One of audio's age old questions is short loudspeaker cables and long interconnects or the opposite. Paul walks us through the possibilities for best sound.
Thanks for explaining this factiod in an easy-to-understand manner Paul. From a pro-sound perspective, the new generation of active sound reinforcement loudspeakers are doing exactly that. And yes, it has to be over balanced interconnects as well. Ideally in you would want a monobloc amplifier or more if you you're bi or tri-amping the speaker to be next to the loudspeaker. Also the higher impedance nature of the signal being driven across balanced interconnects from a preamplifier usually equates to lower loss in detail across the entire frequency spectrum. Big thanks and keep your posts coming. Much appreciated.
I did as Paul suggested. Absolutely true! You must also consider that doing this you move the rack to a side of the listening room and the soundstage gets coniderably larger
Paul, you recently reminded me of your Copper magazine... damn, what a goldmine. Franken-Doris is one helluva resource... eminently qualified. Thank you for that.
In recording sessions in say large stadiums or cathedrals, it’s not uncommon to have microphone cable that runs over 100 yards. But that’s balanced of course.
Paul I'm shocked! Even back in the ancient 20th century when I was running your Linear Control Center & Phono III preamp pairing we knew long ICs & short speaker cables were the way to go if you had a large span from your preamp to your speakers. Welcome to the future...😊
I suppose I am following your advice . My XLR interconnects are one Meter & speaker cables are 3 feet, so 3.4" longer . Analysis Plus Solo Crystal Oval XLR & Silver Oval 2 speaker . Great !
You can certainly use long speaker cables, if you keep resistance below a certain level, and if you use good quality speaker cable (yes, it makes a difference!).
There's a reason why higher-amperage loads need to be paid attention to over long runs wrt wire gauge, while smaller amperage loads are far more agreeable over those distances. That reason amounts to simple physics, wire surface area to transmit electrons in conjunction with impedance per unit traveled. Paul's conclusion (sorry you had to learn it the same way I did :P ) is electrically correct. Of course, you could use - not that I've experimented - #2 copper for speaker wire, which probably wouldn't care about the amp you were using or the length of run, but it'd be difficult to fasten to a speaker terminal.
You can also double the wire, in a “shotgun” configuration. For example, double the gauge 12 wire is about the same as one gauge 9 wire. Benefits: easier to wire and cheaper.
Sure but short speaker cables only works with monoblocks, and the disadvantage of balanced is it adds to the complexity of the signal path. There’s a school of thought that says keep the signal path as short and simple as possible.
I agree. Most of us need to keep our equipment off the floor, in a rack. In this case, short interconnects and longer, proper gauge speaker wire is the answer. What if we like tubes? How many tube amps and preamps have balanced connections?
@@markceci9896 More and more of them, McIntosh, Cary Audio, ELAC, Rogue, to name few. Balanced connections are a prerequisite for me, the Common Mode Rejection just makes too much sense to me; although many more professional AV experts disagree with the premise, they maintain the distance needs to be 25 ft or more to justify a balanced connection; on the other hand, my old boss at Radio Shack (25 years experience) maintained that the effective limit of an RCA interconnect is about 12 feet, coincidentally the longest TCA cable they sold. Offered for your consideration.
@@jeremiahchamberlin4499 I appreciate the info, sir. I still dont really have a dedicated sound room. I have to keep some semblance of decor in my family room...lol. Thankfully, I do have some room treatments and natural absorption diffusion features. Amps on the floor is a no-go with my wife.
In the listening test you cite were the cable types (make, model and connecter types) kept constant and only the lengths varied? If not, other changes could have confounded the results.
If given the choice between long speaker hoses, or long interconnects, I would chose long interconnects every time. The voltage drop with speaker hoses is high compared to the loses with long line level cables.
Interesting, I too have also assumed that short interconnects and long speaker cables is the way to go. I still do it. I have never even thought about the other way around. Well, maybe I'll try it one day when I have mono blocks.
