Audiophoolery - Nov 09 Nuts and Volts magazine - Editor comments

[k4kyv]

My wife bought us a Wave Radio for Xmas, probably 10-15 years ago.  It sits in the kitchen on the top of the fridge.  Audio quality and craftsmanship are better that most radios its size, but I would say it was way overpriced.  The FM radio and CD player both run in stereo, but that is a waste because the speakers are only about 12 inches apart, so to perceive any stereo image you would have to sit less than a foot or two away from the radio.

The digital tuning system sucks.  Instead of a simple straight forward up/down function and a separate scan and seek like what you find on most car radios, it has sort of a combination of up/down tuning and scan.  It is very difficult to land on a weak station and stay there.  Then you have to double click on the channel to keep it from continuing to scan for a stronger station.

But the AM and FM tuner does an above average job with weak signals and rejection of adjacent channel interference, once you manage to get the channel tuned in.  All these years, we have managed to not lose the remote control unit.

Aside from the exorbitant price, I would give it an overall "B" rating.

[W2DU]

Concerning wire size for speaker cables, if the size is too small, and with long lengths, the series inductance of the wire will degrade the bass response; So the larger wire size is necesary for flat frequency response, even though not necessary for high powers.

Walt

[k4kyv]

The resistance of too small a wire is a factor, too.  The amplifier driving a speaker is a  lot like the audio driver for a class B modulator.  The impedance load on the amplifier varies widely with frequency, regardless of the nominal impedance of the speaker.  The acoustics of the enclosure can substantially affect the impedance load that a speaker presents to the amplifier.  So you want the amplifier to be an audio voltage source with as low an internal resistance as possible for minimal distortion, just as in the case of a class B driver.

A long, small gauge wire is in effect wiring a resistor in series with the speaker.  It raises the internal resistance of the audio voltage source, the amplifier.

But that doesn't mean you need exotic oxygen-free copper wire @ $100/ft for speaker cable.  Any heavy gauge copper conductor will do - even the power cord stock that looks just like zip cord, except for #12 or #10 gauge wire instead of #18.  If you don't plan to move the speakers or cable around once installed, romex cable with solid conductors will do just fine.

[Ian VK3KRI]

A capacitor across a speaker wire pair would not do much for anything but differential mode interference.

That idea was in a lot of hambone publications. It must have worked once in a while. 

Typically in a solid state amp one wide will be at ground potential. so the cap will be from the active side to ground. I read somewhere that the RF pickup issue was mainly due to the fact that  RF on the active speaker lead can ride back via the feedback circuit straight to the input stages of the amp.  Ive never been able to determine if thats true  because any amp I owned that picked up RF from the speaker side was presumed to be a piece of crap and thrown away.

                                                                                 Ian VK3KRI

[steve_qix]

Ahhhhhhhhhhhhhhhhhhhhh yes! This has been a very enlightening conversation.  *NOW* I know why my signal has been getting gradually stronger over the past 30 years.  It's because my antenna wire has been properly "broken in".  I have 30 year old wire in my main antenna for 75 meters, but I used "new" wire on 160.  So THAT's why it took a few years for the 160 meter ant to approach the performance of the 75 meter dipole.

Hmmmmmmm. I wonder how much I can get for my "aged and broken in antenna wire"?  30 years of aging.  Gotta be worth something!  Of course, my signal will probably drop 5 dB if I sell the "broken in" wire and use "new" wire......  Time to add more MOSFETs 

[WBear2GCR]

Walt, series inductance will degrade the highs, not the lows...

Too small of a conductor will cause insufficient current to be available, and that will decrease the damping factor, which in turn will tend to actually increase the bass output since the electrical control is reduced.

A small enough conductor might show some heating effects which would yield some non-linear effects, but that's not usually a practical consideration.

A series capacitance would roll off the highs.

When a conductor gets reasonably large for example greater than 10ga. one starts to see some degree of series inductance. There are a number of websites where this effect is measured and documented, usually as a slight (typically 0.1dB or less) droop a 20kHz.

Steve, it is funny to poke fun at it, but for most of what we do in ham radio we are only concerned about 1st order or 2nd order effects. In the world of serious sound reproduction things like the level of 7th harmonics even when the overall level of higher harmonics is seemingly below audible levels is acknowledged by virtually all credible circuit designers (engineeers) as being important and a design criterion.  Higher order effects can not simply be dismissed and mocked.

Snake oil is what happens when people take elements of the science and engineering and use it for advertising and propaganda... a different matter than where the underlying leading edge information really is WRT things that are seemingly subtle and meaningless.

