Speaker length vs. I/C length


I've read that it is better to have longer I/C's vs. longer speaker cables. Is this true? Speakers have a higher level signal, wouldn't this be better for a longer wire? My greatest distance will only be about 6-8 feet.
Thanks,
MaxCast
maxcast
So long as a system has:

1) rock solid connections at all points that are cleaned on a somewhat regular basis

2) cable that does not saturate under load ( i.e. too small of a gauge )

3) cable that doesn't introduces any type of series resistance to the circuit

4) cable that is not highly reactive ( inductive or capacitive )

everything should be fine and pretty straightforward to deal with.

Just as the situation gets more "specialized" or "out of the ordinary", so do the installation requirements. There are some circumstances that a "less than stellar" set of speaker cables can actually out-perform "hi-end" cables, etc.. These typically fall into the "high reactance" /"freak" speaker load category though. Having to resort to this type of installation though is a sign to me that something, probably the amp, is not up to dealing with the load ( speakers ) that it sees and the "junky" cables act as a "buffer".

The biggest problem with people trying to use "heavy wire" to keep resistance down to a minimum and pass a maximum amount of information is that they typically resort to some type of "monster sized" zip cord or other cables with high inductance. High inductance speaker cables resort in rolled off upper mids and treble, lack of detail, loss of air, weak harmonic structure, etc...

Since longer cables would have higher levels of inductance, increased length would only compound the problems that i mentioned. In a situation such as that described, there would be obvious benefits to keeping the speaker cables ( and the inductance ) as short and as low as possible.

In order to avoid that type of pitfall, folks should look for speaker cables that are of a heavy gauge but make use of "fancy geometries" in order to minimize reactance. Another "side benefit" to these "fancy geometries" is that they are typically less prone to acting like an antenna, picking up emi / rfi, etc... Obviously some geometries are going to work better than others and the trick is knowing which is which. I've seen MANY factory designs that are operating WAY below optimum. Reconfiguring the cables and reterminating will typically show marked performance improvements that are audible as well as measurable. I'm not naming names though as i don't need any more hate email or threats regarding lawsuits.

As to your specific installation Ernie, I see no problem with how you set things up so long as the amps are not tucked up into the rafters. They obviously need some room above them to help dissipate the heat / let the heat rise. The fact that you are running balanced negates much of the negativity of my comments ( is that a double negative then ?? ). This is proven by the fact that you have minimal noise even though you are running in close proximity to known sources of emi and rfi. If you would have used RCA's in that installation, i would venture to say that you would have had much poorer results.

On that note, i have to add that not every installation or system can be done "optimally". I myself am running long ic's ( 12 foot ) and short speaker cables ( 4 foot of 7 gauge ). As you might have guessed, i did this because i'm running monoblocks and have them perched behind the speakers. To top it off, the ic's are single ended ( twisted pair of wire with RCA's ) and NOT balanced. The preamp that i chose to use there is fully capable of driving VERY long cables though, as i checked into that right off the bat. Unfortunatly, it does not have balanced outputs or i would be using them. Like anything else though, it is a balancing act. I feel that i have lost some detail by using the longer IC's but i've gained better control of the driver courtesy of the shorter speaker cables. Such is life.... Sean
>
Since I am only working with 6 feet or so I think I will use what works for the configuration I choose to set up.

Interesting story Subaruguru, I bet you had your fingers crossed hooking up for the first time.

Sean, thanks for the searches. I read enough of the JR stuff to give me a head ache. he he

Enjoy the music,
MaxCast
Oy. I hung my monos (no jokes, guys) under my floor's joists directly under my speakers, mostly to escape heat and WAF objections, thus necessitating 3m and 6m ICs, and allowing 8ft speaker cables. The ICs are UNshielded Red Dawn, but are indeed balanced. There is abosolutely NO noise audible within 6" of the tweeters, even at full gain.
These ICs run only a few feet away from the house panel 300 amp, and its webs of romex. Am I lucky, or is it the combo of a quiet Aleph P pre and the balanced ICs?
No, I didn't (and won't--threading through the baseboards was a pain!) try RCAs....
I don't know if Sean and/or John are right, but I thought
the issues revolve around current levels, and that's why speaker wires should be short...especially with low impedence or highly reactive loads.
Oh well.
I think the short speaker cables trick used to be a popular way of helping average output monos drive insensitive (79-ish db) speakers? Shorter wire, juice flows faster & more easily into the crossover... with indifferent results, in my case (I used ear519s back then, to drive Apogee Stages)

