The Audiophoolery Thread.

[stevehertz]

Originally Posted by julie_m

So cables that were "used to", say, jazz, would not sound as good when the system was being used for listening to, say, folk?

That hypothesis would be worth testing, but for the inevitable accusations of insufficiently-sensitive measuring instruments .....

Oh we have such sensitive instruments, they're called ears, and in some respects, contrary to the view of us 'techies', audiophools trust their ears 100%. That's even when the difference either doesn't exist, or is demonstrably so miniscule as to be unmeasurable, as it is in so many cases. No, the problem that audiophools have is that they are either unaware of, or choose to ignore the fact that what they are listening to is actually not just an aural event, but a psychoacoustic one. And that changes everything. The reason? because it is dead easy to convince ourselves (for lots of reasons but one of them being that you just dished out thousands of pounds on this new piece of hifi) that the 'upgrade' that you have just made sounds better. In short, you desperately want, and really believe that this new piece of kit will make your system sound better, so your brain says, there you go, it is better! Seller and audiophool are happy! This satisfying deduction is arrived at on the basis that in truth, your ears have not been able to detect any difference. But you knew the difference would be subtle anyway (didn't you?!), so it all makes sense..

Try the spoken word through most "posh" hifis' it sounds awful. Nuff said.

[Valvepower]

Hello,

Well, with the risk of being drummed off the forum…

Tomorrow morning I’ll be heading to a hi-fi manufacturing factory on the South-East Coast. The company employs over 100 people and in lot of ways this is a very traditional manufacturing environment with the following departments/disciplines:

Tool makers
Electronic and mechanical engineers
Mechanical assemblers
Electronic assemblers
Test engineers fault finding to component level
Service department who will be willing to repair a 40-year-old unit if they can do so
Wire men and ladies
Coil winders
Use local and UK suppliers where they can
It was only up until recently they had a tea lady

And the cable, oh sorry interconnect department, which has gold door knobs, platinum flooring, diamond encrusted chandeliers and silk clad gurus being feed caviar by nubile hand maidens all in harmony with the local lay lines. These gurus are regularly seen enticing electrons the wrong way through cables

But seriously folks I’ll be the first to admit the hi-fi industry hasn’t done itself any favours (the words skeleton and cupboard jump to mind), but isn’t every industry guilty – to some extent – of creative marketing to promote their products!

Anyway, I’m proud to be a part of this company as its basically it’s good ole' engineering and manufacturing with its associated fun games and characters and, thankfully there is not an audiophool in sight, well not in a quarter of mile or so if the 'phool' police have their way

Regards
Terry
Best get my coat….

[kalee20]

Originally Posted by merlinmaxwell

They are rated at a maximum (current, minimum load resistance) and don't care in the slightest if they are not loaded at all, a few k ohms might (very very very much might if the following bit of kit is DC coupled to the input and has a huge input resistance) be useful. And do you use 30 (or is that 33) ohm cable?

Exactly, MM! So Refugee... why load your outputs? It's like saying, the mains cable to all the lights in my house is rated at 3A, therefore I must fit a 750W bulb in every socket!

[mhennessy]

Good audio-quality op-amps show an increase in distortion when asked to drive low impedance loads. As a rule of thumb, 2k is a good minimum load impedance (including the negative feedback resistors, of course). Some, like the good old NE5532, are good to perhaps 600-ish ohms, perhaps.

Meanwhile, the output impedance of a "hi-fi" CD player - set by an output resistor - will be anywhere between 50 and 500 ohms, give or take. Just depends on the whim of the designer - there's no formal standards for domestic hi-fi.

So slapping a 30 ohm load at the output of a hi-fi CD player will not only reduce the output voltage (by an amount that depends on Zout, and so will vary from machine to machine), but will increase harmonic distortion. So, a bad idea every way you look at it (if you actively want less level, use a potential divider with higher value resistors that add up to 2-4k). Matched impedance audio circuits went out decades ago. Just try finding 600 ohms in a modern professional installation - if you do, it'll be for legacy gear like vintage valve preamps or similar...

The output stage of a portable device is a bit different, however. The op-amps in these, working from very low supply rails, will be designed to drive 32 ohm headphones (the actual impedance will obviously vary with frequency, so 32 ohms can only be very nominal, just like it is with loudspeakers). These ICs tend to have pretty high distortion values compared to the usual audio op-amps, but for the application, they're usually good enough.

Some output stages can be upset by not seeing a low-ish load. It's hard to investigate exactly what's going on because such devices tend to be hard to take apart, and once you have, you can't see any of the components in there anyway! Perhaps they need a DC path to charge the output capacitors - who knows? Either way, I've found that a load of a k or 2 is enough - 30 ohms would just take current from the battery needlessly.

Needless to say, there are plenty of other areas where matched impedance is useful/essential. But not audio. Ever seen a power amp with 8 ohms Zout?

[julie_m]

Originally Posted by merlinmaxwell

Try the spoken word through most "posh" hifis' it sounds awful. Nuff said.

Indeed.

