the one that works the best, was a piezo disc glued to the center dome measuring acceleration to do a full PID regulator, reducing distortion and setting the response

With that limit is almost only how much the speaker can handle before it breaks up and the power available

That's all for low frequency like subwoofers but that is also where most of the problems are

-Lasse

It was a joke, but several people were bound to mistake it for a serious proposition.

Perhaps I will make a youtube video and see what people say.

Regards,

Uncle Steve

Maybe

That's crazy. But the best way would probably be laser interferometry with a tiny mirror glued to the driver. Not sure what you'd use as a sensor.

I can see it now. Someone will jam a Raspberry PI in their speakers and run the correcting factor back to the amp over Gigahertz Ethernet

You seem to be saying that the voicecoil in low/mid drivers is the primary target of negative feedback, but would it not also apply to the high-frequency signal components? To what extent I could not guess. I suppose it depends on the design, but people tend to forget that the nominal 20kHz range of human hearing isn't the whole enchilada, and there are several ultrasonic applications where signal fidelity is fairly critical -- like those imaging devices they use to examine a foetus to determine gender and general health of its development.

Regards,

Uncle Steve

Probably a few, I'm a newbie at nuts-and-bolts electronics.

I was referring to the mass of the cone and coil, which will resist acceleration according to the rules of conservation of momentum. There is also the material connecting the cone to the driver chassis, which will contribute it's own retarding influence. The first factor qualifies, but the second may be merely resistance.

I was under the provisional impression that the back-emf, as you call it, would affect the NFB circuitry since the amplifier cannot (or should not be able to) distinguish speaker effects from the distortion introduced by its gain stage(s).

I was not considering the slew rates of the gain stage. The speaker drivers are subject to this phenomenon in the physical sense as you describe above. So far, I am not concerned with signals or transients outside the design envelope of the amplifier.

Never heard of SACD before... I am fairly certain that the amp I have is rated out to 100kHz, although that is unlikely to be linear across the whole range. Momentarily surprising, but there's nothing magical about 20kHz signals in relation to today's electronics components.

I've generally found that CD quality is mostly related to the quality of the studio and engineers that mixed it down from the master tapes. One thing that is interesting in passing is the rich bass characteristic of many recent recordings as compared to the characteristically thin bass from my 80's-era recordings. That may be more a factor of my musical tastes, but it is interesting and may have something to do with the quality of synthesizers and microphones improving over the last thirty years.

You don't really want an equalizer in that path. Better to calibrate the volume against a simple reference signal with a scope.

The easier way to do the experiment you suggest is to use a DSO to measure each amp's output at the speaker terminals and compare. I think if you hooked two amps up to the same pair of speakers (inverting one amp's output) you'd probably damage the hell out of them, but I'm not sure if this is what you are intending.

Of course the problem of using logic in a religious war is that the zealots don't recognize the validity of the logical argument.

I recognize the value of expensive electronics, but I think there is a cost threshold after which the money spent is wasted. Difficult to say what that threshold is, but it is certainly not more than $5000 per component. I do not have the most sophisticated ear in the world and my threshold is probably somewhat lower. My budget is of course much less.

Regards,

Uncle Steve

Are you referring to Philips MFB

As far as I understand, it was used only on quite small drivers (8"), never on big ones.

On a sunny day (Sun, 03 Feb 2013 22:55:34 +0200) it happened snipped-for-privacy@downunder.com wrote in :

I do not have a high opinion of Philips HiFi. In the sixties I bought a 'Philips HiFi International' transistor amplifier, and it had a 4 Ohm and 8 Ohm output socket. Inspection showed the 4 Ohm output was created by adding a 4 Ohm resistor in series with the 8 Ohm output. Their Hi Fi tape deck had about -40 dB noise level, a lot more than my Sony.

Their speaker boxes were OK tough.

I have a big woofer somewhere (not Philips) that has 2 4 Ohm voice coils, you can either use those in parallel, or in series. Perhaps one could be driven, and the other one used as feedback, have not tried that.

But my sincere opinion is that you cannot hear the difference over all the other distortion and acoustic effects, those of the room included, it is all hype.

But then indeed no 2 boxes sound exactly the same. And neither if you move your head an inch. So much for HiFi.

