[Tweaker]

Does anyone ever just do a simple test of your pet interconnects or speaker wires? Connect them to your driving component and a reasonable dummy load, drive the source with high frequency square waves and see how bad the output looks at the load with a scope. You can connect one probe to the source input and the other probe to the load, match levels, and then add and invert the signals and see the difference signal (any is wrong).

Yeah, yeah, I do that all of the time because you should never ever trust your ears. I always like to wait untill someone publishes some graphs and charts to tell me it's ok to hear what I'm hearing and whether or not it's a bad, good, or no difference. Of course you should never trust the data provided by the manufacturer as they could be LYING therefore one must wait for a neutral third party to verify the data and someone else to analyze just what exactly you should be hearing based on the analysis of the third party so that you will be able to know that you should be hearing this or that difference or no difference at all! It may be time consuming (we're talking years or perhaps even NEVER, here) but it could very well save a person a ton of money that might otherwise be wasted buying those expensive Radio Shack Gold interconnects!

[bhobba]

Does anyone ever just do a simple test of your pet interconnects or speaker wires?  Connect them to your driving component and a reasonable dummy load, drive the source with high frequency square waves and see how bad the output looks at the load with a scope. You can connect one probe to the source input and the other probe to the load, match levels, and then add and invert the signals and see the difference signal (any is wrong). You are gonna have   expressions when you see how bad many exotic cable ...

Frank your words of wisdom are as timeless as they are correct.  But most audiophiles are not techos, often shun technical issues and resort to the mantra - have you heard them.  IMHO this is not the correct path - but does seem to be the most favored.  And if one resorts to listening then blind ABX testing is the only valid method.  But even that seems to often get short changed by audiophiles.  The cables I want to get are very simple and I have no doubt they would pass your test with ease
http://www.tnt-audio.com/clinica/piano6_1_e.html
Cat 6 is rated over 200mhz.
The enhancement I would do is place a bybee at the non source end - which would be the major cost.  I know bybees sound like snake oil - but a deep interest of mine is the mathematics of Quantum Field Theory and what I know of that discipline suggests to me it may have some validity.  That is where blind ABX testing comes in to disentangle highly controversial claims that may not be easily measured but could be audible - although it may be possible to detect a lowering of noise or some other objective measure with them.  

Thanks
Bill

[bhobba]

I bought Element Cables and like them just fine, but I'm not a "cable guy". I've heard the differences between $1,000 speaker cables, but would invest in better room treatments or speakers long before putting it into wires.

Very very true IMHO - get a DEQX - it will yield much better bang for your buck.  No problems with Franks measurment philosophy there - flat frequency responce and time alignment is readily measureable.

Thanks
Bill

[bhobba]

What you say is true. However, minimum distortion does not necessarily correlate to good sound. I guarantee a $200 Technics AV receiver will have significantly lower distortion specs than my tube amplifier, but my tube amplifier sounds significantly better. Daniel von Recklinghausen said, "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you've measured the wrong things." (From the VAC owner's manual) One could assemble an uber low distortion system based solely on specs and scope readings, but I'll wager it'd sound like shite.

Please be aware Frank makes some of the finest sounding amplifiers on the planet so he knows what he is talking about.  People like Frank and Hugh Dean who actually design the stuff are only too aware that if an audible difference exists then one can find an objective measurable reason for it.  And if that reason is you prefer a certain type of distortion because it is more musical to you - that is fine - but one is left with the question of what you want as an audiophile - recreating the original or listening to pleasing distortion:
http://en.wikipedia.org/wiki/Valve_sound

It is not wrong to answer pleasing distortion - just be aware that is your preference and judge things accordingly.  Also be aware such distortion can sometimes 'grate' after a while.  My preference is recreating the original with pleasing distortion if I must have it to get other qualities like better transient response that may be damaged by negative feedback to reduce that distortion - or whatever.  That is where guys like Frank come in - they have applied their engineering skill to achieve the best compromise they can.  You may like a different compromise - but be aware that is what it is about.

Thanks
Bill

[Daemon]

AVIHiFi's opinion is pretty much the fact of the matter. If tested that way, it also shows that there are very real discernable differences between cables (ie they don't all have the same distortion). Idealy, you'd think audiophiles would want a cable that best approaches perfect transmission, and such cables are usually relatively inexpensive and simple.  So why isn't this the case? Since Monster began the cable revolution, and since it was deemed that any form or tone control or filtering was evil, audiophiles have been using their cables as extremely costly filters. They're paying for the cable distortion to modify their sound, or so it seems, so they can still pretend their system is 'pure' or 'true' since they obviously don't have any naughty parametric equalizers or such in their system, just interconnects.