And if you can't tell Sheffield from Donny, or Wolverhampton from Dudley, or Stoke from Stafford, or pick any two towns near where you're from, then it might not be that good.

[Craig Sawyers]

Originally Posted by Voxophone

This one is actually true. Spending 48 hours at such a low temperature is ‘character building’ for the valves (similar to DofE or one of those American military schools). They then go on to be well-rounded/superior sounding members of society/your hi-fi system.

Non-cryo-treated valves lack perspective and never realise how lucky they are, whereas those which have been to the North Pole are always thankful for the warmth of their filaments.

Very interesting thread by the way.

Liam

Well, I'm back. I cannot speak on the topic of cryo-treating valves, but I can speak as the ex-Technical Director of Oxford Instruments. We used to build huge and complex superconducting magnets (and other stuff) for research establishments worldwide.

The problem was that if you machined a magnet former (out of a 316LN steel billet), assembled the magnet to the former and then cooled it down to liquid helium temperature (4.2 Kelvin) it did not work - it would never get to the design magnetic field. Disassembly (a week long process) showed that the former has distorted. Every time a magnet did not work - distorted fomer. A fraction of a mm was enough.

We overcame the problem by rough machining the former, and then cycling it in a bucket (it was a big vacuum insulated bucket!) cryostat repeatedly to 77K (liquid nitrogen), and then final machining to dimension.

The background is that cycling to liquid nitrogen results in stress relieving. So at least the plausibility argument is that doing the same thing with a valve changes in some way its characteristics, at least audibly.

Craig

[Craig Sawyers]

Originally Posted by G8HQP Dave

Not true. A coax cable has inductance but near zero pickup. A twisted pair has inductance but near zero pickup. A woven cable, or other poor designs, will pick up RF and for these cables the amount of pickup may vary with inductance.

I was not talking about coax cables, which I hope was clear from my posts. I was talking about loudspeaker cables, which are rarely (if ever) of coaxial construction. For an unscreened cable my argument most certainly holds

Craig

[Craig Sawyers]

Originally Posted by stuarth

I once read a review of a CD player in what I thought was a respectable magazine which made the claim that this particular player had cleaner bass because in had a die cast chassis....
Stuart

You have to be a bit careful here. The tracking requirements for a Red Book CD are quite testing. The laser focus has to be held to less than 1um and 1um in radial tracking with respect to the recorded data. The 1um focus has to be held even with 1.2mm warp. Given the rapid movement of the lens and tracking system, and the associated inertia, a rigid frame to hold the parts together is essential to prevent structural resonances.

OK - I hear you say "This is digital - it is 0's and 1's". Well it might be at one point, but I'd invite you to look at an eye diagram for the analogue output from the photodiodes in a CD mechanism. A CD mechanism is analogue through and through. Focus, tracking and data.

Craig

[Craig Sawyers]

Originally Posted by Radio Wrangler

Were this not the case, pre-burnt-in cables would be available (great value-added opportunity for suppliers), but are in fact conspicuously absent from the market.

Well - just not true.

Just as an example, Russ Andrews offers cryogenic treatment for any of his interconnect and speaker cables for an additional £12.50, regardless of cable cost. Or burn-in for £15.

Craig

[Craig Sawyers]

Originally Posted by stevehertz

Oh we have such sensitive instruments, they're called ears, and in some respects, contrary to the view of us 'techies', audiophools trust their ears 100%. That's even when the difference either doesn't exist, or is demonstrably so miniscule as to be unmeasurable, as it is in so many cases. No, the problem that audiophools have is that they are either unaware of, or choose to ignore the fact that what they are listening to is actually not just an aural event, but a psychoacoustic one. And that changes everything. The reason? because it is dead easy to convince ourselves (for lots of reasons but one of them being that you just dished out thousands of pounds on this new piece of hifi) that the 'upgrade' that you have just made sounds better. In short, you desperately want, and really believe that this new piece of kit will make your system sound better, so your brain says, there you go, it is better! Seller and audiophool are happy! This satisfying deduction is arrived at on the basis that in truth, your ears have not been able to detect any difference. But you knew the difference would be subtle anyway (didn't you?!), so it all makes sense..

You have to be a trifle careful regarding human hearing. I suggest you scan this http://physicsworld.com/cws/article/...ghly-nonlinear

And the paper from Physical Review Letters on which that is based https://journals.aps.org/prl/abstrac...ett.110.044301

Quote "We study human ability to simultaneously judge the frequency and the timing of a sound. Our subjects often exceeded the uncertainty limit, sometimes by more than tenfold, mostly through remarkable timing acuity. Our results establish a lower bound for the nonlinearity and complexity of the algorithms employed by our brains in parsing transient sounds, rule out simple “linear filter” models of early auditory processing, and highlight timing acuity as a central feature in auditory object processing."

Craig

[Argus25]

Originally Posted by GrimJosef

This particular example of what we might call audiophoolery does involve negative consequences for some people. They have to be weighed against the positive consequences for other people. If the question of 'right or wrong ?' is difficult when there's no conflict of interest then it becomes even more difficult when there is one. You'd need the wisdom of Solomon to establish whose enjoyment of a rare valve is 'worth' the most ...