Paul, stop trying to sell us audioquest speaker wire . The only thing that matters is less resistance to the speaker so having thicker speaker cable will make a difference but that's being measured with gear not by ears. If it interconnect is 6 and another one is three feet you aren't going to hear a difference in the sound. That's baloney. So rule-of-thumb use thicker speaker cable because there will be less resistance. Maybe if you used 200 feet of speaker wire versus 5 feet of speaker wire you may hear a difference.
That advice would work only if speakers are connected to some monoblocks. In case of a stereo amp speaker cables will always be longer than interconnects.
My system is running with Audio quest cables from short balance connectors and 8 ft away from each 14 gauge speaker cable no banana plugs and Love the sound
Always remember to use the appropriate Wire Gauge for your speakers. It makes a huge difference. Unbalanced signal cables should be short in order to avoid picking up radio interference, etc. It's not really a matter of signal loss. If you have balanced inputs and outputs on your equipment, great. But if you need to travel great distances, you need a transformer balanced connection.
Hi Paul,awesome as always! So for all us common folk that only have RCA options, stick with short interconnects (6ft or less) and long heavy gauge wire? I also wondered since a speaker wire is basically an antenna
Speaker cables run much higher currents than interconnects ever will, whatever stray EMF the speaker cables pick up are extremely unlikely to strong enough to generate audible sound. If you don't have balanced interconnects, short interconnects and long speaker wires make much more sense.
@@4G12 and even if: it would need demodulation because all that antenna staff is in frequencies far above the audio range while on interconnects and it is a real problem
@@Harald_Reindl It's not about RF transmission, it's about RF reception. If you have say, 2x3m of speaker cable it is the wavelength for 100 MHz in the VHF FM band _and multiples_ of that frequency. Some amplifiers are sensitive to RF interference (entering the chassis via their ouputs!) even when totally unrelated to audio frequencies. That's my understanding of the issue here. I do know that when I RF-proofed my system in the same room as my WLAN router, the sound improved considerably. This is a topic in itself.
All the engineers I have ever known, including my son, love the country of Denmark. It's because the Danish are mean just like engineers - with the exception of my son!. As is often the case, Paul is correct about short speaker cables and long interconnects.
I use super loooooooooooooong interconnects. Input resistance for where those interconnects will go is very high compared to the resistance of the cable. So no worries there. Balanced helps cancel any noise a longer, low signal level cable may pick up. Speaker cable resistance is probably a higher percentage of the resistance of the speaker. Usually still negligible at regular distances. Your inductors will have more DC resistance.
It really only matters if the speaker cable is so long that it becomes lossy and has enough capacitive and inductive reactance to cause the amplifier to want to oscillate. Then using balanced XLRs instead of the long runs of speaker cable would be the way to go.
Those speaker cables are only short when you don't count the cable in the speakers:p. I used to have short cables but when I wanted to place my speakers further apart they were too short, so now I have ample length.
I simply use real OFC speaker cable, not the aluminium clad rubbish that goes bad. I use the length needed to reach my speakers. I use £10 interconnects. The cable i use is 2x79 strand. It cost me £2 a meter. I run a NAD C352 amp with mission 734i speakers in a large room with a high ceiling. It sounds awesome. Clean, dynamic and exciting. I've had several positive comments from others. I don't think I could really do much to 'improve' the sound. It's already great. I always thought that speaker cable lengths was all about the resistance? So longer runs slightly thicker cable? I think I would crack up if I trimmed a couple inches of speaker cable off until I found 'better sound' 😆
Sorry for such a noob question, but what is that black box where the speaker cables are coming out of? I'm having a hard time visualizing the system or why its configured this way.
