Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[danlavry]

locosoundman wrote on Sat, 11 December 2004 15:48

Great topic.  I know just enough basic electronics to get myself really confused about this. Your source impedance is the microphone, right?  So does the impedance of the preamp figure into this at all?  Say I am going from a 150 Ohm mic to a 600 Ohm preamp (I imagine this is pretty common in most live situations with really long snake/cable runs of 100'+).  How does this work with the figures you gave before? Since capacitances in series actually diminish as they add up, wouldn't the longer runs of cable actually have less overall capacitive reactance in their absolute impedance (total resistance would be going up, capacitance down)?  Wouldn't there also be inductances to consider in a cable of this length? Best, Rob

“Your source impedance is the microphone, right?  So does the impedance of the preamp figure into this at all?  Say I am going from a 150 Ohm mic to a 600 Ohm preamp (I imagine this is pretty common in most live situations with really long snake/cable runs of 100'+).  How does this work with the figures you gave before?”

My comments were very general. I used an example of a source with 75 Ohms source and that is a good everyday example for a line amp. I assumed the destination is high impedance compared to the source and that is a good reasonable every day assumption.
The situation regarding cable capacitance requires a case by case evaluation.

Your microphone example is one such case. Again, mics have different source impedance, certainly when going from ribbon to dynamic to condenser. The same is (or should be) for the load presented by the preamp. But the basic concepts apply.

From a circuit standpoint, the driver, cable and load are reduced to a very simple model:
A voltage source driving a network through a series resistor. The network is a capacitor (due to a cable) in parallel to a resistor (the load). The series resistor is the source resistance. Indeed it is a very simplistic model, and one can add the series resistance due to the cable, and some inductance, and skin effect, and proximity effect, and distributed line model…. And that is indeed what we do to look for finer details. For audio (such low frequencies)  the skin effect is no big deal, the inductance has minimal impact, a distributed line model is of little value. For digital audio (higher frequencies in the MHZ) I would go for a more complicated model.

“Since capacitances in series actually diminish as they add up, wouldn't the longer runs of cable actually have less overall capacitive reactance in their absolute impedance (total resistance would be going up, capacitance down)?  Wouldn't there also be inductances to consider in a cable of this length?
:?”

You are not adding capacitance in series. You are “sort of” adding it in parallel, so longer run add capacitance. Take the old capacitance model of 2 parallel plates. You get the same area wit 100 inch by 100 inch square plate, or 1 inch by 10000 plate. The later is more like a cable – narrow and long. The fact that some of the plate is 100 feet away does not matter at low frequencies because the current moves at near the speed of light – for all practical purposes, what happens at the end of the cable is as if it happened at the beginning and visa versa. Now, instead of parallel plates, think of  a circular structure of a cable. There is capacitance there.

Yes there is inductance at play, but again very small amount of it to impact the low frequencies (such as in audio). You go to higher frequencies and inductance matters a lot.

My post was only a limited reply, restricted to the issue of cable capacitance, with emphasis on if and when you get frequency roll off due to audio cable capacitance. I did not even touch on anything else, because cables are a huge subject.

I will try and find time to cover a couple of other basic factors in the unbelievably hyped and full of misinformation subject of cables. It is easier for a cable salesman to sell you on “it sound good”. You ask why, and some of the answers you get belong in the “crazy house”. So many of the statements regarding cables are presented with authority by people that know how to sell, but have less than a remote clue even of the most basic fundamentals of what is going on

Best regards
Dan Lavry  

[button]

May we hear your opinion on digital cabling?  This has also (sadly) become an *audiophile* zeppelin, inflated by the hot air of sales talk:

'the ------- allows more digital information from the source to reach the destination undistorted, and it's audible: subjectively, reduced or eliminated reflections in a digital cable is perceived as more silent background, cleaner overall sound and better "focus" and "attack"'
and
"We have brought forth a product line that is not only the Pinnacle of Accuracy  but also demonstrates the ability to deliver the ambiance of the event along with the emotion of the artist.'

And that's just the digital cables, we haven't started on the AC cords yet!

[Roland Storch]

A digital cable like AES/EBU (110?) is always also a very good connection for analog signals. It is usually just less flexible and more expensive.

I would not use it on location outside the studio because of its mechanical characteristics. But for fixed installations in bigger studios it is common practice to use only digital cables for analog and digital signals because of easier logistics.

[RMoore]

I've been using cable from Gotham - which is 'Mogami'-like, but a bit cheaper, at least over here :

http://Http://www.gotham.ch

They make all kinds of cable & formats..

There's one special double braided shield type which has improved RFI rejection..handy for mic cables.

My feeling is its easy to get hugely paranoid about cables & as long as one uses good quality eg: non-radio shack and keeps lengths short as possible its bound to be good enough for most studio needs...with the good quality ones eg Mogami, Canare, Belden , Gotham etc - it seems the main differences are marketing / pricing and ease of use like soldering, stripping the wire insulation, flexiblity of the jacket, certain formats (eg: fireproof insulation, quad, foil, shielding etc)  as opposed to any significant sonic differences.. (?)..

In my studio I used lots of Gotham cable, some Mogami and some 'recycled' Belden, some Gepco and other random cable that came with multitracks..
and even a bit of some recycled absolutely insane SILVER cable multicore from German broadcast..As opposed to being like a wire or multistrand,  its a solid thin  'rectangular' piece - like a kind of ribbon..
I get the sense you don't want to coil this stuff up or flex too many times..
Apparently it cost a <sick> fortune per meter originally..
I was lucky to score some from someone who was at the right place at the right time when stuff was getting thrown away (!)..
ahh, the practices of state funded radio & TV !

I used it on some 'critical' outboard like my best mic pres.

FWIW - the silver seemed to sound best..the Belden the least, but not 'bad'..
just in comparison.
The easiest cable to work with was the Gotham.
The worst to work with (yet 'special' in its way) was the silver, also the Mogami was a pain cuz the wire insulation is not very 'forgiving' eg: it melts back real quick down the wire..in my limited experience anyway.
The Gotham is very forgiving in that respect,

Good luck,

RM

[fishtank]

Ryan Moore wrote on Sun, 12 December 2004 12:34

I've FWIW - the silver seemed to sound best..the Belden the least, but not 'bad'.. RM

These types of statements can be dangerous.  Did you actually do a double-blind study to determine that the silver sounded better?  

One of the things I really love about this particular forum is Dan's policy about keeping things discussed here based on scientific fact.  The audiophiles always seem to sneak in and rave about some esoteric cables etc. that *seem* to sound better or how they can *sense* something above 20 KHz.

[PookyNMR]

danlavry wrote on Sat, 11 December 2004 11:53

I will try and find time to cover a couple of other basic factors in the unbelievably hyped and full of misinformation subject of cables.

That would be VERY appreciated!

danlavry wrote on Sat, 11 December 2004 11:53

So many of the statements regarding cables are presented with authority by people that know how to sell, but have less than a remote clue even of the most basic fundamentals of what is going on

That's exactly why I posted the question here.  I'd like a technical assessment of what cable properties measurements will work for my application rather than marketing BS.  And with regards to cables, there's so much BS out there, it's hard to even tell if certain advertised properties even matter.

Thanks,

Nathan

[RMoore]

<These types of statements can be dangerous.  Did you actually do a double-blind study to determine that the silver sounded better? >

Hi - thanks for bringing this up & the answer is no I didn't make any scientifically valid test or A/B comparison of any kind at all..
Actually I can't be bothered to !
Although I am concerned with good quality sound - as far as cabling goes, I am much more interested in using some quality materials that are good enough for the application & getting on with the job of recording and mixing...
Otherwise its just too easy to get confused and worry and fret excessively about 'oh jeez maybe brand X would've been .0001% better sounding..
Thats just me - Some people may feel differently.
If my business was mastering for example I would probably be paying more attention to any possible minute differences in cabling.
Anyway - in the case of the silver cable this was based on my 'memory' of how certain units in question were sounding before & after silver cable..
Hardly scientific I know..
I would be interested to know the results of any scientific study in this area. (cable comparison minus the BS and marketing hype!).
From what I understand when people have tried to do this there is little if any audible difference noticeable between quality cables - its only when its really cheap like Radio Shack or lampcord style (anyone remember the 'ol 70's trick of using AC cable for home hifi speakers?!) that anyone can hear a diff..
?
Cheers,
RM

[Sahib]

fishtank wrote on Sun, 12 December 2004 18:28

Ryan Moore wrote on Sun, 12 December 2004 12:34 I've FWIW - the silver seemed to sound best..the Belden the least, but not 'bad'.. RM These types of statements can be dangerous.  Did you actually do a double-blind study to determine that the silver sounded better?   One of the things I really love about this particular forum is Dan's policy about keeping things discussed here based on scientific fact.  The audiophiles always seem to sneak in and rave about some esoteric cables etc. that *seem* to sound better or how they can *sense* something above 20 KHz.

