| AudioREVIEW's Forum Archives - Cables & Interconnects |
|
|   Would you choose long IC&short speaker cable, or short IC... | Tony_Montana
Sep 22, 2003 7:36 AM | | ...and long speaker wire??
Suppose your setup call for either short IC and long speaker wire (>30 feet), or long IC (>30 feet) and short speaker cable. Given using quality IC and speaker wires, which scenario would you be more comfortable with..longer speaker, or longer IC cable length? |
| Long IC of course. | JoeE SP9
Sep 22, 2003 5:44 PM | | It's desirable to keep your speaker cables as short as possible. Keep your power amps adjacent to your speakers and run long IC's from the preamp. |
| What about if you have high impedance RCA output?.... | Tony_Montana
Sep 23, 2003 6:28 AM | | ...Or low quality "high capacitance" cable?
A first-order high-cut (or low-pass) filter is formed by an output's source impedance and the capacitance of the cable. The frequency at which a filter attenuates 3 dB is called its "corner frequency". With short cables (low capacitance) and low output impedances, the corner frequency typically occurs well above the audio band. With longer cables and higher output impedances, the corner frequency drops and may drop into the audio band. The formula for the corner frequency...
F=1/(6.28*C*R)
..where F is the corner frequency in Hz, 6.28 is PI x2, R is the source's output impedance in Ohms, and C is the cable's total capacitance in Farads.
Suppose you have a very high impedance RCA output of 20 k ohms (worst case scenario), and using a very low capacitance cable such as RG-59 which would have a capacitance of 20pf/ft (picofarad per foot).
So going by the formula:
30 feet (IC length) x 20 pf=600 pf (total cable capacitance for 30 feet).
Corner frequency will be: F=1/(6.28*20 k ohm*600 pf)
F=13.26 khz
So as you see from the formula, the 30 feet of IC cable will cause 3 dB drop above 13.26 kHz which is unacceptable. So if one have high RCA output impedance, or not so high quality IC cable (high capacitance), you might want to reconsider running long ICs :) |
| What about if you have high impedance RCA output?.... | mtrycrafts
Sep 23, 2003 11:32 PM | | b Suppose you have a very high impedance RCA output of 20 k ohms (worst case scenario),
Then you have a problem. Stay away from those passive pre amps :) |
| What about if you have high impedance RCA output?.... | Tony_Montana
Sep 24, 2003 4:04 PM | | Hey Mtry
>Stay away from those passive pre amps :)
Or high capacitance cheap cables when lenght is an issue :) |
| What about if you have high impedance RCA output?.... | mtrycrafts
Sep 24, 2003 10:30 PM | | b Or high capacitance cheap cables when lenght is an issue :)
Run the numbers for normal output impedance, a few hundred ohms and input imedance of 100k :) with high cap cables :) |
| What about if you have high impedance RCA output?.... | JoeE SP9
Sep 24, 2003 11:43 AM | | This is usually only a problem with some tube preamps because of their inherently high output impedence. This may also be a problem with passive attenuators in which case you should use a true ladder attenuator. They have a constant output impedence. Why would you use RG-59 as an interconnect? |
| What type of IC would you use? | Tony_Montana
Sep 24, 2003 3:46 PM | | The 20 pf/ft capacitance was very conservative figure. Only the best of coax cables will have that low of capacitance :) |
| What type of IC would you use? | JoeE SP9
Sep 24, 2003 4:55 PM | | I use Belden 9259 with Deltron RCA's. The Belden is sweep tested from 5 to 450Ghz and the Deltron's are very nice RCA's with Teflon insulators. I have very long (20ft.) IC's from my preamp (ARC SP9) but I use a modified "Aunt Corey's" passive preamp as a line buffer (very low output impedence) to drive the long cables. This is not the ideal but I can preserve that glorious ARC tube sound and not be to concerned about high freq. roll off. The Belden is $140 for a 500ft. roll and the Deltrons are $4.99 each. |
| Please 'splain me sumthin..... | FLZapped
Sep 26, 2003 12:01 PM | | i IC's from my preamp (ARC SP9) but I use a modified "Aunt Corey's" passive preamp as a line buffer
Uhm, how can a passive pre-amp act as a buffer, it's passive after all. At 0dB attenuation, it also has 0dB buffering capability.
i (very low output impedence) to drive the long cables.
A "passive pre-amp" that has ouput drive capability? Praytell, how is this?
Or is it not truely passive?
-Bruce |
| Nobody's perfect | JoeE SP9
Sep 26, 2003 8:20 PM | | OK the terminology is not quite accurate. I ment to say buffered. I use only the BUF-03 so that the highish output impedence of my SP9 is not a factor. It is actually a modified version of Corey Greenberg's buffered preamp. |
| What about if you have high impedance RCA output?.... | FLZapped
Sep 26, 2003 11:46 AM | | i Why would you use RG-59 as an interconnect?
You think a prudent manufacturer is going to buy and track in his inventory a seperate cable for audio when RG-59 will function just as well??
-Bruce |
| I think your assumption of output impedence is way off | skeptic
Sep 26, 2003 7:34 AM | | Any good preamp output or output of an electronic device having gain such as a tape head preamp, microphone preamp, phono preamp, tuner output, cd player, etc. deliberately has a low impedence output specifically to avoid this problem. It is almost invariably arranged as a cathode follower output stage if it is tube circuit or an emitter follower output stage if it is solid state. It is usually in the range of 50 ohms, not 20 kohms. This puts the hf 3db down point way beyond the audible range even for high capacitance inexpensive cables and the hf falloff in the audio band well under 0.1 db at 20khz for any reasonable length of wire. In your example, this would put the 3 db down point above 5 megahertz. |
| I think your assumption of output impedence is way off | FLZapped
Sep 26, 2003 11:50 AM | | Somneone posted data here once for a pre-amp that had a measured output impedance of 2000 or so ohms. Doubt it is in the archives any longer since many seem to have been lost, or purged.
Besides, just because it has a low impedance output, does not guarantee that it will drive a high capacitance load well, frequency response aside.
-Bruce |
| I think your assumption of output impedence is way off | skeptic
Sep 26, 2003 1:39 PM | | The calculation is very simple and is applicable for all linear networks. It is a direct application of Thevenin's theorem. The frequency response depends on the equivalent source impedence, network equivalent of the interconnect, and the complex load impedence, and nothing else. When the source impedence is low and the load impedence is high, the network impedence (shunt capacitance and series resistance and inductance) caused by any reasonable interconnecting cable including very cheap and poorly made ones is negligable. A preamplifier output stage or output buffer amplifier with a high (output) source impedence such as 2kohms is a bad design which should be avoided at all times. Generally a poor design in one regard will be a poor design in other regards as well and so would not merit expensive cables in any event even if they were of some benefit. A potentiometer, a handful of jacks, and a switch or two in a box masquerading as a preamplifier or control center or whatever other bogus name someone advertises it as is a very bad idea if for no other reason than it creates this impedence mismatch problem. Buyer beware. |
| I think your assumption of output impedence is way off | skeptic
Sep 26, 2003 2:33 PM | | I've been spot checking a few places for preamp output impedences and power amp input impedences as well as preamp input impedences. For well known reasons, Magnetic phono inputs generally run about 47Kohm or 50 kohms. McIntosh typicaly ran 450 ohms preamp output impedence and 22K input on their power amps. That's a little extreme but still represents a 50:1 ratio which isn't too bad. Dynaco PAS3X a tube preamp ran 1Kohm out looking for a 100k or higher load and supplying 250K at its own high level inputs. Stereo 120 power amp was 100K input. PAT4 preamp had a 600 ohm output looking for a load of at least 10Kohms. Stereo 400 had a 50K input impedence and the PAT5 preamp was rated with a 600 ohm or less output impedence. The inherently wide bandwidth of these units (low internal shunt capacitance) and the high power amp input/preamp output impedence ratios in addition to their absolute values compared to the impedences of interconnect cables attest to their immunity to sensitivity to differences in the impedence of interconnect cables. |
| That example was with really low capacitance cable. | Tony_Montana
Sep 26, 2003 8:37 PM | | The cable which I had used for 3 dB calculation was 20 pf/ft capacitance which only best of coax cables can provide such as Belden 1505F (17 pf/ft). I doubt if off the shelf ordinary IC will have that much of low capacitance due to their slimness. Also we calculate 1 dB corner frequency-instead of 3 dB- to get flat frequency response up to 20 kHz, you will see the cable length might be an issue. |
| I think your assumption of output impedence is way off | FLZapped
Sep 26, 2003 10:04 PM | | i A preamplifier output stage or output buffer amplifier with a high (output) source impedence such as 2kohms is a bad design which should be avoided at all times.
Agreed, but this spec is never published, it is usually uncovered.
-Bruce |
| Long IC of course. | FLZapped
Sep 26, 2003 11:31 AM | | Mr. Engineer, what is your basis for this conclusion??? How do you know if your pre-amp is rated for high capacitance loads? What if your preamp has a relatively high impedance output?
-Bruce |
| Long IC of course. | mtrycrafts
Sep 26, 2003 5:13 PM | | b What if your preamp has a relatively high impedance output?
Buy a better one :) |
| Long IC of course. | FLZapped
Sep 26, 2003 10:06 PM | | Of course, you would have to know this to begin with, the output impedance isn't usually published....I've never seen it....It's usually uncovered through a review process.
-Bruce |
| I have a tube preamp | JoeE SP9
Sep 26, 2003 8:30 PM | | ARC suggested that I not use long IC's because of potential roll off problems. I had the BUF-03's on hand. It was only an hour of work to build the buffer. The IC from my preamp to crossover is 30 ft. |
| I have a tube preamp | skeptic
Sep 27, 2003 9:56 AM | | Manufacturers of separate preamplifiers and power amplifiers expect the two units to be installed within a couple of feet of each other, otherwise, they would take even greater precautions designing the preamp output/power amp input impedence. The sole exceptions in general are powered subwoofers but even there, the output can often come from the power amplifier speaker terminals where the source impedence is a fraction of an ohm (most solid state units.) So cable advocates install their power amps close to the loudspeaker to reduce the length of the speaker wire which was never really a problem and increase the distance between the preamp and the power amp requiring long ICs which might be a problem and then spend a lot of time and money looking for ICs that solve a problem that never should have existed in the first place. Smart, real smart. |
| re: Would you choose long IC&short speaker cable, or short IC... | Norm Strong
Sep 25, 2003 5:31 PM | | 1. Put the speakers where you want them.
2. Put the amplifiers where you want them.
3. Connect everything together.
Any questions? |
| I agree | JoeE SP9
Sep 26, 2003 8:33 PM | | Thats why I have a 30ft. IC from my preamp to my crossover. |
| At least you are using a good quality cable. | Tony_Montana
Sep 26, 2003 9:29 PM | | I checked on Belden 9259 specification which you are using, and it have capacitance rating of 17pf/ft which is pretty good.
But nevertheless, even the editor of Stereophile suggest Buffered Passive Preamplifier will have a harder time driving long cables without losing high-end fidelity.
Copy/Paste follwing link:
"http://www.stereophile.com/showarchives.cgi?54:6" |
| People who look for trouble will usually find it. | skeptic
Sep 26, 2003 7:06 AM | | The best method is to keep the ICs as short as possible. That is where the signal is most vulnerable to induced hum and noise. The power amplifier output is ideal for long runs of cable. It has a very low source impedence especially if it is solid state and the signal level is by far the highest in the system. It is practically immune to induced noise and will rarely if every have it's frequency response compromised unless you find one of the bizzarely conctructed audiophile cables.
Anyone who is stupid enough to buy a so called "passive preamplifier" and then run a long coax to a power amplifier will often get all of the problems they deserve. The cable industry was invented for suckers like these and the owners of these overpriced badly conceived devices will spend endless time and money looking for IC cables that will mitigate the inherent dilemma they have created for themselves. I have no sympathy for them. Shop, shop, shop, spend spend spend. I don't think they will ever wise up and get it. |
| No passive preamp | JoeE SP9
Sep 26, 2003 8:44 PM | | I did not say I had a passive preamp. I have a tube preamp. The manufacturer recommended against a 30ft. IC to my crossover. I mistakenly used pasive in place of buffer. I use only a buffer. 1 BUF-03 per channel. No one in their right mind wants 30ft speaker cables. |
| No passive preamp | Tony_Montana
Sep 26, 2003 9:04 PM | | Hey JoeE
You said;
"No one in their right mind wants 30ft speaker cables."
What about suround speaker's wires? Most of the lenght of wire for suround speakers are over 25 feet long :) |
| No passive preamp | skeptic
Sep 27, 2003 5:15 AM | | I see no problem with 30 foot speaker cables. 16 gage zip cord is probably just fine, 12 gage may be overkill but more than satisfactory and reasonably priced. If cable capacitance for RG-59 or RG-6 were the problem some here have said it is, you would not be able to use it for cable television distribution in your home where dozens of feet carry a signal of many hundreds of megahertz. Significant losses would not only obliterate the color burst subcarrier but the entire video component of the upper band channels as well. Remember, channel 2 starts above 50mhz. Audio which goes from 0 to 20 khz is insignificant bandwidth at the lowest end of the spectrum by comparison. And you can be sure that the output stage of a $25 or $50 video distribution amplifier gets much less quality of design consideration than the output of a $1000 preamplifier. |
| re: Would you choose long IC&short speaker cable, or short I | FLZapped
Sep 26, 2003 11:36 AM | | Longer speaker cables. Some pre-amps may not be able to drive the additional capacitive load and the longer length also makes the system more suceptable to outside interference.
-Bruce |
| Agreed. | Tony_Montana
Sep 26, 2003 8:46 PM | | You said:
"...also makes the system more suceptable to outside interference."
Especially 60 Hz hum if it is running approximately close to power cords :) |
| re: Would you choose long IC&short speaker cable, or short I | JoeE SP9
Sep 26, 2003 8:48 PM | | I have a tube preamp. Tht's why I built the buffer. Excuse my mistake of using passive instead of buffer. I did not want 30ft. speaker cables. |
| |
|
|