It depends entirely on the results you get in your particular circumstance. If an unbalanced connection doesn't suffer any aberrations like hum, or intermittent buzzes or clicks, then no problem. A constant low frequency (mains frequency) hum is usually caused by a common earth path, commonly called a ground loop, or hum loop. That typically happens if both power and pre amps are 3-prong mains earthed devices. Some preamps have a "ground lift" switch to counter this, or a permanent ground lift may be able to be installed as a modification by a qualified audio technician. A ground loop is one of the main problems avoided by using a balanced interconnect. Other hums, buzzes etc can result from the limitations of shielding achievable with unbalanced cables, because the cable shield is also the signal return path and if it gets electrical noise impressed upon it, that becomes part of the signal. Using better quality cables with heavier shielding can improve matters to some degree. Another approach to that problem is to use cables wired as "pseudo-balanced". This uses balanced cable, with the hot wire carrying the unbalanced signal output as it normally would when balanced, but the cold wire being used as the earth connection, and the shield being isolated at one end (i.e the shield only earthed at one end of the cable, this end should go to the earthed unit if one is unearthed) so as not to be part of the signal return path at all. That way the unbalanced earth path is also shielded from electrical fields, and this can considerably reduce induced pickup from long unbalanced cables passing mains wiring, transformers etc. In the end, if unbalanced works without problems, fine, otherwise if possible go for balanced.
I want to compliment Paul on his likeable demeanor and pleasant personality. This is a person who has been in the field for decades yet lays his opinion as such rather than absolute facts of the Gods. Informative videos recorded and presented nicely. Keep it up ☺️
Here's an interesting little snippet from Hi-Fi News back in the 90s. This is Ben Duncan: "... with a single stereo power amplifier, highly local placement of the amplifier channel is impossible, either physically, or to live with. Moreover I maintain that the effect of cable length is likely asymptotic - the counter-intuitive behaviour that Martin Colloms hinted at in his tests of a decade ago. So shortening a cable will emphasise the discontinuity of the final few inches inside the cabinet - including the drive units' tinsel wires." You really need to read this a few times before it makes sense (well, I had to!). I guess that if one has serious high-end speakers and their internal wiring is top notch it might be different story?
the length of the speaker cable don't matter unless you talk about hundreds of meters and everybody pretending the opposite is a fool which would buy audioquest
Naim recommends a certain minimum length of cable. And often it is recommended to use power cables of a certain length. Both for filtering. If that makes any difference, I don't know.
This is because Naim amps used to have a different output stage requiring a minimum so the cable can act as a capacitor (in effect) as they preferred a reduced number of components of a higher quality. I think Naim may have changed the design of their power amps over the years so this requirement may no longer be recommended.
I come from Denmark, and totally agree, Dali, Jamo and Dynaudio are some of the best brands out there. I actually live about 15 away from the Dynaudio factory in Skanderborg, Jutland
Line level cables are driving HIGH IMPEDANCE, while speaker cables are driving LOW IMPEDANCE. Or more importantly, those line level cables are low voltage, low current, while the speaker cables are low to medium voltage and HIGH CURRENT. So the line level cables may be thin gauge wire with low loss, but the speaker cables must be thick gauge wire, or you will get HIGH LOSS. So it should be obvious, put the amps as close to the speakers as possible and run long line level cables to the preamp for best results. BUT those line level cables must be very well shielded or you will pick up noise! Also watch out for ground loops! Balanced lines can help with that ground loop issue, but watch out for the correct phasing? (There ain't no such thing as a free lunch!).
Assuming your question is serious, the suggestion is not that long cable interconnects are in any way desirable. It's if one needs to be long, which is preferable.
@@marcusaurelius2988 Yes, except for some weird outlier situations which most people will not encounter, being just comfortably long enough to reach is ideal.
Its so funny when he says it sounds "SO MUCH BETTER" I bet there is no difference at all. And there is nothing at all you can do to make it sound so much better! There is only slightly better, or for the most part, no difference at all! I bet I could change those cables to 20 feet cables for $5, and you wouldn’t even notice!
Long runs were 100% a base goal when XLR was being designed. What about ole RCA's though? If it aint on a subwoofer, should RCA's go any longer than 5-6 foot?
Ok, that's fine with separate components but what if we are talking about a stereo receiver. I'm helping a friend with his setup and he's not an audiophile at all and in his room the ideal location for the equipment rack is not near his speakers and somewhere on the order of twenty to twenty-three feet will be required. I have already purchased enough nice 12 gauge wire for the job. What do you think?
