Deceitful Electronics / General Chatting / Forums

Forums

4hv.org :: Forums :: General Chatting

« Previous topic | Next topic »

Deceitful Electronics

first previous 2 3 4 5 next

Move Thread

LAN_403

Dalus

Mon May 24 2010, 11:35AM

Registered Member #639 Joined: Wed Apr 11 2007, 09:09PM
Location: The Netherlands, Herkenbosch
Posts: 512

Strangely enough almost none of the audiophiles switch to plasma and electro static speakers that do make a difference. It's just all about the magic wood, gold and silver. And the idea of $$$$ == better sound.

Mads Barnkob

Mon May 24 2010, 11:45AM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

Dalus wrote ...

Strangely enough almost none of the audiophiles switch to plasma and electro static speakers that do make a difference. It's just all about the magic wood, gold and silver. And the idea of $$$$ == better sound.

I know a couple of audiophiles with electro static speakers, and believe me, they are just like all the other with their gold plated voodoo cables :)

Steve Conner

Mon May 24 2010, 01:31PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

The downside with electrostatic speakers is that you can't get much bass out of them. They're great for classical music, but a blast of Venetian Snares would turn them into so much Mylar confetti. You can gang them up with a subwoofer, but then they're not electrostatic any more, they're hybrids.

I'm sure Nicko has at least one pair of Quad ESLs and will comment here. ;)

BTW, does anyone remember the electrostatic headphones that came with their own high-voltage power pack? I wonder if they made your hair stand on end.

Nicko

Mon May 24 2010, 02:11PM

Registered Member #1334 Joined: Tue Feb 19 2008, 04:37PM
Location: Nr. London, UK
Posts: 615

Steve McConner wrote ...

I'm sure Nicko has at least one pair of Quad ESLs and will comment here. ;)

BTW, does anyone remember the electrostatic headphones that came with their own high-voltage power pack? I wonder if they made your hair stand on end.

Nah. No ESLs - I never was a fan.

However, as you pointedly made the comment about electrostatic headphones, you had obviously noticed in that photo are a pair of electrostatic Beyer ET1000s which I renovated last year - I actually bought them new back in about 1978 or so from Audio-T, who, amazingly, are still in business

. The ET-1000s have been in the loft for 25 years, luckily well stored, so really in pretty good condition. Originally, I ran them from a Sansui AU-717 (lovely amp), but in the photo they are being driven off my Perreaux SM2 & 2150B pre/power amp combo - almost contemporaries of the Beyers... I was going to auction them all, but having listened to them again, I've fallen in love (!) so they will stay for a while...

So, close, but no banana!

The Beyers still sound great, and yes, they have their own HV supply - the little black box they are plugged into... as to making your hair stand on end - see my avatar...

Cheers

radiotech

Mon May 24 2010, 08:39PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

If you had a passive column designed for 500 watts with a sensitivity
of about 86 dB/Watt at one meter and an honest response curve from 40-16,000 Hz I think the cable should address skin effect or there would some issues at the high end too for some music. A good test would be an A-B of equal conductivity lengths of litzwire and copper wire setting Rdc to Zspeaker/20 as a starting point.
That would be 12meters of #14 twinlead for 4 ohm speakers.
We would be trying to see if 12 amps at ~~10-16kHz needs litz.

IntraWinding

Mon May 24 2010, 10:02PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

Steve McConner wrote ...

Nothing wrong with listening to music and drinking lots of wine.
And, I think speakers are one part of a hi-fi system that it's worth spending lots of money on. But I'm not sure about the cables.

Radiotech's point about cable resistance and damping factor is a valid one, but the cables don't need to be expensive to address that, they just need to be beefy.

I agree. Speakers are the weak link in the chain.

On the other hand amps aren't as good as the figures suggest: If you put a scope across your amp, play music and slowly raise the volume, it's surprising how soon the audio peaks clip. Even with 100 Watts/channel amp you can only achieve a rather sedate volume before the odd clip of a peak sneaks in, and that's just when it's bad enough to notice on the scope. If you build in a clip detector circuit you'll find you need enormously powerful amplifiers to play music at a respectably loud volume with less than, say, one clip per second. How much of that clipping is required for it to become audible have no idea, but it makes a bit of a joke out of all the other efforts made to keep distortion to a minimum. These peaks don't necessarily occur at perceived loud points by the way, they occur when lots of frequency components happen to be near enough in phase to add up to some value well above the average. It's a statistical thing and happens randomly, but more often in the loud bits.

Steve Conner

Tue May 25 2010, 09:44AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well, I guess that depends on the kind of music you play. Everything except classical music is so heavily compressed now, that the peak-to-RMS ratio is very small. The peak amplitudes are deliberately limited, so they can squeeze more volume into the peak amplitude limitation of CD.

