Hi Tinitus,
As said DF does have difference in the sound provided of course that the speaker is optimally designed as well. It would be hard to expect one component to fix a shortcomings of another.
The problem with calculated/simulated is totally dependant on what parameters is available and how one may apply it. Measuring on the other hand is what is, provided one knows how to measure a specific parameter. Different methods are used to measure low/medium or high impedances.
To try and remain with the focus of the thread, there will be audible differences when amplifiers with different output impedance drive a particular speaker design. It does not mean that an indefinitely high DF will be the answer, however a critically damped system would. I therefore support DF96 by saying the Damping Factor and overall system Q is the most determining factor in how the system performance is perceived.
In my humble opinion one should keep the interconnect as the transmission medium rather than bundle it with either speaker or amp and should be as transparent as possible so that looking into the one side it should reveal the speaker only and while looking from the other side it should see only the amp.
I would not like to start any further debate on speaker cable as it would draw this thread out of context completely.
As said DF does have difference in the sound provided of course that the speaker is optimally designed as well. It would be hard to expect one component to fix a shortcomings of another.
The problem with calculated/simulated is totally dependant on what parameters is available and how one may apply it. Measuring on the other hand is what is, provided one knows how to measure a specific parameter. Different methods are used to measure low/medium or high impedances.
To try and remain with the focus of the thread, there will be audible differences when amplifiers with different output impedance drive a particular speaker design. It does not mean that an indefinitely high DF will be the answer, however a critically damped system would. I therefore support DF96 by saying the Damping Factor and overall system Q is the most determining factor in how the system performance is perceived.
In my humble opinion one should keep the interconnect as the transmission medium rather than bundle it with either speaker or amp and should be as transparent as possible so that looking into the one side it should reveal the speaker only and while looking from the other side it should see only the amp.
I would not like to start any further debate on speaker cable as it would draw this thread out of context completely.
@ Nico.
IIRC from my submarine cable engineering days, all cables made of decent metal and insulator work pretty much the same. What a lot of people don't know is that a cable has a Characteristic Impedance. Phono cables are around 100 ohms, speaker cables (one hopes) are around 8 ohms. If a cable is terminated in its characteristic impedance it is essentially invisible and transparent. If the end is terminated in a high resistance, it shows up a capacitance due to the mismatch. If it is shorted, it looks like an inductance.
This means your source and load impedance on an 8 ohm cable should both ideally be 8 ohms. Otherwise you will get some unpleasant reactive effects at high frequency. Which is why voltage amplifiers successfully use an 8 ohm RC Zobel network on their outputs to avoid HF oscillation.
Back to damping. Infinite damping is evidently not essential in loudspeaker design. You can adjust the box size and filter for a driver to get maximally flat frequency response for various source impedance and driver Q as Morgan Jones does in his masterclass article. How very "BBC" to get flat impedance and low colouration in a loudspeaker! It's a neutral sound a lot of us grew up with.
IIRC from my submarine cable engineering days, all cables made of decent metal and insulator work pretty much the same. What a lot of people don't know is that a cable has a Characteristic Impedance. Phono cables are around 100 ohms, speaker cables (one hopes) are around 8 ohms. If a cable is terminated in its characteristic impedance it is essentially invisible and transparent. If the end is terminated in a high resistance, it shows up a capacitance due to the mismatch. If it is shorted, it looks like an inductance.
This means your source and load impedance on an 8 ohm cable should both ideally be 8 ohms. Otherwise you will get some unpleasant reactive effects at high frequency. Which is why voltage amplifiers successfully use an 8 ohm RC Zobel network on their outputs to avoid HF oscillation.
Back to damping. Infinite damping is evidently not essential in loudspeaker design. You can adjust the box size and filter for a driver to get maximally flat frequency response for various source impedance and driver Q as Morgan Jones does in his masterclass article. How very "BBC" to get flat impedance and low colouration in a loudspeaker! It's a neutral sound a lot of us grew up with.
