UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > General Vintage Technology Discussions

Notices

General Vintage Technology Discussions For general discussions about vintage radio and other vintage electronics etc.

Reply
 
Thread Tools
Old 7th Apr 2019, 1:52 pm   #661
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Audiophoolery?

The persuasive nature of lots of audio, for the want of a better word, snake oil is a real shame. Much of the difference between cables has been known for decades, and is down to:

1. In unbalanced cables, where any unwanted current in the shield simply adds to the signal through ohms law, the use of a heavy braid shield with low resistance is a benefit, as is low contact resistance connectors.

2. Anything to reduce triboelectric noise due to movement of the cable is a good thing (and also for balanced too). Usually this will involve both a conductive polymer inner shield (which is not electrically connected) and a braid outer. It might include fibre fillers or other structures, all with the aim of stopping a cable rustling when moved.

3. With unbalanced however getting a hum and other noise-free connection is a bit of a lottery, particularly if you have a whole lot of bits of kit connected to a preamp, or anything involving a home cinema.

4. If possible go balanced, even with short links. Any hum currents flow in the shield, and appear as a common mode signal in the balanced inner wires - neatly subtracted by the balanced input stage. Depending on the precise design of the input stage, common mode rejection will be in the range 50-90dB, which is a measure of how much more immune a balanced connection to hum etc as compared to an unbalanced connection.

5. All is not entirely sweetness and light with balanced. Anything that breaks the symmetry in the cable causes something called Shield Current Induced Noise (SCIN), in which any shield currents can induce a differential voltage, which is then amplified. The worst construction for this is a foil shield with drain wire, which is inherently asymmetric. High coverage braid is by far the best construction, and for short links is usually not a problem.

However, all these are ancient history in cable construction. Which is why all my signal cables are this stuff https://www.canford.co.uk/Index/Cabl...BLE-Deployable

at £1.73 per metre, plus vat and delivery. The XLR connectors are actually the dominant cost in constructing a finished cable!

Craig
Craig Sawyers is offline   Reply With Quote
Old 7th Apr 2019, 1:57 pm   #662
fetteler
Octode
 
fetteler's Avatar
 
Join Date: Jul 2011
Location: Staffordshire Moorlands, UK.
Posts: 1,464
Default Re: Audiophoolery?

Quote:
Originally Posted by PJL View Post
Audiophiles suffer from a form of OCD perfectionism, as many do to some extent, and the audio market is simply exploiting their obsession. Like gamblers, audiophiles will not all be wealthy. It is down to human behaviour and there is very little logic in that!
Yes, nicely put. It is a feature of their psychological makeup which is being exploited.

I'm not sure that it's morally defensible to take such advantage of human foibles.


Cheers,
Steve.
__________________
Those who lack imagination cannot imagine what is lacking...
fetteler is offline   Reply With Quote
Old 7th Apr 2019, 3:10 pm   #663
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Audiophoolery?

My experience of at least the communal bit of the audio community is that there is just as wide a spread of relationships there as there is in the vintage electronics community. As here, that includes a good deal of camaraderie, generosity and helpfulness. Favours are done for people whose circumstances are tight - equipment is loaned so they can find out for free whether it really makes a difference. Sometimes it's sold at a knock-down price. Sometimes it's even donated. Anyone trying to take advantage gets short shrift.

There is also a great deal of exchanging of second hand kit - I once, briefly, owned a pair of speakers which had had half a dozen different prior owners on the same forum. I sold them on, for the customary 'no more than I paid for them' to another member. I've worked on an amp which has been owned by nearly as many people and is simply known as 'Vanessa'. On balance I'm sure the rewards outweigh any grief and I'm convinced that the people having the hardest time now are the dealers, unless they can export to the Pacific rim where the real demand now is.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline   Reply With Quote
Old 7th Apr 2019, 3:36 pm   #664
fetteler
Octode
 
fetteler's Avatar
 
Join Date: Jul 2011
Location: Staffordshire Moorlands, UK.
Posts: 1,464
Default Re: Audiophoolery?

