Yes, that makes sense. I'd forgotten the return path, so it comes out double!
You're not seriously suggesting 0.5m speaker cables though, are you? Sound a bit MONO, wouldn't it?
I have twisted single pair to treble driver, double to mid driver and triple to bass driver.
The total length from amplifier to driver is ~700mm. The gap between the amplifier/s and the speaker tower is about 20mm.
I can afford that amount of resistance in the speaker driver circuits.
The total length from amplifier to driver is ~700mm. The gap between the amplifier/s and the speaker tower is about 20mm.
I can afford that amount of resistance in the speaker driver circuits.
Last edited:
Hangover from turntable days, but I always like to keep speakers away from the rest to avoid microphonic effects. Even a CD player isn't going to appreciate being rattled by loud bass! Someone I know goes to the extent of floating his electronics on compressed air.
On that stability subject, it is evident that a high impedance load at HF will cause a bit of lift in the amplifier top-end response, perhaps approaching instability. I have noticed that some loudspeakers put a small 50 ohm resistor across the tweeter. I should think that is to avoid that issue. The amp's Zobel network should cover it too. But, yes, it makes you wonder if most of the perceived cable effects are to do with top-end amplifier lift. Gets quite complicated to figure out really.
So with all this cable talk by people who know a lot more about these things than I do I have anything to worry about regarding my speaker cables?
The manufacturer states <4.9Ohm/km resistance and <56pF/m capacitance but no info regarding inductance.
It is 8x4mm^2 for convenience as I use 4way active speakers and is about 5m in length.
Thanks!
The manufacturer states <4.9Ohm/km resistance and <56pF/m capacitance but no info regarding inductance.
It is 8x4mm^2 for convenience as I use 4way active speakers and is about 5m in length.
Thanks!
You can estimate cable inductance for a twin cable as follows:
1. Find or measure the diameter and spacing (centre-centre) of the wires.
2. Using the formula, or an on-line calculator, get the RF characteristic impedance Z - typically 20-100 ohms for speaker cables. (Note: ignore the dielectric as it won't affect the inductance so say the cable is air-spaced)
3. Then L, in henries per metre, is Z/c where c is the speed of light = 3 x 10^8 m/s
1. Find or measure the diameter and spacing (centre-centre) of the wires.
2. Using the formula, or an on-line calculator, get the RF characteristic impedance Z - typically 20-100 ohms for speaker cables. (Note: ignore the dielectric as it won't affect the inductance so say the cable is air-spaced)
3. Then L, in henries per metre, is Z/c where c is the speed of light = 3 x 10^8 m/s
Hmm, not sure you can ignore the dielectric in calculating the characteristic impedance there, DF96.
The calculator I've been using seems a bit ropey at certain conductor ratios, but near enough no doubt:
capacitane/inductance calculations
FWIW, the DC and LF correction that some people mentioned due to internal inductance is a tiny and constant 0.1 uH/m for standard twin lead. While the characteristic impedance is calculated at HF and you ignore it. So, I'll stick with HF characteristic impedance, because it's only about 10% correction at max anyway. I expect it's something that 50Hz mains engineers need to know.
Calculating Cable Inductance of Zip Cord -- Reviews and News from Audioholics
The point about characteristic impedance (and why it interests me) is that if the load and source is close in value to it, you can ignore cable capacitance and inductive effects anyway. Of course, for most speakers at HF and voltage amps, they're NOT! Which is a fun problem with lots of little clever tweaks as solutions.
The calculator I've been using seems a bit ropey at certain conductor ratios, but near enough no doubt:
capacitane/inductance calculations
FWIW, the DC and LF correction that some people mentioned due to internal inductance is a tiny and constant 0.1 uH/m for standard twin lead. While the characteristic impedance is calculated at HF and you ignore it. So, I'll stick with HF characteristic impedance, because it's only about 10% correction at max anyway. I expect it's something that 50Hz mains engineers need to know.
Calculating Cable Inductance of Zip Cord -- Reviews and News from Audioholics
The point about characteristic impedance (and why it interests me) is that if the load and source is close in value to it, you can ignore cable capacitance and inductive effects anyway. Of course, for most speakers at HF and voltage amps, they're NOT! Which is a fun problem with lots of little clever tweaks as solutions.
