[Joe Bloggs]

Quote:

Originally posted by AnsBjork I did not quote that paper at all in my comments, so your "gentle" accusation is uncalled for...

Sorry, it's really because of my lack of patience and expertise in this area that I couldn't give you a better answer

Still, I'll try to highlight the parts in that thread that refute the theory of varying capacitance and inductance:

KikeG
Quote:

I'd say capacitance or inductance don't vary with frequency. What varies with frequency is capacitive or inductive part of impedance.

This is the simplest statement of the truth

TwoJ
Quote:

cap & ind do vary with freq. Since pure cap or ind only exist in theory there will always be a resistive part hence impedance

This guy at first holds this position, but after this

Pio2001
Quote:

No, a capacitor with a capacity of C will have the same capacity at any frequency. Its impedance Z, that is -j/(Cw) varies with frequency, since C stays the same, and w varies.

He changes his position

TwoJ
Quote:

Well blow me down, you're right! I hope they don't revoke my ee degree

Pure capacitance and inductance exist only in theory--this just means there will always be resistance, not that the C and L will change

cabbagerat seems to hold two different positions at the beginning and end of the same post:

Quote:

A number of posters have already stated that Impedence and Capacitance are dependent on frequency, but most of these have been rather vague (but still correct). I will attempt to give a better explaination for those without an EE background.

Quote:

...The Capacitance and Inductance of a cable are not dependent on frequency. However, the effect that they have on the signal passing through a cable does depend on the frequency of the signal.

So which is it???

What he was trying to say was really this

KikeG
Quote:

I'd say capacitance or inductance don't vary with frequency. What varies with frequency is capacitive or inductive part of impedance.

(back at the top of the page)

JeanLuc sums up what was happening in the thread nicely:

Quote:

Seems to me we are confusing capacitance/inductance (which are device constants due to the coil/capacitor layout) and impedance (which in fact is the vector-like addition of both complex resistances, RL and RC) Impedance varies with frequency (RC of a given capacitance hyperbolically falls with rising frequency while RL of a given coil rises with rising frequency) - that's how frequency crossover networks work.

This still manages to happen in that forum because not everybody there is an EE major, either.

Hope I haven't made you dizzy. I've become quite dizzy myself. Excuse me while I find somewhere to puke

 

[AnsBjork]

Quote:

The IBM paper you cited is not really relevant to the cable.

dvw, kindly read my previous post on the subject. It is not supposed to be directly relevant. Read instead Steve L's articles.
Quote:

Steve L's article mentioned only frequency dependent impedance not RLC. Impedance = 1/2/pi/f/c with capacitnace only.

Well dvw, I think I will quote Steve so I can show that I read it carrefully. I think it can settle the question nicely once and for all.

************************************************
What We Can Measure in Speaker Cable?

Resistance (in ohms):

This effect turns electrical flow into heat. It affects all frequencies equally. Changing the size (gage) of the conductor changes its resistance. Resistance usually is specified per unit length (i.e., 100 feet, 100 meters, 1,000 feet etc.) and is additive.

Capacitance (in picofarads):

With speaker cables that are paired, or coaxial, the two conductors with insulation in between form a capacitor. Capacitors hold an electrical charge. The capacitance is small and measured in picofarads. It is additive and usually displayed per foot or per meter and must be multiplied by the actual cable length to get the total capacitance. Capacitance affects the signal level and is frequency-dependant The higher the frequency, the greater the reactance caused by the capacitance and the greater the signal loss.

Inductance (in microhenries):

Inductance is the ability to hold a magnetic charge. All conductors have inductance. It is also frequency-dependent but is in series with the cable, as opposed to capacitance, which is in parallel. The inductance of a cable is small, and the effect is cancelled out by the capacitance. Therefore, inductance rarely is specified in manufactured cables.

Impedance (in ohms):

The total of resistance, capacitance and inductance. As frequencies get higher, resistance becomes less and less of a factor. At frequencies above 10 MHz or so, only capacitance and inductance are left, so the impedance settles to a "characteristic" value.

Skin Effect (in inches):

As frequencies get higher, the signal tends to travel on the outside or "skin" of a conductor. A minor effect for analog audio, skin effect isn’t a major factor until well into the Megahertz.
************************************************

I am also saying that resistance is indirectly frequency dependant because of skin effect, since the apparent section of the cable is getting smaller for higher frequency, resistance must go up.

I hope this settle the question and that we can proceed with the thread.

 

[dvw]

AnsBjork

I appologize for misreading your post. I have been in the IC design for 20 years and can't help trying the clarify the non-relationship between the IBM paper and cable.

Regarding Steve L's article, I think he is trying to put it in layman's term. The effective impedance Z=j/wC and jwL. j is the imaginery number and w is the frquency. L and C are constant (non frequency dependent). So it is accurate to say the effect caused by L and C is frequency dependent but not so to say L and C are frequency dependent.

It is also correct to say skin effect cause the effect resistance to be larger due to more concentrated current density. I can look up the formula for you if you wish. But this does not happens until the signal is very high frequency (multi-MHz and up). If this happens at audio frequency, we will have tremendous problem distributing power (60Hz) or telephony (up to 4 KHz).