Let me be clear, my comment is in regard to those in business who profit hugely from selling audio snake oil - communities of enthusiasts, like any other hobyists usually bring out some of the best and nicest aspects of folk

Cheers,
Steve.
__________________
Those who lack imagination cannot imagine what is lacking...
fetteler is offline   Reply With Quote
Old 11th Apr 2019, 7:13 pm   #665
stevehertz
Dekatron
 
stevehertz's Avatar
 
Join Date: Sep 2011
Location: Rugeley, Staffordshire, UK.
Posts: 8,809
Default Re: Audiophoolery?

Anyone want a pair of phono leads? Black Cat brand, 1m long, just £140 second hand? I just spotted some for sale on another forum. Sounds an absolute bargain to me..
__________________
A digital radio is the latest thing, but a vintage wireless is forever..
stevehertz is offline   Reply With Quote
Old 11th Apr 2019, 8:16 pm   #666
Silvered_Mica
Pentode
 
Silvered_Mica's Avatar
 
Join Date: Aug 2003
Location: Halesowen, West Midlands, UK
Posts: 125
Default Re: Audiophoolery?

I have also been looking at audio forums and something has stood out, the technical folk just use ordinary leads but the less technical folk
use over priced leads or wish they could.

And the audiophool's have a term for us lot and that is measurementalists, apparently we should be using our ears not our meters.
I prefer to be a measurementalist than a pretentious
__________________
Trust Me, It's a Magic Iron (∩ ͡° ͜ʖ ͡°)⊃━☆゚. * ・ 。゚, 📻📺🦔📡
Silvered_Mica is offline   Reply With Quote
Old 11th Apr 2019, 8:32 pm   #667
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Audiophoolery?

Sometimes, when wanting to know how much electricity there is on something, sticking your ear up to it can generate a surprising amount of regret....

The audiophools don't seem to understand that owning a variety of measuring equipment does not disable your hearing. Some of us scope-owning folk have quite normal hearing.

It isn't their ears that we lack, it is their beliefs.

I understand that belief without support of evidence is one definition of a religion. I also understand that recognised churches get tax advantages. If only the audiophiles got themselves registered as a religion they'd be able to save a lot of money, considering the prices of their peccadilloes!

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 12th Apr 2019, 5:45 am   #668
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Audiophoolery?

There is often a reaction against using op-amps rather than discrete circuitry in audio gear, with adverts extolling the virtues of particular design philosophies. That preception tends to be rooted in the past when decent op-amps were just not available.

But two of my go-to op amps for audio design are the LM4562 and the AD797. Because distortion is so low, and beyond the direct measurement limit of any test gear, for the LM4562 they had to devise novel methods for measurement.

Because there are so many zeros involved, it is better in my view to quote parts per million (1ppm = 0.0001%)

For the LM4562 (a dual amp) THD is 0.3ppm, IMD is 0.07ppm. Into 600 ohms load. Noise is 2.7nV/rootHz and slew rate 20V/us. 55MHz Gain bandwidth product.

For the AD797 (a single amp) THD 1ppm (limited by test gear), noise 0.9nV/rootHz (noise resistance of 50 ohms!), slew rate 20V/us

With devices like that in your design armoury, why on earth would you design using discrete devices?

Well I guess if you were designing moving coil input stages a noise resistance of 50 ohms for the LM797 is a little high, so you might precede that with a multiple device input stage to get noise resistance down into the 3-5 ohm level, and that is where some of the design artistry occurs.

Craig
Craig Sawyers is offline   Reply With Quote
Old 12th Apr 2019, 6:50 am   #669
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Audiophoolery?

When you have such good basic gain blocks, you don't want the circuitry around them to spoil things. So the design of PCBs to keep out hum and sundry noises down to a matching level becomes a bit artistic.

Dynamic mike inputs for studio use and moving coil cartridge inputs are the places where the need for optimum noise performance still dictates discrete design. Large width semiconductors come to the fore. There were some rather good individual devices, or multiple more mundane devices can be used to synthesise the behaviour of a single wider device. The problem of availability enters the fray.

The obvious limitation of these amplifiers is the power output, but they could be very beneficial in the input stage of a power amp, offering much improved gain, linearity, pole frequency, common mode and power supply rejection. They would allow the subsequent VAS to be made lower gain, and this could trade into lower stage distortion.