As I said, the dielectric will not affect the inductance. If you want to include the dielectric then the procedure is much harder:
1. As well as measuring the wires, you also need to know the effective permittivity of a mixture of air and insulation. To do this you will probably need to run an RF modelling package.
2. Find the characteristic impedance Z including the dielectric.
3. Calculate or measure the RF signal propagation speed v in the cable, or divide c by sqrt(effective permittivity) then L= Z/v.
I think my original method is easier to use.
Characteristic impedance is not helpful for audio, because of two things:
1. a Z ohm cable is not Z ohms at audio frequencies, but is reactive and frequency-dependent. Note I am not saying that the impedance is reactive (although it is) but the characteristic impedance is reactive with probably equal amounts of resistance and capacitance.
2. a Z ohm speaker is probably not Z ohms at audio frequencies, but will vary especially around the bass resonance.
3. Almost all amps have lowish output impedance, in order to reduce frequency bumps caused by speaker impedance variations.
So for all cables, most speakers, and virtually all amps the concept is not helpful for audio.
1. As well as measuring the wires, you also need to know the effective permittivity of a mixture of air and insulation. To do this you will probably need to run an RF modelling package.
2. Find the characteristic impedance Z including the dielectric.
3. Calculate or measure the RF signal propagation speed v in the cable, or divide c by sqrt(effective permittivity) then L= Z/v.
I think my original method is easier to use.
Characteristic impedance is not helpful for audio, because of two things:
1. a Z ohm cable is not Z ohms at audio frequencies, but is reactive and frequency-dependent. Note I am not saying that the impedance is reactive (although it is) but the characteristic impedance is reactive with probably equal amounts of resistance and capacitance.
2. a Z ohm speaker is probably not Z ohms at audio frequencies, but will vary especially around the bass resonance.
3. Almost all amps have lowish output impedance, in order to reduce frequency bumps caused by speaker impedance variations.
So for all cables, most speakers, and virtually all amps the concept is not helpful for audio.
We'll have to agree to disagree whether cable characteristic impedance has relevance to audio circuits. To me, it's looking like speaker cabling has audio effects for reasons more to do with high frequency (treble) loading affecting amplifier frequency response than anything.
The BBC engineering department had a hobby-horse in designing for flat loudspeaker impedance. The sweet little LS3/5A monitor had a Zobel network on the KEF T27 tweeter, which does this sort of thing to impedance:
Morgan Jones explains how it's done here:
http://www.diyaudio.com/forums/diyaudio-com-articles/158899-arpeggio-loudspeaker.html
Bringing this back to damping, you could say ALL the resistive loading for damping is built into the speaker in BBC designs. Both for treble end and, mechanically, for the bass resonance. Not in the amp. Clever stuff.
The BBC engineering department had a hobby-horse in designing for flat loudspeaker impedance. The sweet little LS3/5A monitor had a Zobel network on the KEF T27 tweeter, which does this sort of thing to impedance:
An externally hosted image should be here but it was not working when we last tested it.
Morgan Jones explains how it's done here:
http://www.diyaudio.com/forums/diyaudio-com-articles/158899-arpeggio-loudspeaker.html
Bringing this back to damping, you could say ALL the resistive loading for damping is built into the speaker in BBC designs. Both for treble end and, mechanically, for the bass resonance. Not in the amp. Clever stuff.
I'm not sure everyone here has grasped the complete picture. Most manufacturers test and define their DF with resistive loads under test bench conditions. Large numbers impress and with lots of system negative feedback largish numbers can be produced easily. It doesn't mean it sounds good, and often high DF systems using excessive feedback to reduce steady state distortion often suffer from transient response issues.
Loudspeakers are reactive, the effective loop inductance a complex result of device geometry, amplifier phase margin and connection series resistance, not to mention any passive crossovers. While the DC resistance of a crossover and associated wiring may well be milliohms, to be effective as a crossover it needs significant impedance at the crossover frequency. Cored inductors may be smaller with lower resistance but inductance is inductance regardless of the construction.
The net complex impedance from all this can range from 0.1ohms up to many ohms at crossover frequency. Consequently effective DF can drop off to near nothing.