Anyone who wants to review a very long thread that over-archs this very subject, participated in by some very top people in the field (JAES writers amongst them) just ask.

                         _-_-

[steve_qix]

One thing has always puzzled me.  I used to be really into high end audio (all homebrew), when my entire house was a single guy's lair, and I could put anything in any spot I wished...  and speakers (literally) as large as an old-fashioned refrigerator were properly placed in the living room, and good sound prevailed.......

... I put the amplifiers in the very immediate vicinity of the speakers themselves, to reduce the wire inductance and resistance.  In some cases, I only had a few feet of #12 wire between the (very low Z) amplifier output and the actual speakers.

That *seems* to be the way to go if one really wants to eliminate wire losses.

Regards,

Steve

[WBear2GCR]

Yes Steve, correct.

But the fly in the ointment is that then you have long runs of low level cable. So the question is which effects the sound more, the low level or the high level... the low level.

Otoh, one could go with low Z balanced drivers and balanced inputs on all ur gear... but that adds cost, circuit complexity and you can't have ur audio friends (most of them) bring their latest "Umpah 1000" over for an audition!

 

                   _-_-

[k4kyv]

Too small of a conductor will cause insufficient current to be available, and that will decrease the damping factor, which in turn will tend to actually increase the bass output since the electrical control is reduced.

That is still a form of distortion.  This is exactly what I was saying about the desirability of maintaining the audio driving voltage from as low impedance (lowest internal resistance) source as possible, just like you would do in the case of a class B driver.

A series capacitance would roll off the highs.

But the question of capacitance vs impedance remains.  A .001 capacitor across a hi-Z crystal microphone would kill the high frequency response, if not all the audio altogether.  But strapped across an 8Ω speaker, at what frequency would a 3 dB rolloff occur at .001 mfd?

When you are talking about 7th harmonics and overtones, it may be visible on the scope, and possibly audible in actual sound pressure from a live concert.  Since the human ear is a non-linear device it is entirely possible that intermodulation products amongst tones, all at frequencies well beyond the reach of human hearing, might be perceptible, but how many loudspeakers or headphones that exist in the real world would have any response whatever at the 7th harmonic of 10 kHz musical tones generated at an orchestral concert?

[Opcom]

From a 4 Ohm source, across a 4 Ohm load, about 48 MHz. But maybe not. it depends on the output Z of the amp.

The amp driving a 4 Ohm speaker better not have a 4 Ohm output Z.

More like an upper limit of 0.4 Ohms, to behave half-decently. Except for professionals and picky people's amps, then the limit might be much less, like less than 0.01 Ohms.

I don't think the kind of people that have to call into radio talk shows for advice will have good enough hi-fi gear to be able notice the degradation caused by the capacitor, or good enough (if any) test gear to measure it. The host shouldn't have cut you off, obviously he did not really know better and was used to helping with problems like 60Hz hum and turntable rumble.

and I was foolin about the wire. Although in response to the inductance issue, if it is some kind of twin lead, the L should mostly cancel, should it not?

[VE3GZB]

Breaking in a cable? What the heck?? Are these people nuts?

Hmmm, well in this economy it's clear there are people who have more money than brains!

[VE3GZB]

When I mentioned this to the talk show host, he went into a tirade about how you would never place any kind of capacitor across the speaker output terminals, because it would totally ruin the the performance of the amplifier.  When I tried to question his statements and present my side of the story, he was like some of to-day's AM political talk show hosts; he got downright angry and any further discussion was like arguing with a brick wall.

And this was long before the advent of voodoo electronics involving breaking in speaker cables and the sound quality of power cords.

Sad how $$$ from advertisers has infiltrated and so corrupted honest science and electronics.....so today you have thieves and charlatans peddling their rubbish - and the uneducated masses fall for the hype.

It makes me mad really - but trying to correct someone whose bread and butter comes from the packaging and marketing of dishonesty is no simple task (as you've observed).

Hopefully, if current economic troubles continue due to inbred corruption and greed in the system, then more rats like this will fall into the economic sewer and then have to learn how to earn money honestly. Something good can come out of trouble as long as people try to avoid such foolishness.

[Opcom]

Someone once said "The music business is a cruel and shallow money trench, a long plastic hallway where thieves and pimps run free, and good men die like dogs. There's also a negative side."

I suppose the 'phoolery industry has also embraced that ideology.

Since we know better we make fun of it but I don't like noobs being intentionally cheated. One has to wonder of the FTC could ever level a fraud charge.