8 ohm at ANY frequency. Listen and learn. We are talking CHARACTERISTIC IMPEDANCE OF CABLE here. Aerial coax is 75 ohm unbalanced, IIRC LAN wiring is 50 ohm coax, or FM aerials often 300 ohm balanced. ProAudio uses 600 ohm balanced cable. RF engineers need to know this stuff.
Last edited:
When the voltage changes in a time interval comparable to the time it takes for the signal to travel down the wire, the length becomes important and the wire must be treated as a transmission line. Stated another way, the length of the wire is important when the signal includes frequency components with corresponding wavelengths comparable to or less than the length of the wire.
A common rule of thumb is that the cable or wire should be treated as a transmission line if the length is greater than 1/10 of the wavelength. At this length the phase delay and the interference of any reflections on the line become important and can lead to unpredictable behavior in systems which have not been carefully designed using transmission line theory.
Does this describe audio?
I think not.
"8 ohm at ANY frequency"
See above.
"RF engineers need to know this stuff. "
I work with L-band myself, it tends more towards microstrip and stripline.
"ProAudio uses 600 ohm balanced cable"
Hasn't for many years, and never in the sense that you seem to mean.
600Ω ladder line (RF):
Current practice is to drive from a low source impedance into a high impedance load. Typical source impedances are on the order of 100Ω (or less) and load impedances on the order of 5KΩ (or higher).
A common rule of thumb is that the cable or wire should be treated as a transmission line if the length is greater than 1/10 of the wavelength. At this length the phase delay and the interference of any reflections on the line become important and can lead to unpredictable behavior in systems which have not been carefully designed using transmission line theory.
Does this describe audio?
I think not.
"8 ohm at ANY frequency"
See above.
"RF engineers need to know this stuff. "
I work with L-band myself, it tends more towards microstrip and stripline.
"ProAudio uses 600 ohm balanced cable"
Hasn't for many years, and never in the sense that you seem to mean.
600Ω ladder line (RF):
Current practice is to drive from a low source impedance into a high impedance load. Typical source impedances are on the order of 100Ω (or less) and load impedances on the order of 5KΩ (or higher).
speaker cable would not have any "characteristic impedance" unless it's a significant fraction (generally considered to be between 1/10 and 1/8) of a wavelength at the frequency of operation. the idea that any cable has a "characteristic impedance" at audio frequencies is a total misunderstanding and mis-application of the term. you are not trying to radiate audio frequency "radio" waves using an antenna (at 30hz, the antenna would have to be 5 million meters long for a dipole, 2.5 million meters for a 1/4 wave vertical). 1/10 wavelength would be 1 million meters (1000km), and would be the length of cable required before speaker wire even BEGINS to act like a transmission line. so let's NOT treat any normal length cable as a lumped-constant transmission line... it just doesn't behave as such.
in fact, even if you use 50 ohm coax for audio, it does not behave like a transmission line, but a plain shielded cable. there just is not enough length to begin to exhibit the resonant properties that make it a transmission line at radio frequencies.
in fact, even if you use 50 ohm coax for audio, it does not behave like a transmission line, but a plain shielded cable. there just is not enough length to begin to exhibit the resonant properties that make it a transmission line at radio frequencies.
Last edited:
OK, let me give you an example of cable mismatch not too far back in the history of audio. 1/2 a meter of 100 ohm phono cable between a moving magnet record cartridge and the standard 47k load of an amplifier input.
If you know your transmission lines and characteristic impedance, you can calculate a capacitance of around 200 pF. That combined with the horrid ca. 500 mH inductance of a MM cartridge produced a high resonance around 15 kHz.
This was considered a GOOD THING by most people, because it equalised the HF output up to 20 kHz. Alas, with a second order characteristic, it tended to make the stylus move in the opposite direction to the groove (180 degree phase), which IMO caused an unpleasant mistracking which was why they sounded rough.