Whether there is "less bass" or "more bass" depending of the DF is really a function of the speaker acoustic loading and natural resonance.
I recall a time 30 years ago when I produced an amp for personal use but some colleagues thought it was pretty good and it ended up in the local audiophile store being compared with the latest and greatest. It scored a first equal with the comment "bass a bit tight". If I recall, the output stage used multiple feedback paths with a DF around 50 at 1kHz into a resistive load with exceptional transient response. It was also excellent with reactive loads. I wound up a couple of 300uH chokes and put them in series with the speaker leads. The results were a "warmer sloppier bass" (to my ears) which those listening thought was "perfect". Equivalent DF at 150Hz was around 5.
Loudspeakers are reactive, the effective loop inductance a complex result of device geometry, amplifier phase margin and connection series resistance, not to mention any passive crossovers. While the DC resistance of a crossover and associated wiring may well be milliohms, to be effective as a crossover it needs significant impedance at the crossover frequency. Cored inductors may be smaller with lower resistance but inductance is inductance regardless of the construction.
The net complex impedance from all this can range from 0.1ohms up to many ohms at crossover frequency. Consequently effective DF can drop off to near nothing.
Whether there is "less bass" or "more bass" depending of the DF is really a function of the speaker acoustic loading and natural resonance.
I recall a time 30 years ago when I produced an amp for personal use but some colleagues thought it was pretty good and it ended up in the local audiophile store being compared with the latest and greatest. It scored a first equal with the comment "bass a bit tight". If I recall, the output stage used multiple feedback paths with a DF around 50 at 1kHz into a resistive load with exceptional transient response. It was also excellent with reactive loads. I wound up a couple of 300uH chokes and put them in series with the speaker leads. The results were a "warmer sloppier bass" (to my ears) which those listening thought was "perfect". Equivalent DF at 150Hz was around 5.
I think you may be confusing resistance, which provides electrical damping, with total impedance. Reactance does not provide damping, so the fact that a crossover inductance has high reactance does not affect damping. Away from the bass resonance electrical damping has little effect anyway.otiake said:
It is true that the DC resistance of crossover inductances can limit effective DF. Cable resistance does this too. For these reasons there are diminishing returns as the bare amp DF goes much above 20 or so.
The best bass I've ever heard came from a bjt ss amp with a damping factor of >400.
A question: I understand DF is frequency dependent (some manufacturers state a DF of X at 100Hz, others at 1000Hz), is it possible to guesstimate what it would be at a frequency other than the one chosen by the manufacturers?
In other words does it vary at a roughly predictable rate?
A question: I understand DF is frequency dependent (some manufacturers state a DF of X at 100Hz, others at 1000Hz), is it possible to guesstimate what it would be at a frequency other than the one chosen by the manufacturers?
In other words does it vary at a roughly predictable rate?
Thanks guys!
Only asked because in my youth I read a german HiFi rag and they tested for DF at three frequencies: 100, 1000 and 10 000Hz.
As far as I remember it was always higher the lower the frequency so I was just wondering as it would make comparing specs (for what they are worth) easier.
Only asked because in my youth I read a german HiFi rag and they tested for DF at three frequencies: 100, 1000 and 10 000Hz.
As far as I remember it was always higher the lower the frequency so I was just wondering as it would make comparing specs (for what they are worth) easier.
There's a LOT of different information, experiences, and opinions in the
Audio Industry, aren't there? Well... MY experience, and my opinion, about DF's is the same about every other aspect of this wonderful arena: It's ALL about the synergy of the components involved, the macro components and the micro. Change the amp output filters, 2 lil resistors and a tiny cap: you’ve altered the sound of the entire system. It’s the synergy of combining energies and types.
I personally like the well rounded sound and feel of the bass my tube amp gives my tuned port speakers vs. the bass from my SS amp.
Audio Industry, aren't there? Well... MY experience, and my opinion, about DF's is the same about every other aspect of this wonderful arena: It's ALL about the synergy of the components involved, the macro components and the micro. Change the amp output filters, 2 lil resistors and a tiny cap: you’ve altered the sound of the entire system. It’s the synergy of combining energies and types.
I personally like the well rounded sound and feel of the bass my tube amp gives my tuned port speakers vs. the bass from my SS amp.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.