The snag in power amps is that opamps are normally found within 15v of ground, while VAS stages live out by the supply rails. Marrying the two, or finding an alternative architecture can be fun. The trad discrete pair actually fits in rather well, and parallelled-input PNP and NPN stages are a very neat fit.

I wonder about twin opamp input stages for a power amp. One running from supplies up near the positive rail, one down near the negative rail, configured as diff amps with carefully matched resistors to attenuate what they would see as extreme common mode input voltage. There would be noise, offset and gain disadvantages, but the opamps start with advantages in those areas. Might still be a good trade-off.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 12th Apr 2019, 7:39 am   #670
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Audiophoolery?

Absolutely to all of that David.

The VAS is the most highly stressed block of any in a power amp. It not only has to swing between the rails, but also supply the base drive current to the output stage.

But using an opamp as an input stage is not an unreasonable idea. Quad pioneered that in the 405 - in early amps the LM301A and later the TL071.

But an opamp could be combined with a push-pull output stage running close to the rails, with overall feedback. That is commonly used to drive between rails in the hundred volt range, and should work fine in a power amp.

The lowest noise discrete devices are pnp. However the semiconductor industry seems to be hell bent on discontinuing manufacture of low noise pnp devices, which are lower noise and lower 1/f knee that npn. First to go was the superb Rohm 2SB737 with an rbb' of 2 ohms and 0.4nV/rootHz. More recently the pnp dual (with 0.7nV/rootHz) SSM2220. That is by no means the only ones. At least Linear Systems have remanufactured the long discontinued Toshiba 2SK170/2SJ74 low noise JFETS as the LSK170/LSJ74 (0.9nV/rootHz)

Craig
Craig Sawyers is offline   Reply With Quote
Old 13th Apr 2019, 12:39 am   #671
Glowing Bits!
Octode
 
Glowing Bits!'s Avatar
 
Join Date: Oct 2011
Location: Wrexham, North Wales, UK.
Posts: 1,457
Default Re: Audiophoolery?

Quote:
Originally Posted by stevehertz View Post
Anyone want a pair of phono leads? Black Cat brand, 1m long, just £140 second hand? I just spotted some for sale on another forum. Sounds an absolute bargain to me..
£140? Since when?
A perfectly good pair could be had for a fiver in Woolies in the past.
I can't tell the difference between something expensive versus cheap, it's just a wire, minus the snake oil!

Craig, all your figures in the above post mean sweet FA to me, it may do to someone who understands it, I just listen to music, not calculate it.

Rick, the annoying object roaming the forum.

One day, I'll set that as my signature!
Glowing Bits! is offline   Reply With Quote
Old 13th Apr 2019, 6:51 am   #672
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Audiophoolery?

No, there is a perfectly valid point in Craig's post.

There are circumstances where different ways of connecting things DO make a difference. Engineering and maths handle it correctly. Holy water, magic crystals do nothing.

There have been some hilarious incidents with 'extreme' cables:

The designer/owner of one upmarket expensive amplifier firm decided that the small network usually found on the output of a power amplifier in order to stop the amplifier seeing cable capacitance and losing stability margin, was hurting the sound. So he never included them.

Along came a fashion for expensive speaker cables. Some people theorised that as speakers were 8 Ohms-ish then designing a cable with a characteristic impedance ought to be good. Their cables interwove lots of isolated strands, half carrying the signal, half carrying the return. Up went the cable capacitance! Those posh amplifiers started smoking!

Instrumetalist's instruments would show the problem as oscillation at low RF frequencies, an important clue for what to do about it.

Subjectivists would diagnose the problem as a burning smell, then a lack of music. The subjectivist would then hit the magazines to find what he needed to upgrade so his next amplifier wouldn't go the same way. 'mains cable' would not be a good answer. amp or speaker cable might be better.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 13th Apr 2019, 8:45 am   #673
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Audiophoolery?

Quote:
Originally Posted by Glowing Bits! View Post
... Craig, all your figures in the above post mean sweet FA to me, it may do to someone who understands it, I just listen to music, not calculate it ...
The fact is that Craig's calculating the equipment, not the music. The aim is to make sure that when you're trying to listen to the music the kit isn't singing along in the background (actually humming and hissing and singing out of tune).

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline   Reply With Quote
Old 13th Apr 2019, 11:09 pm   #674
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Audiophoolery?

Quote:
Originally Posted by Radio Wrangler View Post
Their cables interwove lots of isolated strands, half carrying the signal, half carrying the return. Up went the cable capacitance! Those posh amplifiers started smoking!

David
Kimber. I've actually done quite a few measurements on these, both in my own lab and also at an EMC test lab.

Yes they are high capacitance, but also have a tiny inductance. And yes the characteristic impedance is comparable to a loudspeaker impedance. But that is accidental coincidence and is not the point. The main benefit is that they are very immune to conducted and radiated RFI as compared to a kettle lead construction. In a sense that is pretty logical, because the very low inductance implies the immunity. And that is important because any RFI that gets into a power amp's feedback network by whatever means can cause all sorts of horrors.

I actually use them for a different very good reason. I use an active crossover and a bunch of power amps. So I have five drive unit connections on each side. Because the inductance of each cable is low, you can run five woven cables right next to each other in the same techflex sleeve without any interaction (crosstalk between drivers). The low inductance of each cable means that the mutual inductance between adjacent cables is low.

The power amps are Doug Self's Blameless design, and they cheerfully drive the high capacitance of the woven cables. But they very sensibly have an inductor and Zobel on their outputs.

Quad also got that quite right when they stated that their amps were "unconditionally stable". They were of course interested in whether their amps would drive the horrid load impedance of an electrostatic speaker, but I know that a Quad 303 will drive 5 metres of woven cable without issues (to the ceiling speakers in out kitchen).

Anyone bonkers enough to manufacture an amp without inductor and Zobel deserves all the magic smoke and warranty repairs they get.

The usual giveaway is when a power amp manufacturer says they will only work with their own cables. I can think of several, past and present, but think it might be prudent not to mention their names

Craig
Craig Sawyers is offline   Reply With Quote
Old 14th Apr 2019, 3:24 am   #675
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Audiophoolery?

The interwoven cable construction gives very low subtended area for a magnetic field component to induce voltage in the connection loop, which in general is a good thing. It also gives very high shunt capacitance, and what is particularly nasty is that that capacitance is low loss even at high RF frequencies. This is what would drive an unprotected amplifier unstable.

Otto Zobel was a filter designer at Bell Labs, heavily involved in the "Image Parameter" style of filter design maths (lots of things owe a lot to this man). In the sixties the"Modern Filter Design" style evolved from image-parameter, and the big name in this was Anatol Zverev. However, image parameter filter methods still survive because they work very well for diplexers... like a sort of RFy crossover design.

It is good practice to include an RF diplexer on the output of a power amplifier, so that by the sort of frequency that the amp would oscillate at if capacitively loaded, it doesn't see the applied load at all, instead it is made to see a nice resistor as a dummy load.

To do this a small inductance (small in Henries, but thick in wire for low loss at audio) is fitted between the amplifier proper and its output terminals. This would naturally resonate with the cable capacitance and cause worse trouble. Planting a low-ish value resistor across this inductor kills the Q of the resonance. Its value is a compromise. Too high leaves too much Q in the resonance, too low spoils the isolation from the cable capacitance. To make things easier, a dummy load resistor is connected to the output of the amplifier proper, via a small decoupling capacitor. The small capacitor is a good connection at RF, but prevents the resistor being hit with the full AF signal.

All well and good? These networks should have their values properly calculated to suit the characteristics of an amplifier for optimum effect, but most usually other amp designs values are copied. They usually aren't too far off.

Now, the diplexer consists of the series L and the shunt C. The R in series with the C is the dummy load and the R in parallel with the L is to spoil the Q of the inevitable resonance.

Of course, the resistor-capacitor series pair could be wired in either element first. If drawn with the R on the amp output and the C to ground it makes it unobvious that the purpose of the resistor is as a load. Confusion is generated by this.