I'm sure some cables can make some systems in some rooms sound better (psychoacoustics and spending justification asside), but surely you'd be far better off buying reasonable quality cheap cables and doing the corrections in a controlled and measureable manner with a DEQX, which though pricey, isn't much diffferent in cost to a full suite of exotic cables (probably cheaper), and will surely do a better job as a filter system. I'm not one of those that subscribes to the "any cheap old amp can be made to sound just like any other by a skilled sound technician using a parametric EQ" theory, which is just two bit sound tech' ideology and can be shown mathematically to be untrue, but surely using such a device has got to be better than trying to use cables as a hit or miss filter system when it comes to getting your sound just how you like it (since that's what you're doing with cables, all pretence of purity and accuracy asside).

As an engineer and a scientist for both military and private sector, I feel pretty sure that Bybees are BS and have pretty much nothing to do with quantum field theory, QED, QCD or anything else with mathematical or theoritical validity that can reasonably be prefixed with the word 'quantum'.  I'm sure there are alot of theories that apply to them that can be described with words like 'marketing' and phrases like 'burden of proof'. but that's just my theory.

[kfr01]

Does anyone ever just do a simple test of your pet interconnects or speaker wires?  Connect them to your driving component and a reasonable dummy load, drive the source with high frequency square waves and see how bad the output looks at the load with a scope. You can connect one probe to the source input and the other probe to the load, match levels, and then add and invert the signals and see the difference signal (any is wrong).

You are gonna have   expressions when you see how bad many exotic cable ...

 
:clap:

[bhobba]

Up to this point we are in perfect agreement

.  As an engineer and a scientist for both military and private sector, I feel pretty sure that Bybees are BS and have pretty much nothing to do with quantum field theory, QED, QCD or anything else with mathematical or theoritical validity that can reasonably be prefixed with the word 'quantum'. I'm sure there are alot of theories that apply to them that can be described with words like 'marketing' and phrases like 'burden of proof'.  but that's just my theory.

Well I am no expert but do know something of it and am not quite that sure.  I regularly post to physics forums and some actual scientists might like to give an opinion - I will give that a whirl and see what they say.  One point not in their favor is that Jack Bybee claims to have developed them through extensive measurements and listening tests - I have never seen those measurements published.  Jack if you are reading this techo types like me would really like to see those measurements.  If Jack or anyone else knows where those measurements are published please let me know - I would dearly love to see them.  But what is not theory - what is an actual fact - is properly conducted blind ABX listening tests.  The proof of bybees may need to come from that.  I have seen enough information from technical people who I do trust for me to spend money conducting such a test one of these days.  I can provide stronger evidence one way or the other then.

Beyond that I suspect the cheap cables I linked to without bybees will be very hard to beat.

Thanks
Bill






Thanks
Bill

[TomW16]

Mmmm, your "understanding" differs from mine!!  

Irrespective of the wire connecting the two components (shall we say preamp to power amp, just for argument's sake?), you should have the inoput impedance of the destination at least 10 times the output impedance of the source. At least 50 times is better.

Hi Andy,

I should have used better terminology when I said "matched" impedance.  Your stated 10 fold and 50 fold ratios were exactly what I meant.  Sorry for the confusion.

Tom

[Steve]

I agree tvad4.

Not to get off discussion, but I routinely check components using a square wave, checking for abnormalities that should not be present.

So again, I connected an IC to the output of a preamp, loaded by 100k and 75pf, Frequency 50 khz, and tried connecting the 2 probes, first, to the preamps Output and the second probe to the IC output (load), just looking for any change caused by the IC. Then I connected the probes to the preamp Input and the load at the end of the IC, so both the preamp and IC are checked.

The conclusion is that no difference was found measuring between the output of the preamp and the load. So, conservatively, the cable produced no difference up to 1mhz. No ringing etc noted.

However, this does not mean other cables might not produce distortion artifacts. So each brand and model needs to be checked if possible.

Next, connecting the probe to the preamp input and the 2nd probe to the load did produce a difference. (The preamp is included in measuring any changes in the squarewave.) A "spike" with a trailing slope occurred. To someone doing his first measurements, at first glance, something looks horribly wrong.  