Philips just sold their consumer electronics branch .. cannot remember to whom, but it will continue under the Philips brand. OK, let we look: Funai

series with the 8 Ohm output.

At least in the 1970's, the Philips designs were quite good, but the electro mechanical implementation quite awful. Huge (up to 50 x 30 cm) PCBs were used, suspended on a very flimsy light weight metallic frame.

Lifting it from one end or being stressed by other bending motion caused the metallic frame to bend and also bend the PCB. Since the connectors, potentiometers and switches were also directly mounted to the PCB, in addition to a spaghetti of single stranded connection wires crisscrossing between modules, so sooner or later of a large number of flexing event, the systems became quite unreliable.

While I did some electronic servicing in the early 1970's for my friends, I absolutely refused to touch any Philips audio gear.

Later on, when Philips produced mobile phones, they had to have a 24 h replacement service all over the world. I have never hear any other company forced to offering such service to the general public. I think my friend had his 5th replacement Philips, before he switched to an other producer.

So even you can design a good product, the actual implementation can destroy the credibility.

On a sunny day (Mon, 04 Feb 2013 00:30:06 +0200) it happened snipped-for-privacy@downunder.com wrote in :

series with the 8 Ohm output.

Yes, perhaps management changes. I have worked for Philips several times, once in a service center. I was offered a job in Eindhoven lab, but they did not want to pay enough. They have been very creative as far as technology goes, and they CAN make very good products. Think of the invention of the optical disk (compact disk). I was told in Eindhoven they only have 2 weeks from design to production, so there is a stress level there that is not only mechanical of nature.

They seem to be selling everything, not sure what the philosophy behind it is, these days banks go belly up, good things are sold, maybe too many non-technical types in management, bean counters, not inventors. For consumer stuff of course China is a very strong competitor, bit of a recession in Europe and the US does not help. Some other flower will bloom I am sure.

There is good comedy value in some of the hifi mags.

Years ago I read about a very conscientious speaker designer in the UK who had the goal of making a good cheap speaker. He came up with a good design, but couldn't sell them. Finally (on advice) he had to resort to pushing the selling price up significantly, then they started selling. He was quite discouraged by this experience, while for others, of course, it represented a commercial opportunity.

Remember, most audio phools have air shocks with auto levelers in their cars to make sure the woofers are gravity balanced at all times to avoid unwanted harmonics from what could be eccentric resonating woofers.

Can't have those cones wearing unevenly and generating parasitic material now.

Jamie

It sounds like you screwed up the tweeter sensitivity throwing the speaker out of whack. The only think an amplifier might change is woofer damping, and that does not usually occur. It cannot control tweeters distortions.

Greg

Just loosing one drive coil will decrease damping. Changes parameters, and using one for a feedback is not a great idea. Better to use separate detector. About all you can get is more control of resonance, damping, but then it also depends on box type.

Greg

You are more likely to increase the low end bump. In some cases building speakers, you can add resistance, say, .5-2 ohms, to decrease Q, to get better low end. One way to alter driver parameters.

Greg

I should point out that I didn't read about this in a hifi mag.

Increase Q value I meant.

Greg

Actually I melted the ribbons. I paid to replace the tweeters, and recieved two replacements, one of which was weak. I returned that, and got another weak replacement. I returned THAT, and got a replacement that seemed to be manufactured properly but which was weak. The first two had visual assymmetry of the installation of the ribbon. The third looked O.K., but as I found out today the chassis was warped out of square. At that point I KNEW the local distributor and the store was for sure committed to sending every defective item in their no doubt comprehensive inventory of defective drivers. The email exchange corroborates.

So I decided to tell the store manager to go f*ck himself and started a project to rehabilitate the two spares I ended up with. Early results are encoraging since I now have one speaker that is ten times better sounding than what I originally started with, and the second is due to open up any time now.

So what you think may have happened is completely bogus.

Regards,

Uncle Steve

In the first .com boom, local restaurants discovered that the more they charged for a bottle of wine, the more they sold. Restaurants like Boulevard had several wines in the $1000 to $3000 range.

Beer is the new wine, and The Monks Kettle usually has something in the $80 range.