Turntable Forum • View topic - Cartridge loading explained
Low output moving coil cartridges worked into a matched 100 ohm load through a 100 ohm cable. They sounded MUCH better because they had a mere 1st. order (90 degree phase) filter response. In fact it's quite incredible that we terminate 100 ohm cables in 47k in audio. Make no mistake, 47k to 100 ohm is a HUGE mismatch.
Everybody who has experimented knows that some speaker cables sound better than others. My own view is that characteristic impedance is the thing that makes sense of this. Nothing to do with gold or copper. Just simple capacitance and inductance per unit length.
If you know your transmission lines and characteristic impedance, you can calculate a capacitance of around 200 pF. That combined with the horrid ca. 500 mH inductance of a MM cartridge produced a high resonance around 15 kHz.
This was considered a GOOD THING by most people, because it equalised the HF output up to 20 kHz. Alas, with a second order characteristic, it tended to make the stylus move in the opposite direction to the groove (180 degree phase), which IMO caused an unpleasant mistracking which was why they sounded rough.
Turntable Forum • View topic - Cartridge loading explained
Low output moving coil cartridges worked into a matched 100 ohm load through a 100 ohm cable. They sounded MUCH better because they had a mere 1st. order (90 degree phase) filter response. In fact it's quite incredible that we terminate 100 ohm cables in 47k in audio. Make no mistake, 47k to 100 ohm is a HUGE mismatch.
Everybody who has experimented knows that some speaker cables sound better than others. My own view is that characteristic impedance is the thing that makes sense of this. Nothing to do with gold or copper. Just simple capacitance and inductance per unit length.
I WAS trying to, but unclejed613 and djk WILL argue about established matters of cable characteristics!interesting, but please try and respect the topic(thread title)
Let's summarise this baby and put it to bed. Damping factor isn't all that important in itself. You can design a speaker and connecting cable to work well with almost any amplifier topology, damping and related output impedance. Just a matter of understanding the Thiele-Small parameters for a loudspeaker driver.
I'd admit my analysis of valves versus solid-state is a bit shaky, and I really don't know if one is better than the other. So I'd say, have fun with your HiFi. When it's good, it's very good. Love and Peace, all.
Can I get away with saying that at audio frequencies it is very unlikely that an '8 ohm' cable has a characteristic impedance anything like 8 ohms. This is because for most of the audio band resistance dominates over inductance, so the characteristic impedance is high and capacitive. Anyone who talks about 'matching' an 8 ohm cable to an 8 ohm speaker is probably demonstrating how little he knows about both transmission lines and speakers. Submarine cables may be different, because they are deliberately designed to have low dispersion, probably by adding inductance in a distributed or lumped form.
Not quite on topic, but closer than cartridges!
Not quite on topic, but closer than cartridges!
Last edited:
I just used this cable calculator to do some rough and ready checks on cable characteristic impedance:
capacitane/inductance calculations
To my surprise, it's very difficult to make thin or fat twin-lead speaker wire with an impedance of anything other than about 70 ohms, which means only the inductive component can have any effect on 8 ohm speakers. Yup, I was wrong, no such thing as 8 ohm cable! Sorry.
A typical Morel 1" dome tweeter has an inductance of 0.07 mH, and a 8" Fostex Full range FF225WK driver 0.068 mH. A meter of cable has about 1uH max. So, yes, it's hard to see a significant effect on damping. Can't be more than 10% even with long cables.
Roger Russell of McIntosh opines that even 10 feet of hugely flimsy 24 guage speaker wire (remember that DIN stuff they used to fit to music centres?) shouldn't have anything more than 5% resistive effect on damping of 8 ohm speakers. Negligable, in other words.
Speaker Wire
I must try this. I'll hard solder it to some speaker plugs to get a good contact though.
capacitane/inductance calculations
To my surprise, it's very difficult to make thin or fat twin-lead speaker wire with an impedance of anything other than about 70 ohms, which means only the inductive component can have any effect on 8 ohm speakers. Yup, I was wrong, no such thing as 8 ohm cable! Sorry.