If you have a decent system that is actually capable of a high dynamic range, this gets annoying after a while. Record producers call it the "loudness war". So there's another electronic con: when you buy a CD, you pay for 16 bits, but you only get about 5!

I built an amp of about 130 watts/channel with a clip detector circuit, and it could get painfully loud before any clipping was detected. Since I moved house, I can't get it loud enough to clip any more! Some of you may have seen it before:

I've also got a 30 watt per channel valve amp, and that does start to sound obviously dirty when you ask it to reproduce heavy bass at high volumes.

IntraWinding

Tue May 25 2010, 10:35PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

Did your clip detector produce a visible signal for even a single momentary excursion to one of the rails? In other words did a clipping event that lasted a fraction of a millisecond do something like turn an LED on for a 100 milliseconds? If so Iâ€™m surprised you didnâ€™t find the same effect I described.
But thinking back I might have been using a record player as a signal source, which could be the explanation. With vinyl, the signal from the pickup is passed through an RIAA correction filter and perhaps this reintroduced the large spikes I saw even if they donâ€™t appear physically in the vinyl by way of reshuffling the relative phase of the different frequency components of the audio signal.

Iâ€™ve noticed this effect in the excellent audio editing software Cool Edit Pro. A CD has its signal carefully compressed to fit in the available number of bits whilst playing back at reasonable volume, but after you've done a bit of work on a track, letâ€™s say filtering out a bit of mains hum and hiss, for example, suddenly you'll notice these large peak have appeared which I think are a result of the various frequency dependant phase shifts caused by the filters occasionally bringing various frequency element in phase at just the right (wrong?) moment for their voltages to add up to something well above the average. If I now want to cram the signal back into 16 bits with a similar playback volume to the original without clipping these peaks I have to use some technique to bring them back down closer to the average as inaudibly as possibly, such as using a compressor. In practice you can get away with just clipping and filtering at 20KHz, if there arenâ€™t too many. What this tells me is that they almost certainly had to use a similar technique when they mastered the CD, so they've already removed the peaks, which may explain why you didn't see them if you where using CDâ€™s exclusively.

Frosty90

Wed May 26 2010, 01:17PM

Registered Member #1617 Joined: Fri Aug 01 2008, 07:31AM
Location: Adelaide, South Australia
Posts: 139

Lots of you have probably seen this one before:

but to me it takes the cake.

GeordieBoy

Wed May 26 2010, 01:38PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

IntraWinding: CD mastering and in particular broadcast audio processing have both become horribly complex and quite extreme in the things they do to the audio spectrum.

A typically processing chain might consist of something like this:

1. Filter out an sub-bass or excessive HF content that is out of the audible range anyway. (This just wastes valuable dynamic range!)

2. Rotate the phase of the low-frequency components using a cascade of allpass filters. This is said to correct for the typically assymmetric waveform of the human voice. (If you mess with the phases of the components to modify the waveform to be symmetrical, you can compress it much harder later to maximise the volume before both sides clip equally badly!)

3. Split the audio up into somewhere between 3 and 10 seperate frequency bands.

4. Apply slow acting AGC to each of these frequency bands so that all music source material is forced to a consistent tonal signature. (This is what gives "Rock FM" its signature sound regardless of what they actually play!) Think of it as a dynamic graphic equaliser.

5. Apply fast acting compression to all bands to raise the average power level in each slice of the audio spectrum relative to the peak amplitudes.

6. Clip off the peaks. Yep, that's right, just clip them right off in each band! Sure, it will cause distortion but clipping off the top 10% of the signal allows you to turn up the volume another 10% futher and that's worth the added distortion. The multi-band processing also makes the distortion not as bad as it could be because the inter-modulation products are confined to each frequency band instead of spreading right across the audio spectrum.

7. Combine all of the bands together

8. Process the combined audio with a look-ahead limiter or clipper to limit the peak modulation depth or signal amplitude if recording to CD.

It's all quite complicated. As you said, even filtering something can alter it's peak amplitude in a way that you might not expect. For instance if you consider a square wave, it's fundamental component is actually of greater peak-to-peak amplitude than that of the resulting square wave with all of it's harmonics present. So, if you filter a square wave to remove all of the components except the fundamental, then the peak-to-peak amplitude will actually increase by something like 17% if I remember correctly. For me that result is not intuitive!

Years ago all of this processing used to be done by analogue electronics. These days it's done by DSPs in boxes like TC electronics's Finalizer or Orban's Optimod series, or even as a software plug-in running on your PC!

My person feeling is that however the processing is achieved, it is still to the detriment of sound quality. But like Steve C said it's a loudness war and until one record company or radio station backs down we have to put up with everything sounding like it's compressed to mush!!!

-Richie,

first previous 2 3 4 5 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob