Evening All,
decided to post here as I've an F5 and ACA, both perform their best with 4 ~ 8 ohm loads.
for argument sake, lets assume I'm using either amp and my speakers are 8 ohm.
in the perfect world, if I could tap the amp post's directly onto the speaker post's, meaning I don't use speaker cables, then cable losses are eliminated. However as soon I use speaker cables the system is no longer a perfect amp speaker match correct, particularly if my cables are high impedance, say up around 60 ohm ?!?
If however, the speaker cable was a perfect 1 ohm impedance, am I correct in saying losses would be at their minimum and I'm effectively nearly in the same boat as not using cables at all? In other words, 1 ohm cable and 8 ohm speakers would see 9 ohm load at the amp?
For this scenario, lets assume the speaker cable capacitance and inductance are average values for speaker cables
Thanks AL
decided to post here as I've an F5 and ACA, both perform their best with 4 ~ 8 ohm loads.
for argument sake, lets assume I'm using either amp and my speakers are 8 ohm.
in the perfect world, if I could tap the amp post's directly onto the speaker post's, meaning I don't use speaker cables, then cable losses are eliminated. However as soon I use speaker cables the system is no longer a perfect amp speaker match correct, particularly if my cables are high impedance, say up around 60 ohm ?!?
If however, the speaker cable was a perfect 1 ohm impedance, am I correct in saying losses would be at their minimum and I'm effectively nearly in the same boat as not using cables at all? In other words, 1 ohm cable and 8 ohm speakers would see 9 ohm load at the amp?
For this scenario, lets assume the speaker cable capacitance and inductance are average values for speaker cables
Thanks AL
Last edited:
1) WHY would your cables be 60 ohm???
Even hair thin ones won´t go that high at normal speaker to amp distances.
2) why do you consider 1 ohm a perfect speaker cable impedance?
3 meters/10 feet of parallel 18ga wire show 0.13 ohm
4 mm² speaker cable have about 0,005 Ohm / Meter
Pretty close to perfect world 😉
Nothing to be worry about.
But if you are after all worried about the way between your amp and your speakers you have to consider electric fields
Cosmic interference radiation and maybe need something like this:
https://www.in-akustik.de/en/cables...ache=1&cHash=8c4edb032c0ff346ee247a515097cae5
And 10mm² speaker wires --> 0,002 Ohm / Meter 🤏
Pretty close to perfect world 😉
Nothing to be worry about.
But if you are after all worried about the way between your amp and your speakers you have to consider electric fields
Cosmic interference radiation and maybe need something like this:
https://www.in-akustik.de/en/cables...ache=1&cHash=8c4edb032c0ff346ee247a515097cae5
And 10mm² speaker wires --> 0,002 Ohm / Meter 🤏
It's possible for a typical parallel pair of conductors that might be used for aloud speaker cable to have a 'characteristic impedance' of 50Ω - 150Ω. However the characteristic impedance is not at all relevant to cable loss for a loudspeaker cable of any realistic length at audio frequencies. As JM pointed out above the DC resistance is what is relevant, and that would normally be a small fraction of 1Ω.
Characteristic Impedance of Transmission Lines
Characteristic Impedance of Transmission Lines
http://k9yc.com/TransLines-LowFreq.pdf
The summary is at the end of the paper. For cable length less than about 2000 ft the resistance and capacitance of the cable dominate.
The summary is at the end of the paper. For cable length less than about 2000 ft the resistance and capacitance of the cable dominate.
Don't confuse the transmission line impedance with resistance. The wavelength of a 20KHz electrical signal is about 15Km, so anything less than 1/4 of that is essential a direct connection. Can I assume your listening room is less than 3.75Km wide?
The stability of your amplifier may be compromised by the capacitance and/or inductance of the speaker cables. A very effective way to reduce cable inductance is to cross wire a 4-conductor cable. However, a bit of series inductance may improve the response of your speakers by preventing voltage drive at high frequencies which aggravates series resonances.
The stability of your amplifier may be compromised by the capacitance and/or inductance of the speaker cables. A very effective way to reduce cable inductance is to cross wire a 4-conductor cable. However, a bit of series inductance may improve the response of your speakers by preventing voltage drive at high frequencies which aggravates series resonances.
I have a pair of Kimber cables that have a characteristic impedance of ~60ohms, measured using a VNA and verified mathmatically. I'm travelling OS at the moment and don't have the link with me but I came across a YouTube video where a cable manufacture used several 50 ohm coax cables in parallel to reduce this down to just under 10 ohm. He said a lot of his clients raved about the cable, so I made a pair up and concur they without doubt improved the sonic character of the system... it seemed happier whereby everything it had to do was done with very little effort, almost like it just got over the flu and could finally breath again.