On top of that, the R-C pair are spotted by people and called a "Zobel Network" and the series L doesn't get a mention. For this diplexer to work, you do need both parts. If anything, the series L does most of the work.

Zobel analysed and documented an awful lot of network meshes, just about every one there is, there is equal justification for planting his name on every one of them. Zobel lives on in the image-parameter maths for designing the full diplexer Two resistors, one inductor and one capacitor.

Back to cables...

There is another cable construction with very low subtended area - coaxial cable, and this has appreciably less capacitance per metre than the woven stuff. Its Zo is higher than 8 Ohms, but unless you plan on using miles (literally!) of the stuff, it won't matter.

All speaker cables, whether interwoven bank-busters, lengths of RG213 coax or Woolies bell wire will act as antennae. Fancy construction affects the differential RF voltages acquired, but it does ABSOLUTELY nothing for avoiding common-mode pick-up.

Good audio design is normally taken to mean a single-point ground and srar connection. Good RF design wants ground wires and screens hard-connecting to the metal box right at the point of entry. The two ideals do not go together well. Taking the view that RF is harder to get right than hum, I go for grounding at the point of entry, and using differential signals to void the hum.

Common mode pick-up can be fixed with a ferrite core surrounding both speaker cable conductors at the amp end of the cable. This works well and isn't expensive and fixes the biggest RF pick-up mode which rather ironically, the boutique cables usually ignore.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 14th Apr 2019, 8:26 am   #676
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Audiophoolery?

Great detail David - I must do more reading about Zobel.

I spent many hours at university designing things like Cauer filters, which were on the syllabus. Along with Chebychev filters. Not a lot about more regular and simpler filters like Butterworth.

All this was theoretical - we never had to build one and measure it in a practical session, which seems in 40-odd year hindsight a bit of an omission.

Back to output filter networks. On mine first 0.1uF/10 ohm and then 2uH//10 ohms. The 2uH is out of 1.5mm diameter wire, for all the good reasons you set out.

And you are correct - there is no native attenuation of common mode in any speaker cable. And there is a horror in the audio world of any ferrite clamps on cables (although beads are often put on mains lead ground wires). Which misses the point that for regular audio band differential signals the ferrite disappears - there is no field from the wires that pass through it.

Craig
Craig Sawyers is offline   Reply With Quote
Old 16th Apr 2019, 9:45 am   #677
Alvin
Heptode
 
Alvin's Avatar
 
Join Date: Nov 2003
Location: East London, UK.
Posts: 761
Default "Hi-Fi-Fo-Fum" Vintage Audiphoolery

Just came across this brilliant BBC film from 1959
https://www.facebook.com/watch/?v=363622534242894
Alvin is offline   Reply With Quote
Old 16th Apr 2019, 10:57 am   #678
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: The Audiophoolery Thread.

Superb! I loved the bit of them all shuffling into the sweet spot of the stereo speakers.

Flanders and Swann did a skit on "high fidelity" in 1957 https://www.youtube.com/watch?v=f_DptPvj7ts
Craig Sawyers is offline   Reply With Quote
Old 16th Apr 2019, 11:39 am   #679
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: The Audiophoolery Thread.

Quote:
Originally Posted by Craig Sawyers View Post
And yes the characteristic impedance is comparable to a loudspeaker impedance.
At audio frequencies the cable characteristic impedance will be nothing like a speaker impedance. At radio frequencies the speaker impedance will be nothing like its nominal impedance. Hence in both cases there will not be a match but fortunately, as someone said, this does not matter too much.

Just to avoid any confusion, it may be worth repeating here that the characteristc impedance of a cable at audio frequencies is not sqrt(L/C); that is the radio frequency approximation, which is only valid when resistance can be ignored. At audio frequencies the characteristic impedance is approximately sqrt[ R/( j 2 pi f C ) ] - which is frequency-dependent and a mixture of resistance and capacitance.
G8HQP Dave is offline   Reply With Quote
Old 16th Apr 2019, 12:25 pm   #680
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: The Audiophoolery Thread.

All that is absolutely true. And as you say irrelevant in loudspeaker cables other that at RF frequencies.
Craig Sawyers is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 6:43 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.