A typical Morel 1" dome tweeter has an inductance of 0.07 mH, and a 8" Fostex Full range FF225WK driver 0.068 mH. A meter of cable has about 1uH max. So, yes, it's hard to see a significant effect on damping. Can't be more than 10% even with long cables.
Roger Russell of McIntosh opines that even 10 feet of hugely flimsy 24 guage speaker wire (remember that DIN stuff they used to fit to music centres?) shouldn't have anything more than 5% resistive effect on damping of 8 ohm speakers. Negligable, in other words.
Speaker Wire
I must try this. I'll hard solder it to some speaker plugs to get a good contact though.
Last edited:
Roger Russell of McIntosh opines that even 10 feet of hugely flimsy 24 guage speaker wire (remember that DIN stuff they used to fit to music centres?) shouldn't have anything more than 5% resistive effect on damping of 8 ohm speakers. Negligable, in other words.
My speaker cables are 24g solid copper. They work VERY well in my system.
dave
That cable calculator only calculates the HF characteristic impedance so not very useful for audio. At audio frequencies the characteristic impedance is high and frequency dependent, while the propagation speed is lower than normal and gets complicated by dispersion.
Z = sqrt( (R + jwL ) / (G + jwC ) )
For RF L and C dominate, so we get the simplified formula Z = sqrt( L/C ).
For audio R dominates over L (roughly - except top few octaves) so Z = sqrt( R/jwC ) - I guess you pick the root with positive real part. Perhaps slight cheat here, as you no longer get wave propagation but diffusion through distributed RC network.
Near DC R and G dominate so it becomes Z = sqrt ( R/G ) - resistive again but very small and probably meaningless as you just have a tiny attenuator.
Z = sqrt( (R + jwL ) / (G + jwC ) )
For RF L and C dominate, so we get the simplified formula Z = sqrt( L/C ).
For audio R dominates over L (roughly - except top few octaves) so Z = sqrt( R/jwC ) - I guess you pick the root with positive real part. Perhaps slight cheat here, as you no longer get wave propagation but diffusion through distributed RC network.
Near DC R and G dominate so it becomes Z = sqrt ( R/G ) - resistive again but very small and probably meaningless as you just have a tiny attenuator.
FWIW, I checked out planet10's single strand 24 AWG cables, and got 0.1 ohm per meter. That looks promising.
American wire gauge - Wikipedia, the free encyclopedia
Equivalent to puny 7X 0.2 mm hook up cable:
Miniature Loudspeaker Cable : Speaker Cable : Maplin Electronics
This twin strand 3A lighting cable is 16X 0.2mm, so is also a cheapie candidate.
Oval 2-Core 3A Mains Cable 2192Y : Mains Wiring Cable : Maplin Electronics
I also looked at very exotic Nordost cables and my own thickish Richer Sounds stuff, and to my surprise the characteristic impedance (root L/C) was really quite high around 150 ohms. The cheapie cables do better around 75 ohms because the conductors are quite close together. I know, it doesn't matter, but I still like the idea of getting lower.
American wire gauge - Wikipedia, the free encyclopedia
Equivalent to puny 7X 0.2 mm hook up cable:
Miniature Loudspeaker Cable : Speaker Cable : Maplin Electronics
This twin strand 3A lighting cable is 16X 0.2mm, so is also a cheapie candidate.
Oval 2-Core 3A Mains Cable 2192Y : Mains Wiring Cable : Maplin Electronics
I also looked at very exotic Nordost cables and my own thickish Richer Sounds stuff, and to my surprise the characteristic impedance (root L/C) was really quite high around 150 ohms. The cheapie cables do better around 75 ohms because the conductors are quite close together. I know, it doesn't matter, but I still like the idea of getting lower.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.