I completely agree on the amount of hype that is made on cables
but in terms of silver vs copper, of course silver is better as it is a better conductor than copper. Hence you do not need to see a proving formula to accept it. However, how much difference does a silver conductor make in audio frequency  and how possible it is to hear that change is another matter. But I can assure you that it would make a difference.

Cemal

[danlavry]

Sahib wrote on Mon, 13 December 2004 20:29

fishtank wrote on Sun, 12 December 2004 18:28 Ryan Moore wrote on Sun, 12 December 2004 12:34 I've FWIW - the silver seemed to sound best..the Belden the least, but not 'bad'.. RM These types of statements can be dangerous.  Did you actually do a double-blind study to determine that the silver sounded better?   One of the things I really love about this particular forum is Dan's policy about keeping things discussed here based on scientific fact.  The audiophiles always seem to sneak in and rave about some esoteric cables etc. that *seem* to sound better or how they can *sense* something above 20 KHz. I completely agree on the amount of hype that is made on cables but in terms of silver vs copper, of course silver is better as it is a better conductor than copper. Hence you do not need to see a proving formula to accept it. However, how much difference does a silver conductor make in audio frequency  and how possible it is to hear that change is another matter. But I can assure you that it would make a difference. Cemal

My ITT Reference Data For Radio Engineers tells me that resistively at room temperature (20 degree C)
Copper resistivity is 1.7241 Ohms-Cm X10^6.
Silver resistivity is 1.62 Ohms-Cm X10^6.
Gold resistivity is 2.44 Ohms-Cm X10^6.

Using copper as a reference, silver is about 6% lower and gold is about
41% higher resistivity for the SAME DIMENSSIONS.

In other words, increasing the cross sectional area by 6% will make copper resistivity equal to silver. In terms of diameter it translates to about 3% more diameter for the copper wire. In other words, just slightly thicker wire will more than overcome the material differences.

The reason I included gold is to expose some of the hype. In terms of resistance, gold is worse than copper. Of course you can add 41% more gold and be as good as copper, and it does cost money!!!

Again, one needs to examine the specific application. There are cases (very high frequencies) where the current “defies” the diameter of the material, and concentrates on the skin (skin effect), and coupled with long runs, one can loose a lot of energy. Such cases may call for silver coating. There cases where silver oxidation is not welcome and gold plating serves as protection, and so on.

For this post I am only taking about materials and resistivity.
Audio is low frequencies and one need not worry about skin effect. To say that silver vs copper matters, one needs to more than just assume that we are talking about the same exact dimensions. The basic question is at what point will the resistor begins to impact the signal.

At low frequencies, the simple model of the cable resistance is as if it were just a single resistor (component). We need to account for both signal and return path (2 wires, a wire and shield or what not). Say we have a 2 wire cable (signal and return). The wire used is 18 gauge annealed copper at room temperature (20 degree C). We get 6.4 ohm per 1000 feet in each direction, or a total of 12.8 Ohms round trip per 1000 feet. For a 16 gauge it is 8 ohms round trip and so on. So does it matter?

Clearly 12.8 Ohms is a hack of a lot if the load is an 8 Ohms speaker. Most of the energy would be wasted on the wire, and so we do not run 1000 feet of 18 gauge for speakers. But say the load is 10 KOhm. The loss in voltage is about .13% (0.011dB). That is for 1000 feet!

As a rule, at audio frequencies, high current cases call for attention to cable resistance. Sending say a 5V DC supply line to a 5 Ohm load will simply not work. But most audio cases have relatively high load impedance, where is no longer about resistance, and most often, as explained in my previous post, not about capacitance either.

I am not saying any cable is just as good as the other. I am saying: For most audio cases (low drive impedance, high load impedance) it is not about silver vs copper, or issues relating to resistivity. Of course I am assuming proper wiring of the audio cables, to carry only audio, and not be a part of some high frequency chassis return path or some AC power ground loop…

Everything I said changes at higher frequencies (MHz and above).  

BR
Dan Lavry

[LumenStudio]

Wow, lots of great info so far.  I'm really interested in this topic as I need to add additional wiring to my setup.  

So far I understand lower capicitance is better, and resistivity matters, but not enough to be concerned about (so just go with silver neutriks and be done with it).