I wired the conductors in parallel.. centre conductors together and same with the shields. Going to try the cross connected method next
... just trying to put some theory behind it
Last edited:
^ so I really should be comparing the parallelled coax capacitance and inductance against the Kimbers and forget about characteristic impedance
The characteristic impedance of the cable in this application is irrelevant. In fact at audio frequencies the characteristic impedance of the Kimber cables will be substantially higher that 60 ohms, unless of course Ray Kimber has been granted an exemption from the laws of physics (hint: he hasn't). At around 20Hz the cable will have a characteristic impedance >40kΩ falling to around 120Ω at 20kHz; in fact the characteristic falls asymptotically towards 60Ω as the frequency approaches 1MHz.
That transmission line behaviour is irrelevant in loudspeaker cables is fortunate indeed, or else the mismatch between amplifier output impedance and loudspeaker cable characteristic impedance would mean that very little amplifier power would make it to the loudspeaker, even if the cable's characteristic impedance matched that of the loudspeaker. In the case that the amplifier output impedance is matched to the characteristic impedance of the cable the voltage delivered to the load would be halved, therefore the power delivered to the load would be reduced by 75% and the power dissipated in the amplifier output stage would be doubled, neither being a desirable outcome.
So what is happening when cables are paralleled? Of course the lumped electrical properties are paralleled, so for example two cables paralleled will have half the series resistance, half the series inductance and twice the shunt capacitance (and half the characteristic impedance or around 20kΩ at 20Hz for the Kimber 60Ω cable). Will that change the sound? A: Quite likely yes. Is the change due to a change in the characteristic impedance of the lumped cable conductors? A: No, it is not.
Yes.
I don’t know about the OP’s situation but this trick has been used in the past. A particualr example was Fostex showing one of their horns that really wants a high Rout amplifier, butthe Lux they were using was low Rout so they used Tungsten speaker wires with very high resistance to compensate.
dave
And why would these Fostex horns need a high output impedance amp, other than to compensate fundamental design flaws in the low end?
No extra sellers involved actually. The exact circumstances, for the sake of interest, went like this. Back in 2000 or 2001 Fostex were doing a couple of US shows with Accuphase, demonstrating their then new FE208e∑ wideband. The drivers & the back-horn in question were designed (I know one of the designers, he confirmed when I was working with him in HK in '17) assuming they would be used with amplifiers of relatively high output impedance, SETs being all the rage for the target market at that time. Remember, this is not mainstream hi-fi we're dealing with, but specialist interest products sold primarily to a particular audience in, predominantly, their home market.
This being the case, they had a bit of a mismatch, since the amps were a couple of Accuphase's big top-end monoblocks -fine for regular systems, not so good for speakers of this type. So, rather than farming anything out, or simply shoving a big power-resistor in circuit to get the desired alignment, the lead driver designer decided to build a complete, highly resistive wire-set to go with it. Remember, this is Fostex, and Japan -they have their own ways of working, and playing about with materials etc. is rather more common to them, especially around that time, than to many others. The wire set was tungsten, lead-shielded. Full lot -speaker wire, interconnects and power leads (!) presumably 'just because' / to try it in the latter case. I gather a power lead weighed in at about 60lbs. These were never sold, or a product -just a somewhat left-field solution, at least in the case of the speaker wires, to a practical mismatch / 'problem'. And distributed series R is arguably more interesting. You could call it a more elegant solution than just sticking a resistor in there, but given the weight and lead shielding, I'm not entirely convinced this particular application was quite that. 😉
This being the case, they had a bit of a mismatch, since the amps were a couple of Accuphase's big top-end monoblocks -fine for regular systems, not so good for speakers of this type. So, rather than farming anything out, or simply shoving a big power-resistor in circuit to get the desired alignment, the lead driver designer decided to build a complete, highly resistive wire-set to go with it. Remember, this is Fostex, and Japan -they have their own ways of working, and playing about with materials etc. is rather more common to them, especially around that time, than to many others. The wire set was tungsten, lead-shielded. Full lot -speaker wire, interconnects and power leads (!) presumably 'just because' / to try it in the latter case. I gather a power lead weighed in at about 60lbs. These were never sold, or a product -just a somewhat left-field solution, at least in the case of the speaker wires, to a practical mismatch / 'problem'. And distributed series R is arguably more interesting. You could call it a more elegant solution than just sticking a resistor in there, but given the weight and lead shielding, I'm not entirely convinced this particular application was quite that. 😉
Last edited:
Thanks for correcting the details Scott. It was Scott’s story that is was trying to convey.
dave
dave
Tungsten has about 3 times the resistance of copper. I suspect Fostex's objective was to raise the source impedance for their loudspeaker system from a fraction of one ohm to an ohm or two. The reason they wanted to do that has nothing to do with the characteristic impedance of the cable.
No, it had nothing whatsoever to do with characteristic impedance, and I didn't say (or imply) it did. My apologies if I wasn't clear on that point. They were simply compensating for the fact that, as noted, the speakers and drive unit were designed to be used with amplifiers that had a relatively high output impedance (between about 2.5 - 4ohms) and in the absence of that with the Accuphase amplifiers used for the shows, were adding equivalent series R to bring the alignment to what was originally intended.