Cheers,

GJ

I'd have to disagree with that, turning the right-wrong question about "truth or dishonesty" into an ethical dilemma is merely a smoke screen to allow some people, with the positive consequences you refer to, to benefit and avoid facing the consequences of dishonesty. It doesn't take the wisdom of Solomon to figure this out, Sheldon on the Big Bang Theory did (or at least the script writers)..he said, : "wrong is an absolute state and its not subject to gradations".
Of course some people in the industry who have been successfully coached to believe the lies about things like component sonics, think that what they are saying is entirely true and its hard to argue they are doing anything wrong, but for those who know, its shameful.

[GrimJosef]

Originally Posted by Craig Sawyers

... The background is that cycling to liquid nitrogen results in stress relieving. So at least the plausibility argument is that doing the same thing with a valve changes in some way its characteristics, at least audibly.

Temperature cycling can indeed relieve stress, but it works just as well if the parts are heated and then cooled. It's just a shame that there's no way for the internal metalwork in a valve to be heated up and cooled down a few times.

Oh ... hang on ... .

Seriously though, during the pumpout procedure for almost all valves the metalwork has to be heated much hotter than the normal operating temperature of most of it to get it to release any trapped gases. By the time it leaves the factory it will have experienced much greater temperature excursions than the 200-odd degrees that an LN2 bath will give it.

I know what you mean about 316 stainless though. When I was a graduate student I used to have TEA laser electrodes machined from it and we went through the same process of rough cutting followed by annealing followed by closer working followed by more annealing followed by even closer working etc etc.

Cheers,

GJ

[GrimJosef]

Originally Posted by Argus25

I'd have to disagree with that, turning the right-wrong question about "truth or dishonesty" into an ethical dilemma is merely a smoke screen to allow some people, with the positive consequences you refer to, to benefit and avoid facing the consequences of dishonesty. It doesn't take the wisdom of Solomon to figure this out, Sheldon on the Big Bang Theory did (or at least the script writers)..he said, : "wrong is an absolute state and its not subject to gradations".
Of course some people in the industry who have been successfully coached to believe the lies about things like component sonics, think that what they are saying is entirely true and its hard to argue they are doing anything wrong, but for those who know, its shameful.

We'll have to leave it there then. You're clearly confident in that position.

Cheers,

GJ

[Refugee]

Originally Posted by merlinmaxwell

They are rated at a maximum (current, minimum load resistance) and don't care in the slightest if they are not loaded at all, a few k ohms might (very very very much might if the following bit of kit is DC coupled to the input and has a huge input resistance) be useful. And do you use 30 (or is that 33) ohm cable?

Some may have noticed I have not included the output impedance of the unit which, in most cases, is near zero not matching any cable.

Using the lower value is done to load out mobile phone noise.
The amplifiers have heavy metal cases and are out of reach of mobile phone noise.
There is a short stub of two core from the jack plug and then two bits of 50 ohm.
The total length is about 3 to 4 feet.

The source is designed to take 30 ohm headphones so loading it to the designed resistance is the correct thing to do surely.

[PJL]

Is the conclusion that audio reproduction has reached perfection and technological advances can do nothing to improve the listening experience? Or is it that audiophiles are a dying breed as people only listen to music when they are texting or looking at facebook?

My hi-fi can still get a bit of an oooh from my daughters friends (30) especially when I tell them that I was about 5 when most of the equipment was made.

[Refugee]

I have found a web link that I could not resist.
The sound of solder.
http://www.diyaudio.com/forums/parts...er-alloys.html

[Craig Sawyers]

Originally Posted by Refugee

I have found a web link that I could not resist.
The sound of solder.
http://www.diyaudio.com/forums/parts...er-alloys.html

The problem is that there are truths lurking under all the pseudo science. Sure thermoelectric effects are DC in nature, and at face value have no impact on an AC signal, such as audio. And indeed for domestic audio that is really pretty much the case.

But if you were designing something using a 20 bit A-D (they are now available to 24-bit) thermoelectric effects are very real and can easily mask the lower bits with DC offset. See for example the attached.

Craig

[Craig Sawyers]

Originally Posted by GrimJosef

I know what you mean about 316 stainless though. When I was a graduate student I used to have TEA laser electrodes machined from it and we went through the same process of rough cutting followed by annealing followed by closer working followed by more annealing followed by even closer working etc etc.

Cheers,

GJ

Where were you working on high power CO2 lasers? A Culham studenship?

Craig

PS for those not initiated in laser technology TEA stands for Transversely Excited Atmospheric (pressure).

[kalee20]

Originally Posted by Refugee

The source is designed to take 30 ohm headphones so loading it to the designed resistance is the correct thing to do surely.

Why?

The output stage will almost certainly use complementary NPN/PNP emitter followers. So loading with less than 30ohms would overly stress them. But loading with HIGHER will give them an easier life - less heating; less thermally-induced distortion, everything happier!

A DC path might be necessary, but probably isn't.

Have you tried switching out your 33ohm loads and compared results?