Now if lower capacitance is better, I'd buy the digital AES cables, for all my stuff if it is that easy.  I only have a project studio and cables around here are between 5-10 ft.   I also have 3 CRT (not LCD) monitors, a few computers, a bunch of outboard gear, a digital mixer, etc.  Now with all this stuff, should I be worried about interference noise and get quad cables?  Or do I go for the lower capaitance AES cables?

Much appreciated.

Cheers,
Lumen

[RMoore]

FWIW - it strikes me that the cable subject voyages into a realm similar to 'faith' and political beliefs where there are many options but no clear single 'best way' for all people - so my thought is research what's out there, make your choice based on what seems best for you & move on to make the hits of 2005 and beyond..

RM

[locosoundman]

One more stupid question from me...

Since capacitive reactance has an inverse relationship to frequency and capacitance (I think it is -1/(6.28fc), then as frequency or capacitance increases, the overall capacitive reactance will decrease, which in my mind translates to less of an effect on the signal passing through both in phase and impedance.  Intuitively, I know this logic is wrong, I am just not sure why.

It stands to reason that cables should be transparent, but taking the above into account, if a capacitance does exist it almost seems as though a larger capacitance would be better.  Can someone please shed some light on this?

Thanks,
Rob

[danlavry]

locosoundman wrote on Tue, 14 December 2004 17:03

One more stupid question from me... Since capacitive reactance has an inverse relationship to frequency and capacitance (I think it is -1/(6.28fc), then as frequency or capacitance increases, the overall capacitive reactance will decrease, which in my mind translates to less of an effect on the signal passing through both in phase and impedance.  Intuitively, I know this logic is wrong, I am just not sure why. It stands to reason that cables should be transparent, but taking the above into account, if a capacitance does exist it almost seems as though a larger capacitance would be better.  Can someone please shed some light on this? Thanks, Rob

The general impedance is 1/sc for all signals. When you apply a sine wave of a known frequency it get simplified to 1/(i*2*pi*f*C)  and your 2*pi is about 6.28, but the i is very important (engineers often call it j), because it signifies that the impeadance is an imaginary number. Imaginary in this case means that the current leads the voltage by 90 degrees.

as frequency or capacitance increases, the overall capacitive reactance will decrease, which in my mind translates to less of an effect on the signal passing through both in phase and impedance. Intuitively, I know this logic is wrong, I am just not sure why.

It is wrong because you are viewing it as if the cable capacitance were in series with the signal. In fact it is in parallel with the load.

BR
Dan Lavry

[locosoundman]

Thanks Dan.  It goes back to the error I was making in my first post.  The capacitance is in parallel with the resistance, not in series.  Thanks for your patience.

Best,
Rob

[ssltech]

Steve Lampen at Belden has authored several excellent common-sense papers on these very issues.

An ideal cable should be low resistance, low inductance, low capacitance, just a straight black line on a schematic.

I've found that certain 110Ω digital audio cables can be microphonic in certain audio applications (load-impedance can surpress this effect to a greater or lesser extent) but generally, copper (and lots of it) is good, becoming of increasing importance where the load impedance is lower. Capacitance is bad, becoming of increasing importance where the source impedance is higher. Inductance is bad, becoming of increasing importance where the load impedance is lower.

Many domestic devices have "anti-idiot" build-out resistors so that Joe Q. Public can short left and right together to get mono, without blowing anything up and returning his "defective" CD player to best buy for a refund. Sometimes these build-out resistors, coupled to higher-capacitance cables, (often made with cheap aluminium foil conductors.. I kid you not!) can result in a loss of sonic quality (it's a simple R-C network after all). In such instances, comparing a $50 monster cable to a 25-cent cheapie cable will make you buy the monster. -Removing the build-out resistor would also have likely improved the signal with the cheap cable to about the same extent!

So the really important question is: what are the interface impedances and conditions? -That's an excellent reason why speaker cable is bad for mics (no matter what "boutique" brand you use!) and vice-versa!

Hi-Fi magazines and other works of fiction usually do not have the readership brainpower to factort in these issues, and are advertiser-influenced to the degree that they just say "Buy a $500 power cable, some $300 speaker cables, $185 interconnects, and marvel at the improvement".

Me, I say buy decent cable, use good connectors (Neutrik are innovative and convenient, but Switchcraft are electrically better... trust me!) and consider the impedances of the devices which are being connected together while making your choice.

That really is all there is to it.

The 'Voodoo' is Doo-doo!

Keith Andrews