Naked Science Forum
Non Life Sciences => Technology => Topic started by: Geezer on 09/05/2010 02:59:23
-
I'm a bit out of touch with the latest and greatest when it comes to audio systems. I used to fancy myself as a bit of an Audio Phil, but I rather lost interest when I realized that there was a lot of
utter codswallop slightly dubious technology involved.
What I'm on about is the idea that you spend a lot of money on the lastest amplifier that faithfully produces a voltage across a resistance that is, to all intents and purposes, identical to the voltage that was stuck into the input of the amplifier, just a bit bigger. (Isn't that what an amplifier is supposed to do?)
Then you shove the very expensive amplified voltage into something that doesn't look anything like a resistor (it's a bunch of loudspeakers, filters and stuff) because it has a impedance plot that resembles the Himalayas, and you hope that it sounds really good (good luck!)
Your amplifier may "think" it is providing the ultimate in hi-fidelitee, but it really has not the faintest in terms of the actual sound that is being delivered to the lugholes of poor old Audio Phil who is now stuck with paying off the substantial loan that was used to buy all this expensive kit.
Are things any better today?
-
Electrostatic speakers (http://en.wikipedia.org/wiki/Electrostatic_speakers) were in vogue a few years ago : they double as air filters. [:)]
-
Electrostatic speakers (http://en.wikipedia.org/wiki/Electrostatic_speakers) were in vogue a few years ago : they double as air filters. [:)]
"a few years ago"! Well, OK, if you consider a third of a century to be a few years. [;D]
A former boss who was an EE and a reader at a certain UK University (so he really should have known better) had a set of these.
I never had the nerve to tell him he had wasted his money!
-
Amplifiers can be made very good but speakers are a problem. As I said in the other thread, IM distorton is almost impossible to eliminate. Electrostatic speakers reduce the effect by have a large area with a small movement but the disadvantage is they don't give a lot of power and the large area gives a loss of positional accuracy to the sound source. The modulated air stream tweeter (Ionophone) does avoid IM distortion but to be honest I have never heard one. I think the problem here is also getting volume out without a significant continuous (and presumably noisy) airflow.
I used to be keen on hi-fi years ago but find I enjoy the music from a medium fi system just as much. It makes little difference to me providing it is simply not totally rubbish.
-
In my experience, coming at this from a radio angle, when it comes to electrical kit, "you gets what your pays for," up to a point.
In other words, there is a significant law of diminishing returns, meaning that there is a point beyond which further spend does not translate into appreciable benefit for the majority of listeners because their listening platforn, or the compression applied to the audio they are hearing, renders the material at a lower spec than the system upon which it was recorded.
Therefore, if you try to get by on really cheap gear then your sound will suffer, and you will frequently fall victim to unreliability. Things are usually expensive for a reason and I have certainly found, occasionally to my cost, that going cheap is a false economy. A good piece of gear lasts a long time and tolerates a lot of punishment.
That said, for certain applications there are some makers who turn out good gear at a very competitive price, so sounding good need not cost the Earth.
Chris
-
You get what you pay for, but I would expect a £20 amplifier with £200 speakers to sound a lot better than a £200 amp driving £20 speakers. Also, loudspeakers have moving parts and they wear out in a way that solid state electronics can't.
Incidentally, I have always felt the sound quality was better with headphones and I presumed this was because they didn't need to move much air so they could do it properly. Was I on the right track?
-
Thanks all! Actually, it's pretty amazing what you can get these days for not much money.
I was wondering if anyone is using servo driven loudspeakers. Amplifiers make use of negative feedback to ensure that the output is a faithful reproduction of the input, but the amplifier has no means of knowing what the loudspeaker is actually doing.
By taking a signal from a transducer on the loudspeaker that represents the actual displacement of the voice coil, it should be possible (in theory) to close the loop and have the amplifier really control the displacement of the speaker.
I think Philips did something along these lines about forty years ago, but it does not seem to have gone very far. The technology available today would probably make this relatively simple to accomplish.