However, the difference shown simply indicates the finite frequency response of the preamplifier. This is quite normal as the spike would essentially disappear if the preamp's gain were flat to, say, around 500khz on up. So, what I measured and saw was simply an indication that the FR was finite, and couldn't produce constant gain to at least 500 khz or higher. A sine wave would also be useful in measuring the frequency response.
No other artifacts/distortions present, such as ringing.

However, this may not be the case for other components, so each should be measured if possible.

I think this shows that cable capacitance is a consideration, lower is better in most cases. One could also add a buffer stage to handle higher capacitance cables, but then we are adding another stage, with more colorations, just to use a higher capacitance cable and keep the frequency response ultra high.

The specs may look good, but adding another stage will obviously add more colorations.

I would support tvad4's position.

What you say is true. However, minimum distortion does not necessarily correlate to good sound.

I guarantee a $200 Technics AV receiver will have significantly lower distortion specs than my tube amplifier, but my tube amplifier sounds significantly better.

Daniel von Recklinghausen said, "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you've measured the wrong things." (From the VAC owner's manual)

One could assemble an uber low distortion system based solely on specs and scope readings, but I'll wager it'd sound like shite.

[avahifi]

Are you sure that the spike you saw was not caused by the capacitive element of the cable used causing the preamplifier's output stage to slow down trying to drive the load, causing an excess signal in its feedback loop, causing the voltage amplifier stage to generate a correction signal too large and late, making the feedback go positive, showing up as an underdamped high frequency spike?  Of course this would indicate that the whole preamp circuit was messed up while this was happening.  You can certainly hear this, for better or for worse.

Try the test again with a cable with a lower capacitance per foot rating and see if the results do not change for the better.

If your cables are modifying your equipment's performance, the results are bad, even if you like the effect.  Remember that the guy driving by with the car radio putting out window rattling whale farts likes that best too.

Oh well.

Frank Van Alstine

[Harmon]

Get Gregg Straley's Interconnects at [email protected].  They will put a very big smile on your face!!!    

[guest1632]

I agree tvad4.

Not to get off discussion, but I routinely check components using a square wave, checking for abnormalities that should not be present.

So again, I connected an IC to the output of a preamp, loaded by 100k and 75pf, Frequency 50 khz, and tried connecting the 2 probes, first, to the preamps Output and the second probe to the IC output (load), just looking for any change caused by the IC. Then I connected the probes to the preamp Input and the load at the end of the IC, so both the prea ...

"Next, connecting the probe to the preamp input and the 2nd probe to the load did produce a difference. (The preamp is included in measuring any changes
in the squarewave.) A "spike" with a trailing slope occurred. To someone doing his first measurements, at first glance, something looks horribly wrong.

However, the difference shown simply indicates the finite frequency response of the preamplifier. This is quite normal as the spike would essentially disappear
if the preamp's gain were flat to, say, around 500khz on up. So, what I measured and saw was simply an indication that the FR was finite, and couldn't
produce constant gain to at least 500 khz or higher. A sine wave would also be useful in measuring the frequency response.
No other artifacts/distortions present, such as ringing."


Hi Steve, Interesting stuff here, Now, this same test if no cables were attached and you did this last test, no spikes? Then Frank's assersion  would be correct. If there was a spike, well, go from there. If it's your cable that you are selling, and the spike occurrs, then you gotta look at ways to lower the capacitance of the cable. If you read the stuff for tboth Pear and the Revelation cable, Capacitance is only one part of the equation. The question for both you and Frank is how much of that junk is just that junk, versus real artifacts that can make a cable sound better or different.

Ray

[Steve]

"Are you sure that the spike you saw was not caused by the capacitive element of the cable used causing the preamplifier's output stage to slow down trying to drive the load, causing an excess signal in its feedback loop, causing the voltage amplifier stage to generate a correction signal too large and late, making the feedback go positive, showing up as an underdamped high frequency spike? Of course this would indicate that the whole preamp circuit was messed up while this was happening. You can certainly hear this, for better or for worse.">>

Two very basic points clearly show your assumptions to be false concerning my experiment.

1) Any feedback used in the preamp would not begin to cover the amplitude that was seen/measured on the scope.

I don't know of any preamp that uses high levels of feedback, global or otherwise that would follow your logic. So how does your comment relate to real life?

2) The squarewave shows exactly what the sinewave measurements show. They are in agreement.

Every component will show this type of waveform to some degree, unless the frequency response approachs infinity, which I don't believe any does.