Tell him he needs to upgrade to either (1) a receiver with XLR out for the speakers, or (2) mono-blocks placed close to the speakers. If the receiver has pre-amp outs, he may get away with active speakers. Provided the budget is unlimited and the wife approves of all changes to the interior of her house. 🙂 But seriously, use what he has and try to ameliorate whatever faults you or he might perceive. It is probably a good practice to use cables of equal length.
Ok. Now, what would be the longest I can go on XLR Interconnects? Speakers in one room, equipment in another (retired to a condo) can I go 40 ft. on XLR’s? I shake because it’s going to be expensive for custom made.
@@laurencel.2493 surely because I own a working brain, common sense and technical understanding while most audiophiles are even two meters around of their head dumb
Long interconnects are more prone to interference simply because they are operating in a much higher impedance circuit compared to the speaker cable circuit. Good quality screened RCA interconnects are perfectly adequate for local connections.
very interesting thanks. I recently moved and the new set up (to avoid long speaker cables) speaker cables about 4 ft from 2 quad 405 bi-ampd into B&W Matrix 3e loudspeakers with a long (about 15 ft) Y interconnect to quad 44. The interconnect uses quad 4 pin DIN plugs. There is a fair bit of noise which i'd love to eliminate if possible. I'm thinking of a balanced Y cable but the quad 44/405's are unbalanced by default. any ideas pls thanks
Paul, you mean real "Balanced" correct. Don't some manufactures put a balanced "connector" on the unit, while not being wired correctly as a real balanced circuit ??
There can be what is called an "impedance balanced" output - one of my USB audio interfaces has impedance balanced output. This supplies a single hot signal, and the cold connection is simply returned to signal ground through a resistor of identical value to the series resistor supplying the hot signal. This works for noise rejection, as common mode noise cancellation depends on both hot and cold circuits having identical impedance, so any noise voltage induced is identical in both. The out-of-phase arm of the balanced input then still cancels this to the degree of its CMMR (common mode rejection ratio) as normal. So you still get common mode noise reduction, but you do not get the doubling of signal level you would get if the same output was implemented as a full balanced output, so for the same output voltage drive, the output stage has to deliver twice the voltage swing (potentially resulting in higher distortion).
@@PlatypusPerspective How do you figure you lose signal with a balanced connection? I feel you are sowing FUD (Fear, Uncertainty, and Doubt). A balanced connection purely makes sense, it could as easily be argued that an unbalanced connection delivers half the power available through a balanced connection. I’m sorry, but I need more than your word for it that everyone else I’ve heard on this topic is wrong and you are right.
@@jeremiahchamberlin4499 Hi Jeremiah, you seem to have misread or misunderstood the information I've given, since I don't figure you lose signal with a balanced connection. Nothing in what I wrote says that. In fact what I mention requires that a true balanced output inherently delivers twice the voltage (and hence four times the power) of an "impedance balanced" output. It's vital to recognize that my comments are specifically answering Henry's question, with an explanation of the different behavior of an "impedance balanced" output configuration. I will edit my final sentence so this is more clearly evident - note that in that sentence, referring to an impedance balanced output, I am saying that with that type of output, you do NOT get the double signal level that you WOULD get with a true balanced output. Interestingly, with either balanced or unbalanced connections, the FUD figures remain at zero!
@@PlatypusPerspective OK. I misunderstood, and actually, I’m not exactly sure what “impedance balanced” means in this context. I’ve heard from others recently that one wants a high impedance source feeding an amplifier, but that at the output, one wants a low impedance from the amp feeding the speakers, hence the use of output transformers in tube amps taking the relatively high impedance from the tubes down to the relatively low impedance of the speakers. The transformer not only helps drive the speaker, it also helps protect the amp. If that is what you mean by ‘balanced impedance’ I’m all in with you.
I will be buying new speaker cables within the year and I was looking at the price of the cables that I want to purchase. There was no difference in cost between six, eight and 10 foot lengths. I already knew that short speaker cables are the way to go and I need 6 feet. Then I thought that what happens if I want to separate the speakers even more that I might as well by eight feet but now I’m reversing my position again. Six feet it will be.
