Fantastic
Well-Known Member
For many of us cable capacitance should not pose any practical problems due to loss in HF response.
Lets look at a cable with 150 pF per m capacitance. That's high !
The loss of HF with a 2 meter coaxial cable ( 300pF) and an amp with 22K input impedance with a 20 K volume pot and a 100 ohm out impedance of the preamp or the source is as follows.
-0.1 dB at over 360 Khz ,
-0.5 dB at over 800 Khz ,
-1dB at 1.2 MHz
However for a tube amp with source resistance of 3Kohm ( like a single tube preamp ) and an input impedance of 1 Meg ohm and volume pot of 250K ohms , the response is as follows.
-0.1 dB at 23 khz ,
-0.5 dB at 52 Khz ,
- 1 dB at 77 Khz.
( All calculated values and cross checked with simulation )
HOWEVER , using different cables could sound very different even though they don't seem to have any significant effect at 20 Khz on steady state continuous signals! Dynamic performance is still not fully understood as you can see on the various cable comparison threads.It doesn't seem to be the capacitance which is the culprit. More to do with dielectric ( storage?) performance on dynamic signals.
So with solid state amps the possibility of cable capacitance causing a major problem appears lower than with a tube amp using high impedance pots.
We have not accounted for stray capacitance which could reach significant values.One reason why point to point wiring appears to sound better than pcb mounted components in tube circuits. The 47Lab chip amp with PP wiring also supposedly did sound very good. I have never seen anyone comparing it with a pcb based design which by themselves are quite good.
If anyone wants to plug in their own values of capacitance and loads I can give you the XLS file that I made. There is no calculation for 20 KHz exactly but I can add it if you want.
Some other interesting data.
Typical capacitance of RCA connectors .
The simplest chromed chassis mount type was 2.8 pF (MX) and the gold plated one with insulators for chassis mounting was 5.7 pF ( Jap) !
Chetan Cables single core shielded cable was 150 pF per meter. ( 5mm dia)
No name balanced cable had 150 pF/m to shield and 76 pF between the two signal cables.
A Srexact balanced cable ( from SP road Bangalore) was 95 pF /m from signal line to shield and 54 pF between signal cables.
Typically I find most balanced cables are around 120 pF/m between signal and shield. However the plastic and quality of the shield used varied widely and so would the sound I guess.
RCA plugs.
MX silver type 4.1pF
MX chromed case/gold connector end ( expensive ) 2.9 pF
So I guess IC cables better be picked after listening tests rather than selecting by looking at their specs.
Lets look at a cable with 150 pF per m capacitance. That's high !
The loss of HF with a 2 meter coaxial cable ( 300pF) and an amp with 22K input impedance with a 20 K volume pot and a 100 ohm out impedance of the preamp or the source is as follows.
-0.1 dB at over 360 Khz ,
-0.5 dB at over 800 Khz ,
-1dB at 1.2 MHz
However for a tube amp with source resistance of 3Kohm ( like a single tube preamp ) and an input impedance of 1 Meg ohm and volume pot of 250K ohms , the response is as follows.
-0.1 dB at 23 khz ,
-0.5 dB at 52 Khz ,
- 1 dB at 77 Khz.
( All calculated values and cross checked with simulation )
HOWEVER , using different cables could sound very different even though they don't seem to have any significant effect at 20 Khz on steady state continuous signals! Dynamic performance is still not fully understood as you can see on the various cable comparison threads.It doesn't seem to be the capacitance which is the culprit. More to do with dielectric ( storage?) performance on dynamic signals.
So with solid state amps the possibility of cable capacitance causing a major problem appears lower than with a tube amp using high impedance pots.
We have not accounted for stray capacitance which could reach significant values.One reason why point to point wiring appears to sound better than pcb mounted components in tube circuits. The 47Lab chip amp with PP wiring also supposedly did sound very good. I have never seen anyone comparing it with a pcb based design which by themselves are quite good.
If anyone wants to plug in their own values of capacitance and loads I can give you the XLS file that I made. There is no calculation for 20 KHz exactly but I can add it if you want.
Some other interesting data.
Typical capacitance of RCA connectors .
The simplest chromed chassis mount type was 2.8 pF (MX) and the gold plated one with insulators for chassis mounting was 5.7 pF ( Jap) !
Chetan Cables single core shielded cable was 150 pF per meter. ( 5mm dia)
No name balanced cable had 150 pF/m to shield and 76 pF between the two signal cables.
A Srexact balanced cable ( from SP road Bangalore) was 95 pF /m from signal line to shield and 54 pF between signal cables.
Typically I find most balanced cables are around 120 pF/m between signal and shield. However the plastic and quality of the shield used varied widely and so would the sound I guess.
RCA plugs.
MX silver type 4.1pF
MX chromed case/gold connector end ( expensive ) 2.9 pF
So I guess IC cables better be picked after listening tests rather than selecting by looking at their specs.
Last edited:
