Not sure if this is the right place for this, but here goes. I'm in computers with a software background, but limited knowledge of electronics, which I am now trying to address. I started off trying to understand something I thought was really simple - cables. Interconnects first.
So it's impedance and capacitance (innit?)
According to my humble calculations, a co-axial cable with a 1mm diameter core and a shield at 1mm radius (ie 2mm diameter) should have a capacitance of about 60 picofarads per metre. This agrees at a sort of order-of-magnitude level with what I've read. If I increase the cable thickness to 4mm (about the same as TV coax) then the capacitance goes down to 38picofarads per metre. So that would seem a good cheap way of achieving an improvement, if a difference of 22picofarads in my interconnects is relevant. So how to find that out? I'm of the school that feels if it's not there, you can't hear it.
The only decent, understandable facts I can find on signal attenuation or interference based on capacitance are at MHz frequencies, here: http://www.littelfuse.com/data/en/Application_Notes/ec624.pdf.
So although they show very little interference at 10 picofarads, the problem is:
a) The voltages. What is the voltage range in an interconnect?
b) The frequencies. There is a perceptible difference when the frequency is multiplied by 20 (from 12MHz to 240MHz), but obviously one can't extrapolate to KHz
c) Capacitance. The difference between 10picofarads and 550 picofarads at 12Mhz is huge. Where would 60picofarads sit?
d) Materials. Nothing in my limited research provides factual support that capacitance in a cable is affected by the materials it's made of, although there are masses of sales bumf and subjective warblings . Reference here: en.wikipedia.org/wiki/Capacitance#Stray_capacitance
So my data, although interesting, is not relevant. So I'm looking for better data, where at least one of frequency and voltage is more within listening range(Not surmise, experience or conjecture - there is bags of that available!)
I have a good analogue oscilloscope, so my next step is to do some blind tests on generated waves through various levels of capacitance.
Anyone out there to help?
So it's impedance and capacitance (innit?)
According to my humble calculations, a co-axial cable with a 1mm diameter core and a shield at 1mm radius (ie 2mm diameter) should have a capacitance of about 60 picofarads per metre. This agrees at a sort of order-of-magnitude level with what I've read. If I increase the cable thickness to 4mm (about the same as TV coax) then the capacitance goes down to 38picofarads per metre. So that would seem a good cheap way of achieving an improvement, if a difference of 22picofarads in my interconnects is relevant. So how to find that out? I'm of the school that feels if it's not there, you can't hear it.
The only decent, understandable facts I can find on signal attenuation or interference based on capacitance are at MHz frequencies, here: http://www.littelfuse.com/data/en/Application_Notes/ec624.pdf.
So although they show very little interference at 10 picofarads, the problem is:
a) The voltages. What is the voltage range in an interconnect?
b) The frequencies. There is a perceptible difference when the frequency is multiplied by 20 (from 12MHz to 240MHz), but obviously one can't extrapolate to KHz
c) Capacitance. The difference between 10picofarads and 550 picofarads at 12Mhz is huge. Where would 60picofarads sit?
d) Materials. Nothing in my limited research provides factual support that capacitance in a cable is affected by the materials it's made of, although there are masses of sales bumf and subjective warblings . Reference here: en.wikipedia.org/wiki/Capacitance#Stray_capacitance
So my data, although interesting, is not relevant. So I'm looking for better data, where at least one of frequency and voltage is more within listening range(Not surmise, experience or conjecture - there is bags of that available!)
I have a good analogue oscilloscope, so my next step is to do some blind tests on generated waves through various levels of capacitance.
Anyone out there to help?
You seem to be doing rather well on your own.
Below are two good pages, but each site as many other pages that you can search for.
http://www.st-andrews.ac.uk/~jcgl/Scots_Guide/audio/Analog.html
Characteristic Impedance of Cables at High and Low Frequencies
Maximum audio signal voltage is 2 V RMS.
But audio is at such low frequencies that any reasonable cable will do just fine under any reasonable circumstance.
Below are two good pages, but each site as many other pages that you can search for.
http://www.st-andrews.ac.uk/~jcgl/Scots_Guide/audio/Analog.html
Characteristic Impedance of Cables at High and Low Frequencies
Maximum audio signal voltage is 2 V RMS.
But audio is at such low frequencies that any reasonable cable will do just fine under any reasonable circumstance.
You might want to look up transmission line theory.
This looks at sending signals down transmissions lines which is basically the same as sending signals down capacitive/inducitve cables.
This looks at sending signals down transmissions lines which is basically the same as sending signals down capacitive/inducitve cables.
Transmission line theory only starts to work at audio frequencies when the interconnect approaches a kilometer long.
Hi,
Its only related to radio frequencies not audio but the characteristic
impedance of a phono plug / socket is about 180 ohms. For audio
frequencies your talking bulk effect loading, i.e. cable capacitance.
Voltage levels are near irrelevant.
Dielectic constant of the insulating material affects capactitance.
The effect of a cable is defined more by the the driving point,
which is never an ideal voltage source, than the terminating
point, which can be treated as a resistive load.
rgds, sreten.
Its only related to radio frequencies not audio but the characteristic
impedance of a phono plug / socket is about 180 ohms. For audio
frequencies your talking bulk effect loading, i.e. cable capacitance.
Voltage levels are near irrelevant.
Dielectic constant of the insulating material affects capactitance.
The effect of a cable is defined more by the the driving point,
which is never an ideal voltage source, than the terminating
point, which can be treated as a resistive load.
rgds, sreten.
Last edited:
Thanks for the links
Thanks for the links. Does anyone have a data source for the effect of capacitance or inductance on a square wave at audio frequencies and line voltage?
I've read some of Malcolm Hawksfords papers and the articles by Rod Elliot, but I just can't find basic experimental data.
Thanks for the links. Does anyone have a data source for the effect of capacitance or inductance on a square wave at audio frequencies and line voltage?
I've read some of Malcolm Hawksfords papers and the articles by Rod Elliot, but I just can't find basic experimental data.
It's not that hard to do the math.
Total cable capacitance in parallel with the load impedance.
Total cable inductance in series with the load impedance.
Remember that almost all musical recordings and bandwidth limited below 20kHz.
Total cable capacitance in parallel with the load impedance.
Total cable inductance in series with the load impedance.
Remember that almost all musical recordings and bandwidth limited below 20kHz.
To a first approximation the effect of a cable at audio frequencies is just a single-pole low pass filter formed by the output resistance of the driver and the total capacitance of the cable. You ought to be able to find the effect of a low-pass filter on a square wave.
Don't believe everything you read from people selling exotic cables. Some of it is nonsense, although they may sincerely believe it themselves.
The inner insulation is a dielectric, and dielectrics are not perfect, so it could have some effect but this will be small. Ignore all talk of transmission lines at audio frequencies, unless your amp is at the opposite end of the town from your source.
Don't believe everything you read from people selling exotic cables. Some of it is nonsense, although they may sincerely believe it themselves.
The inner insulation is a dielectric, and dielectrics are not perfect, so it could have some effect but this will be small. Ignore all talk of transmission lines at audio frequencies, unless your amp is at the opposite end of the town from your source.
- Status
- Not open for further replies.
- Home
- Source & Line
- Analog Line Level
- Help newbie please - Basic data on interconnects