Remember the home audio power ratings of the '70's? 200 watts IPP, 2 watts RMS. That has been straightened out, but I still find $40 car stereos claiming 50W/ch X4 and for a true example of car audio amps, the LANZAR VCT 2220
with 1175 Watts x 2, ... $159.95

All it means is the home hi-fi industry has found a new way to lie because the old way was forbidden, and the automotive hi-fi industry has taken note of the opportunity.

The poor sucker buying a car stereo, besides getting scroo'd on the magical wire, he's getting it on the amps and speakers.

[k4kyv]

The same thing happened in amateur radio with  p.e.p..  The concept was introduced to hams in the late 50's and early 60's as slopbucket began to be highly promoted in a high-pressure sales campaign, and the manufacturers of leen-yar appliances used it to inflate the power ratings of their amplifiers in the ads.  A couple of sweep tubes would claim 500-watt capability, and the standard advertised rating for a "legal limit" linear was 2KW (p.e.p.)  Sometimes there was only a small asterisk next to the "2 kw" with a tiny footnote at the bottom that read "p.e.p." Of course, it was all input power back then.

[flintstone mop]

If you don't plan to move the speakers or cable around once installed, romex cable with solid conductors will do just fine.

ohhhh Don I bet some audiophiles are cringing with that statement!!!! ROMEX WIRE HA!
And I agree that 20G stranded might be stressed for runs longer than 40 feet.
WE're back to need ing test equipment to show any change in a signal that can't be detected by the best of ears. It's the grey matter in between the ears the "presception of what's being heard" that starts all of these myths.

Fred

[WBear2GCR]

Don,

The 7th harmonic is not the 7th harmonic of a 10khz. tone. Clearly that would be above audible frequencies, and unless it aliased or effected the performance in the audio band that would be meaningless.

It is both measurable and audible, with an appropriate system and source material, of course.

Meaningless for >95% of all existing audio systems & listeners, maybe a higher number.
That doesn't change the issue of if it can be heard, or is audible in an absolute sense.

The 0.001ufd capacitance is not a problem. I did not object to them being used. I did say that an NPO would be a better choice in that application than a Z5U...

The desireablility of "maintaining a low Z driving source" depends almost entirely upon the load being driven, it can not be stated as an absolute.

Opcom & Don, the Damping Factor of an amp is a measure of the output resistance vs a given (8 ohm) impedance. So a Crown Macrotech that has a DF of 10,000 has an output Z of a small fraction of an ohm. The QSC USA1300 (~300wrms 8 ohms) has a DF of 80. A typical tube amp with feedback has a DF of about 10. A tube amp such as a SE triode without feedback has a DF of 1! In otherwords the DF of 1 means the output Z = load Z.

For hifi applications one wants to "match" the Qts (the Q of the woofer in the box) which equates to the filter response, and so the frequency response of the system to the amplifier & room (some rooms have boost at some low frequency) such that the total in room response (Q) is not underdamped (unless you like "boom bass"). In some instances a low DF amp is optimum, and in some the opposite is true. At most all other frequencies the output Z of audio amplifiers (never being greater than the load) makes almost no difference whatsoever... fwiw.

Fred, 20ga is too small for anything but short runs to a very low power speaker. It is pretty simple to check the circular mils vs. current capability... the agreed upon standard for speakers by all engineers and experts in the audio field is ~>14ga. and 12ga is about optimal.

Romex and zip cord are usable for speaker wire, but they are both somewhat capacitive, and so may not really sound that good on a low distortion system - on an average system like most people use, and like I use in the lab and the bedroom, just fine.

Opcom, modern car amplifiers can make big watts! because they have built in switching supplies (up converters) that make high voltage to run the amp rails!! The rating is in to 2 ohms. SO for example 800watt rating = 400 into 4 ohms and 200 into 8 ohms. Some are rated into one ohm.

I know someone who is a consultant who "fixes" the designs for manufacturers, and some of these car amps are actually quite good as amplifiers... also some of them today are going class D...

Breaking in wires -  way beyond the scope of this forum.

                      _-_-bear

[Opcom]

Opcom & Don, the Damping Factor of an amp is a measure of the output resistance vs a given (8 ohm) impedance. So a Crown Macrotech that has a DF of 10,000 has an output Z of a small fraction of an ohm. The QSC USA1300 (~300wrms 8 ohms) has a DF of 80. A typical tube amp with feedback has a DF of about 10. A tube amp such as a SE triode without feedback has a DF of 1! In otherwords the DF of 1 means the output Z = load Z.