I have had better results the other way round, as per Sean & Fineberg. I play single-ended.
I personally think that Mgottlieb's message points to the source of the confusion. Balanced ICs were designed to cover long distances. In recording situations there are sometimes 50 feet or more between even a mic and its pre-amp (this is a signal much weaker than line level). Single ended (RCA) interconnects were designed for under 2 meter runs. Speaker cable was also designed for longer but still relatively short runs, but because of its higher level it can usually take runs up to about 25 feet without much audible effect (providing the cable has good electrical properties at that distance). So if you have balanced (XLR) interconnects, longer ICs (within reason) and shorter speaker cable may work well (though you will need very good ICs). But if you are using single ended ICs they should be as short as possible and speaker cable can be lengthened a bit. All that said if your setup has everything under 2 meters or so there won't be much of an issue either way. Personally I use very short single ended connections to the pre-amp (.5 meter) and a .5 meter digital cable. I use a slightly longer cable to the power amp (1.5 meters) and am forced to use quite long 25' speaker cables. I have very good results with this configuration.
Well, the science of it all is a bit confusing, but halving speaker cable length (e.g., 8 feet to 4 feet) will frequently make an audible difference, while doubling IC length (e.g., 1 meter to 2) will rarely have much effect, at least on balanced cables. I've tried it with various speaker-amp-cable combos, and it always seems to work out that way.
Try checking out this link over at the Cable Asylum archive. I found it after doing several searches:

http://www.audioasylum.com/audio/cables/messages/29011.html

To sum it up briefly:

Jon states: keep interconnects as short as possible unless you know for certain that the component can drive a long length of cable. Most preamps should be kept below 2 meter runs for best sonics and 3 meters at max. That pretty much confirms my previous statements of 6' to 8' at max for single ended ( RCA ) cables.

Thorsten Loesch states: Keep interconnects as short as possible. Long interconnects introduce noise into the system, act as antennas and are responsible for ground loops.

Audio Engineer states: Keep interconnects as short as possible due to problems with dielectric absorption, phase shifts, high frequency roll-off, etc...

Even though i received a few private emails contradicting the statements in my post, i think i'm in pretty good company with my statements and findings. There are always "exceptions" to every "rule" or "generalization" though..... Sean
>
Max, i have ALWAYS promoted the longer speaker cable theory as can be evidenced by doing a search on the Cable Asylum. Jon Risch USED to promote the longer IC / shorter speaker cable ideology. I have seen him do a 180* turn on that subject though and as far as i know, he is now supporting the same ideas that i do. He may have changed his mind again though, so who knows.

The bottom line is not to take my word or anyone else's as "Gospel". Find out for yourself. Keep in mind that it is the same signal between the preamp and amp and the amp and the speakers. As such, keeping the budget equally divided amongst the two will give best results. Many of the benefits of going to longer IC's stem from the fact that the user typically steps up to a higher grade of IC than what they previously had. If you do not get an improvement from upgrading ANY component or cable, then something is wrong. Besides that, the use of shorter speaker cables can increase damping factor IF the speaker cables were inadequate to begin with.

Like i pointed out above, what works best in terms of science and logic only apply if EVERYTHING is up to snuff to begin with. If something is out of whack or less than optimum, "band aids" of various natures can be applied and show benefits. Sean
>
I have always thought the rule was longer speaker cables. Then I read some things by Jon Risch. Don't know what to believe.

The reason for the inquiry was because I want to move my mono blocks to some custom made speaker stands which will hold the amps under them. when I read it is better to have longer I/C's I thought this was right up my ally.

Thank you for you help,
MaxCast
Let's look at this ONE more time using both science and logic. You pick the most correct answers and make your own decisions.

1) Which signal contains the least voltage ?

Speaker Level or Line Level

2) Which signal contains the least current ?

Speaker Level or Line Level

3) Which signal is typically transmitted via a smaller gauge / higher resistance connection ?

Speaker Level or Line Level

4) Which signal is most susceptible to line loss ?

Speaker Level or Line Level

5) Which signal is most likely to be interferred with by Radio Frequency Interference ( RFI ) ?

Speaker Level or Line Level

6) Which signal is most likely to be interferred with by Electro-Magnetic Interference ( EMI ) ?

Speaker Level or Line Level

7) Which signal can be distorted or lost and then further amplified, further colouring the sound ?

Speaker Level or Line Level

8) Which signal do you want to keep as short as possible in order to minimize the potential problems that were previously listed ?

Speaker Level or Line Level

I will only add that if you MUST use long interconnects, i would HIGHLY recommend the use of balanced cables. The use of single ended cables ( RCA's ) that are of an RFI / EMI resistant design ( braided, twisted, spiral wound, star quad, active shielding, etc ) should be fine for lengths up to appr 6 to 8 foot at most. Not only do they become more susceptible to signal degradation at points longer than that, some preamps simply won't drive / deal with cables that long in a positive manner. Due to differences in circuit design and cable reactance, you may experience frequency response aberrations, phase shifts, etc... in extreme situations. If your system or component layout absolutely requires interconnect runs notably longer than those mentioned, you should consider the use of balanced cables if at all possible.

If you obtain noticeably better results by using longer IC's and shorter speaker cables, chances are that something in the system is not properly matched and / or working outside of what would be considered its' optimum "window of operation". This may be due to many variables such as impedance mismatches, lack of conductivity or surface area, using a cable as a "buffer" to correct / absorb reactance generated within the system, etc...

Hope this helps and gives others a better idea as to why many folks have these beliefs. Sean
>
6 to 8 are not too bad, but yes it is better to send a longer IC rather than drive current over long lengths.