To suggest that a buffer/cathode follower is necessary for excellent fidelity when only the ultra high frequency response is affected in a good component, is incorrect, because only the ultra high frequency response is affected. We are talking about 100khz and up.

Simply use a low capacitance cable and get rid of the colorizing buffer stage, and you'll hear the sonics improve, unless the basic design is very poor. In otherwards, with a buffer stage added, one gets sonic degradation whether one wants it or not. Your stuck even if one uses low capacitance ICs, which is advantageous.



>>"Try the test again with a cable with a lower capacitance per foot rating and see if the results do not change for the better.">>

Of course it will, no matter what stage is used because the frequency response is increased. But remember, we are dealing with ultra high frequencies, well above 50khz.

>>"If your cables are modifying your equipment's performance, the results are bad, even if you like the effect. Remember that the guy driving by with the car radio putting out window rattling whale farts likes that best too.">>

 Even a buffer, cathode follower would show a change. And that is my point. One should strive for minimum cable capacitance rather than adding a buffer stage and using high capacitance cables. As long as one is well outside the audio band. By the way transformer coupled outputs should always be checked.

[Steve]

HI Ray,

     How low of capacitance do you want? 40pf is pretty low. And yes, Frank's assertions are correct, at 100khz and above. But it is simply amplifude rolloff, not added artifacts at this high frequency.

And if one can get the music to sound the same with the component either in the system, or out of the system, isn't that perfection?

     Secondly, how high of a frequency response do you want? 1 mhz, or higher? No preamp will do that, unless one gets into video stages, as an example.

In fact, many preamps with a buffer won't have any higher frequency response than a traditional stage. They will just handle higher capacitance cables, BUT why, when one wants low capacitance cables to begin with.

[avahifi]

What I was trying to say was that if you add a capacitve load to the output of the driving device, this will normally show up as a spike in the square wave response, caused by the load modifiying the linearity of the driving device.  There have been jillions of bench tests published showing this. This is a bad thing.

I have seen some standard off the shelf Beldon shielded cables with a capacitive rating of 10 pF per foot or thereabouts.  You can measure yours with a simple multi-meter, those doing everything from capacitance to thermal temperature measurements are really cheep these days and are highly recommended for any hobbiest.  A Tenma 72-7740 from MCM Electroncis for example.

We would not expect to see a square wave spike for the device driving a resistive load, only a gentle high frequency rolloff if the equipment is properly designed.  Another simple bench test, if you drive your unit to near full power on a 1 KhZ sine wave into a resistive load, and then run the frequency up higher and higher, does the sine wave simply decrease in amplitude at some very high frequency, or does it turn into a triangle wave shape?  If it does the later, is is subject to slew rate limiting, and that is another topic to talk about.

Adding an additional buffer stage to drive the load may or may not be a good thing, depending upon how badly the source behaves on the load in the first place, and how difficult the load is, and how much the buffer detracts from the music in and of itself.

My advice always has been to minimize the capacitive loads your equipment has to drive.  Now what pray tell is in one of those Bebee filters anyway?

Frank Van Alstine

[Steve]

>>"What I was trying to say was that if you add a capacitve load to the output of the driving device, this will normally show up as a spike in the square wave response, caused by the load modifiying the linearity of the driving device. There have been jillions of bench tests published showing this. This is a bad thing.">>

I can see an amplifier behaving oddly. A good preamp may have problems if an output tranny, etc is used. Otherwise, a good preamp shouldn't have any spike, just a gentle roll off of the frequency response, which of course, occurred with the preamp under test. The spike only showed the frequency response was not infinity, which ANY preamp would also show.

I have a tektronics capacitance meter. My 1 meter ICs measure 40pf total, counting 10pf for each plug's capacitance.

>>"We would not expect to see a square wave spike for the device driving a resistive load, only a gentle high frequency rolloff if the equipment is properly designed.">>

I reread your comment, and have to disagree. The spike, between the source and output, is simply a result of the Difference between a "perfect" square wave and the frequency response of ANY preamplilfier. This is a simple comparison between source input and output. A simple rolloff would occur if one were using a sine wave as a source, not a square wave.

>>"Another simple bench test, if you drive your unit to near full power on a 1 KhZ sine wave into a resistive load, and then run the frequency up higher and higher, does the sine wave simply decrease in amplitude at some very high frequency, or does it turn into a triangle wave shape? If it does the later, is is subject to slew rate limiting, and that is another topic to talk about.">>

I would suspect an amp is more suseptible, but then I haven't tested alot of competitor's preamps. I used both a resistive and capacitive load on the preamp under test, and the sine wave stays sine till the generator tops out, around 600khz, with the amplitude gently rolling off.