Whether anyone would notice an improvement in sound quality is another matter, but that never stopped the Hi-Fi industry before. What was it Flanders and Swan said about attracting bats?
-
Incidentally, I have always felt the sound quality was better with headphones and I presumed this was because they didn't need to move much air so they could do it properly. Was I on the right track?
High frequency response of headphones is better than loudspeakers because the low/mass inertia of the headphone diaphragm. The small diameter of the headphone diaphragm means the bass response is poor compared with a subwoofer type loudspeakers. The stereo image of headphones is better than loudspeakers because there is no unintentional cross-over, (left channel entering right lughole and vice versa).
-
BC, you are right about headphones sounding better. The reason is the reduction in IM distortion because of the reduction in absolute velocity of the diaphragm thereby not doppler shifting, so much, the higher frequencies.
RD, I disagree about the poor bass response in headphones. Because the energy needed is so much lower than for speakers, the area of the diaphragm does not have to be very large at all to give a flat response over a wide frequency range. There is no need to have folded exponential horns or any of the other techniques to get the required power. But, unfortunately, the stereo image is not actually as realistic from headphones though some of the other qualities compensate for this. The shape of the human head and ears contribute to our perception of directionality and this cannot be reproduced by just headphones.
Control of diaphragm movement by overall feedback was indeed done by Philips and others. It does seem a good idea but does not solve the IM distortion problem unless it is accompanied by some very fast digital processing. I don't know whether this has ever been done but I would be surprised if nobody had tried. A lot of work on sound reproduction has been done by the Fraunhofer Institute with DSP techniques. This was mainly aimed at auditorium use and very clever it is too - giving everyone a near perfect stereo image for example - but I don't know whether they have done anything in this area.
-
Thanks Graham.
I never understood the IM problem with speakers until I reread your post on the other thread.
Do you think it would it be possible to compensate for the IM problem by deliberately modulating higher frequencies with lower frequencies (in the opposite phase of course) that were destined to go to a particular speaker? It would require a lot of number crunching, but it would not necessarily have to be done in real time.
We could make a few quid generate some revenue by flogging purveying the necessary software (optimized, of course, for the punter's customer's speaker configuration, air temperature, humidity, altitude etc). Shoot, Dolby is still raking it in, and nobody uses tape anymore.
I can see it now! - IM Cancellation by 2G Industries
The logo will be GG (obviously one will be reversed) and the motto, "Giddyup"
-
I think you need to have very rigid diaphragms on the speakers and employ an optical feedback control system. Then, with a fast DSP it should be possible to accurately compensate for the Doppler generated IM distortion. Whether this degree of fidelity matters is another matter but I reckon it would sell for a lot of dosh to audiophiles. I will look out for the Geezer name in the HiFi world.
-
You get what you pay for, but I would expect a £20 amplifier with £200 speakers to sound a lot better than a £200 amp driving £20 speakers.
The general consensus is that, perhaps counter intuitively, the opposite is true. The audiophile philosophy is that you spend more money at the beginning of the audio chain, than at the end, on the basis that information lost earlier in the chain cannot be retrieved later, regardless of the quality of later links in the chain. Indeed, having a better quality item later in the chain can just make all the flaws in the earlier links all too apparent. With a £20 amplifier driving a £200 pair of loudspeakers you run the strong risk of just hearing the flaws in the £20 amplifier all too clearly.
Also, loudspeakers have moving parts and they wear out in a way that solid state electronics can't.
Well, it's certainly true that loudspeakers have moving parts, and that those parts will degrade over time. The question is how long will they take to degrade though? In any case, solid state electronics are not entirely immune to degradation over time either; dielectrics, from those in capacitors, to cable insulations, are certainly susceptible to degradation.
Incidentally, I have always felt the sound quality was better with headphones and I presumed this was because they didn't need to move much air so they could do it properly. Was I on the right track?