For hifi applications one wants to "match" the Qts (the Q of the woofer in the box) which equates to the filter response, and so the frequency response of the system to the amplifier & room (some rooms have boost at some low frequency) such that the total in room response (Q) is not underdamped (unless you like "boom bass"). In some instances a low DF amp is optimum, and in some the opposite is true. At most all other frequencies the output Z of audio amplifiers (never being greater than the load) makes almost no difference whatsoever... fwiw.


Opcom, modern car amplifiers can make big watts! because they have built in switching supplies (up converters) that make high voltage to run the amp rails!! The rating is in to 2 ohms. SO for example 800watt rating = 400 into 4 ohms and 200 into 8 ohms. Some are rated into one ohm.

I know someone who is a consultant who "fixes" the designs for manufacturers, and some of these car amps are actually quite good as amplifiers... also some of them today are going class D...


                      _-_-bear

I know about damping factor, but I don't worry about it too much. I mostly want to avoid clipping and crossover distortion and other definitely bad sounding stuff.

I suppose I could measure the output Z of some of my tube amps, that would be fun to see where they are. The amps I use for bass, those are solid state shaker table amps by Ling. It's only because I already have them, and it is not inexpensive for me to make 800W/ch with tubes. The output Z on the Lings is not specified, only the voltage, current, and freq response: 34V RMS, 30A RMS +/-0.5dB 5Hz to 10KHz. They need a 1-ohm load.  With that, above mentioned 12 GA romex speaker wire becomes a bigger issue than the damping factor.

What would you think a Quad or (less commonly) Mcintosh tetrode tube amp using a separate winding for cathode feedback would have as a damping factor? I think much lower than without K FB. The iron is very important for it.

I can imagine what 32 6550's would look like (and cost). That would definitely need to be a rackmount panel with the tubes sticking out horizontally to the front, just for visuals. All that stuff, it is for those who need to have the best. For me, if the old junk sounds good when cranked, I am happy.

Car stereo amps are fine, once you find out what the true ratings are.

[KC2IFR]

The damping factor of an amp always brings many opinions. I look at the damping factor sorta like the shock absorbers on a car. The shocks control the action of the springs........no shocks and the springs will continue to go up and down until they run out of energy.
The damping factor tries to control the action of the speaker. If u apply a quick a pulse to a speaker, when the pulse stops, u want the speaker to stop......its sort of like the short one puts on a meter when u ship it.....the short bucks the back emf when the needle tries to move because of movement.
The damping factor does the same thing.........it tries to control any movement of the speaker cone when the input stops.
This could get more complicated but Ill stop here.

Bill

[k4kyv]

Romex and zip cord are usable for speaker wire, but they are both somewhat capacitive, and so may not really sound that good on a low distortion system - on an average system like most people use, and like I use in the lab and the bedroom, just fine.

How is the capacitance of "speaker" wire different from zip cord of the same gauge?  Most of the speaker wire sold commercially looks to have exactly the same construction as zip cord, except the insulation is transparent plastic instead of opaque plastic.  Could there be that much difference in the dielectric constant?

For my main living room stereo amp, I use "Monster Cable" that I picked out of  Lowe's wire scrap barrel several years ago for pennies on the dollar of the normal selling price.  Evidently someone had purchased a 50 ft. length and returned it, and instead of restocking it, since it was detached from the main spool, they put it in the scrap barrel. IIRC, I paid about 40¢ a foot for it. It looks like about #8 gauge stranded copper with the typical transparent zip cord-like speaker wire insulation.  The conductors are heavy in weight and very flexible, much like wire rope, made from a large bundle of hair-size strands of copper wire braided together.  I had trouble attaching the cable conductors to the amplifier and speaker terminals because it was too big to fit the receptacles, so I had to fabricate some homebrew adaptors for the ends of the cables.  I couldn't tell any  difference in the sound from the speakers, except that I eliminated some flaky intermittents - my old speaker cable was several lengths of smaller gauge stuff spliced together, and occasionally the splices would make poor contact or the electrical tape would fall off and they would short, so I was constantly having to fiddle with those old JS cables.

[WBear2GCR]

Romex and zip cord are usable for speaker wire, but they are both somewhat capacitive, and so may not really sound that good on a low distortion system - on an average system like most people use, and like I use in the lab and the bedroom, just fine.

How is the capacitance of "speaker" wire different from zip cord of the same gauge?  Most of the speaker wire sold commercially looks to have exactly the same construction as zip cord, except the insulation is transparent plastic instead of opaque plastic.  Could there be that much difference in the dielectric constant?