>>"My advice always has been to minimize the capacitive loads your equipment has to drive. Now what pray tell is in one of those Bebee filters anyway?">>

Amen. And I have never even seen a Bebee, so I couldn't lend a thought?

[avahifi]

My generator goes out to three meg, and it is very interesting to sweep equipment from the one to three meg range and see all the lumps and bumps in the responses show up on some equipment.

Our suggestion that any underdamped resonances at any frequency from your equipment is bad bad bad.  How they actually reflect back into the audio range that you hear is something not clearly understood by us, other than that the musicality is definately improved when you get rid of them by careful design work at all frequencies, not just at audio frequencies.

Frank Van Alstine

[guest1632]

What I was trying to say was that if you add a capacitve load to the output of the driving device, this will normally show up as a spike in the square wave response, caused by the load modifiying the linearity of the driving device.  There have been jillions of bench tests published showing this. This is a bad thing.

I have seen some standard off the shelf Beldon shielded cables with a capacitive rating of 10 pF per foot or thereabouts.  You can measure yours with a simple multi-meter, those doing everyt ...

Hi Frank,

Well, capacitance sure seems to be part of this cable making equation. What about the rest, like diaelectrics, the geometry of the cable, all of the matherials in the making of the cables, the type of solder used, the connectors, and so on. What we ought to do is to restart a thread and discuss for us poor hobbyist, who don't have a clue, what makes a good cable and what stuff to consider versus what is just garbage. It does seem that cables do make a difference, and now as to how much well, that's probably the $64000 question.

You and Steve at the end of the last few posts pretty much seemed to be in agreement with each other.

Ray

[guest1632]

My generator goes out to three meg, and it is very interesting to sweep equipment from the one to three meg range and see all the lumps and bumps in the responses show up on some equipment.

Our suggestion that any underdamped resonances at any frequency from your equipment is bad bad bad.  How they actually reflect back into the audio range that you hear is something not clearly understood by us, other than that the musicality is definately improved when you get rid of them by careful design work at all frequencies, not just at audio frequencies.

Frank Van Alstine

Hi Frank, Ok, I have always felt that a high bandwidth amp is the best in the long run, but back to cables. So if we are seeing spikes of resonances in the 1 to 3 MHZ range, then aren't we perhaps looking at harmonics induced from those spikes that would show up in the audio range as perhaps noise?

Ray

[bhobba]

My advice always has been to minimize the capacitive loads your equipment has to drive. Now what pray tell is in one of those Bebee filters anyway?

Bybee filters are a very controversial element (probably composed of proprietary ceramics) that is supposed to reduce noise.
http://www.bybeetech.com/
Take what is said about how it works with a grain of salt (I mean nano technology carbon?).  Jack Bybee has mentioned elsewhere no one really understands exactly how they work.  I have an interest in the math of quantum stuff and on that basis think there MAY be some theoretical validity in it (but probably not what the site says - my feeling is it MAY be related to superoconductivity and cooper pairs and stuff like that - but that is just a guess).  But the fact he does not publish measurements when he claims he developed them using measurements is not a good omen.  They are claimed to be originally developed to reduce battery noise to a minimum for military power applications.  

The controversy is that evaluation is done purely on listening tests which are of course notorious unless done blind.  You will find technical guys who should know their stuff swear by them and others say they can hear nothing.  I just trust the guys that claim they can hear a difference enough to fork out the dosh and one day construct a cable with them in to see if I can hear a difference and suggest others interested in cables may consider doing the same as well.  If it fails you wasted nowhere near what they charge for uber expensive cables and it is claimed they wallop those:
http://www.enjoythemusic.com/magazine/equipment/0702/bybee.htm
'Jack Bybee was there, and I tagged along while he went into various rooms and persuaded the proprietors to stick his cables into their systems in place of the megabuck wires they were using. To near universal astonishment, Jack's wire was in usually decidedly better; sometimes it was a close call. What was Jack's miracle interconnects? They were some thirty-buck Mogami wire with a Quantum Purifier at each end, just before the RCA plug.'

Evidently ordinary measurements show vanishing small inductance and capacitance and a very small resistance.  

My advice for what its worth is to get a couple and see if you can detect some benefit either audible or measurable which is exactly what I plan to do one day.

Thanks
Bill