I think you probably are on the right track, but the amount of air being moved has effects back down the chain. Because the amplifier isn't having to work as hard to drive the headphones, compared with driving the loudspeakers, it will be operating within a narrower region, which being narrower, will probably be more linear and with less distortion. It will also effectively increase the dynamic range and reserve of the amplifier because it can still expand into the wider range normally used when driving loudspeakers.
In addition to the amplifier being able to reproduce the signal better, because the reduced power demands enable it to operate in a more linear range and with less distortion, it will also be able to control the headphone driver with more control and authority, resulting in higher fidelity.
One relatively recent, and interesting but not universal, trend in amplifier design has been to move away from over-sized and multi-strand internal wiring and cabling to using single solid-core wire, gauged according to the current it has to carry. The reasoning, if iirc, is that passing a signal down a multi-strand wire means that the signal takes multiple paths. Now not only are these paths all of slightly different lengths, but if copper wire is used then a small degree of oxide will inevitably build up on the surface of each strand, forming a dielectric between each one. The signal then, is being passed along multiple paths of different lengths, resulting in phase incoherency, but in addition, the signals passing down each strand will start to interfere with each other. Moving to single strand wire prevents the signal from interfering with itself, but if it is over-gauged there will still be phase issues because that proportion of the signal that propagates near the surface of the conductor has to travel slightly further than the part of the signal that propagates down the center of the conductor.
This might just all seem too extreme but then you need to remember that if you're using a moving coil cartridge on a turntable, and you're listening via moving coil loudspeakers, you'll be needing to amplify the signal from the cartridge somewhere in excess of 30,000 times if you want to play the signal reasonably loud (let's say requiring a 10V input to the speakers. A moving coil cartridge puts out a few hundred micro volts at most, let's say 300μV = 0.0003 V, and the standard reference voltage for 1 W sensitivity tests in to an 8Ω impedance is 2.83V. So, playing music reasonably loud on typical loudspeakers is likely to exceed 10V inputs).
-
I think you need to have very rigid diaphragms on the speakers and employ an optical feedback control system. Then, with a fast DSP it should be possible to accurately compensate for the Doppler generated IM distortion. Whether this degree of fidelity matters is another matter but I reckon it would sell for a lot of dosh to audiophiles. I will look out for the Geezer name in the HiFi world.
I was thinking of pre-processing and saving the digital recording to include the compensation for the IM distortion. I'm not sure a DSP would even be required. It might just be software that runs on a PC.
-
The distortion depends on the velocity of the speaker cone so preprocessing could only be done if the speaker design was precisely specified and consistent - not varying much over manufacturing tolerances.
-
The distortion depends on the velocity of the speaker cone so preprocessing could only be done if the speaker design was precisely specified and consistent - not varying much over manufacturing tolerances.
We could characterize each speaker by playing a series of test tones and capturing the otput with a microphone, or have the user specify when the IM was no longer discernable. That might actually be better, because the user would know that the system was working even if he could not detect the IM distortion while listening to music. It would also allow the user to fiddle with it.
-
I doubt that many would be good judge of audible distortion. To be cynical, it is more that people want to believe they have a perfect system - if you demonstrate they can't actually tell the difference it would not be good for business :-)
As a matter of interest, as I've got older I find loud music makes my ears distort the sound. I was at London South Bank yesterday listening to various choral groups. I reckon most sound engineers have damaged their hearing and are deaf. The sound was amplified ridiculously loudly in some cases, to the extent I had to move away from the huge, and unnecessary, speakers.
-
As a matter of interest, as I've got older I find loud music makes my ears distort the sound. I was at London South Bank yesterday listening to various choral groups. I reckon most sound engineers have damaged their hearing and are deaf. The sound was amplified ridiculously loudly in some cases, to the extent I had to move away from the huge, and unnecessary, speakers.
I don't like going to movies much either because the sound systems are far too loud.
What did you say? I can't hear you. The soundtrack's too loud!
I wonder why they are still called movies? Shouldn't they really be talkies?