It's not different by very much if the geometry is about the same.
So-called "fancy" speaker cables are not usually clones of zip cord using better materials.

there are a variety of ways to make speaker cables - they have some effect on the perceived sound.

If ur interested PM me and you can do some home DIY versions and decide for urself if it has any effect or not.

 

              _-_-bear

[WBear2GCR]

I know about damping factor, but I don't worry about it too much. I mostly want to avoid clipping and crossover distortion and other definitely bad sounding stuff.

I suppose I could measure the output Z of some of my tube amps, that would be fun to see where they are. The amps I use for bass, those are solid state shaker table amps by Ling. It's only because I already have them, and it is not inexpensive for me to make 800W/ch with tubes. The output Z on the Lings is not specified, only the voltage, current, and freq response: 34V RMS, 30A RMS +/-0.5dB 5Hz to 10KHz. They need a 1-ohm load.  With that, above mentioned 12 GA romex speaker wire becomes a bigger issue than the damping factor.

What would you think a Quad or (less commonly) Mcintosh tetrode tube amp using a separate winding for cathode feedback would have as a damping factor? I think much lower than without K FB. The iron is very important for it.

I can imagine what 32 6550's would look like (and cost). That would definitely need to be a rackmount panel with the tubes sticking out horizontally to the front, just for visuals. All that stuff, it is for those who need to have the best. For me, if the old junk sounds good when cranked, I am happy.

Car stereo amps are fine, once you find out what the true ratings are.

Ya sure... you need to send me those Missing Link amps... they're not being "exercised" enough now...

The McIntosh output design is very good. Requires the xfmr to be wound for the application.
Edit: oh, the DF? Not much different than a usual tube amp, since the DF of a usual tube amp is due to global feedback lowering the output Z... but with zero FB the McIntosh will have slightly lower output Z due to the winding that is on the cathode, being in parallel to the one in the plate, the entire primary also being lower Z than the transformer used in the plate of the same tube in a standard circuit.

But how would that compare to the output Z of a transformer output designed to run off a Ruskie 6C33C?? 

Not a fan of the Quad "current dumping"...

There is an commercial amp (VAC, iirc) that had a slew of tubes, probably EL34s or else they were 6650s...
probably you can find it online.

Fwiw, global feedback around a not so good output xfmr that "flattens" the response is not so good...

               _-_-bear

               _-_-bear

[flintstone mop]

The damping factor of an amp always brings many opinions. I look at the damping factor sorta like the shock absorbers on a car. The shocks control the action of the springs........no shocks and the springs will continue to go up and down until they run out of energy.
The damping factor tries to control the action of the speaker. If u apply a quick a pulse to a speaker, when the pulse stops, u want the speaker to stop......its sort of like the short one puts on a meter when u ship it.....the short bucks the back emf when the needle tries to move because of movement.
The damping factor does the same thing.........it tries to control any movement of the speaker cone when the input stops.
This could get more complicated but Ill stop here.

Bill

OK BILL Right on!! The higher the DF the tighter the bass response. Now some speakers (older) will sound better with low DF from a TUBE amp. And some tube amps will have muddy bass with speakers demanding higher current or speakers with lower SPL's.

Fred

[KC2IFR]

Thanks Fred,
U are also correct. I have found that a high DF is more noticeable when  using vented bass enclosures.
The sealed enclosures help to control the cone excursions by there very nature but are not as efficient.

Bill
 

[flintstone mop]

Now now, Bear,
You can't slight McIntosh amplifiers. Their freq response was real and good to 100kc. I know! Coz during my pirate FM days, I modulated a grid dip meter that coupled 6mhz stereo FM to an aircraft transmitter (10W pushing the hell out of 5763's) I drove the grid dip meter with an MC-30 and connected to the MAC was a stereo generator. Pretty good stereo separation,19khz pilot, the whole thing.
KRAP, the little brown spot on your dial, 108.5.........busted by the FCC in 1973. We were actually around 109mhz. National airport near Washington DC had concerns about music programming being heard.

Fred

[WBear2GCR]

Fred,

Did I say a single disparaging word about McIntosh amps? Nah.

I did say that I was not fond of the Quad system...

The output stage design of the McIntosh is one of my favorite designs of all time. By a fair margin too.
How they executed the rest of the amp depends mostly on the specific model - check out the schematic for a K-107 - it also goes by another number - it has 8005 tubes in the output and a slick bootstrapped driver circuit. Mega power.

                          _-_-bear

PS. I have a 50-W (I think that is the number) an original McIntosh amp, uses xfmr coupled driver stages, and the whole thing is potted in the chassis - not much more in there other than parts and potting!