Assuming you are doing as Michael in Denmark is doing... meaning, your speaker cables will be used with monoblocks, then: If you choose to separate your speakers even more, then move your monoblocks to accompany the move of your speakers. The length of your speaker cables will remain unchanged.
In my 7.4.4 system I used just over 100 metres of speaker cable, no other choice of shortening them at all, sometimes I think you just have to work with what you have and the space/length that's needed for speaker placement.
Same here. Ran a good 100 meters total of in-wall rated 12 gauge oxygen-free copper for 7.2.4, plus another 2 lines of 50 feet of shielded RCA for each sub. Amp and sources are in the rack in the mechanical room, a 50 foot run from the front speakers. Short speaker cables are only an option for people with media rooms that are never entered by their spouse. :D
@@Harald_Reindl there is no visible positioning of amplifiers or sources that is acceptable to non-audiophiles. The speakers themselves are already "bad enough". I think it's worth noting that even Paul has his home theater rack in a separate room behind the the rear wall of his home's theater room. So he may preach short speaker cable runs for demo rooms that exist for no reason other than 2-channel listening and which can have components and cables in an unsightly mess on the floor behind the speakers, but in his own home he's running short interconnects, long in-wall speaker cables, and a tidy room appearance that can be accepted by the other members of his family.
@@roadfordays well said, if I want to use two channel with my system, then my cables to polk signature series s60 fronts, are just over 2 metres each. In a system that is explicitly stereo, it might be beneficial having shorter leads as Paul has said.
As far as "sounding different", if it really does, then you would want to investigate carefully with an oscilloscope. Otherwise all those fancy amplifiers are going to be letting out that nice, juicy audiophile grade magic smoke.
Going by the same logic, shouldn't we design powered/active speakers all around with XLR inputs (used with a good DAC, a great preamplifier and balanced interconnects - thus doing away completely with speaker cables), instead of using a whole set of monoblocks and two types of interconnects/cables? If not, please elaborate why?
That's what I use, sounds fine! Doesn't suit everybody though, I guess. Makes for a simple setup though - Event powered studio monitors, Cambridge DAC, optical in from the player.
It kind of makes sense, speaker cables impact the damping factor of a system whilst interconnects drive high impedance loads with a low impedance output. You could call that voltage mode.
Yep, difference between current and voltage source. Speakers are a current load and much harder to match and drive than a low current/low voltage (~1V) low level signal in an interconnect. I would further say, that being balanced has nothing to do with the ability to drive the interconnect. It only has better noise rejection. Often, the additional buffers needed to invert one side of a balanced line and back again add additional degradation of the signal.
I can never understand that setup there ... all the real punch comes from the 6 speaker subs with their own class D power amps .. but why all that power to drive those tiny electro magnetic ribbon drivers ? Talk about overkill ! I know I know there's headroom to consider but ...!!
Hi Paul, but there loads of people that think cables are snake oil, I'm batting them away everyday, I think these people are either jealous or have very poor hearing, they prefer to Hook their equipment up with Bell wire because they don't think there's anything better, what do you think.
"there loads of people that think cables are snake oil" I can agree that there are lot's of people that cables are snake oil. My view is that there are very real issues on cables, strange beliefs on them and lots of snake oil marketing related to them. "these people are either jealous or have very poor hearing" Not necessarily. "they prefer to Hook their equipment up with Bell wire because they don't think there's anything better" Why don't music studios where people really know about all things on sound quality and different cables don't use expensive special cables? Maybe because they don't offer them any improvement compared to good quality "normal" cables? Or what they offer is so tiny that it is not worth of the high price boutique high end cable maker ask of their products?
@@epanorama Hello Tomi, the biggest thing with hi-end cables are, they are ridiculously overpriced, the core materials in making these cables are peanuts, especially in high volumes, now that's where the snake oil comes into it. But I'm showing how to make true hi-end cables for penuts. The gains are really good when you're making hi-end cables for peanuts, but when it comes to spending vast amounts of money on cables, I can see why some people would say that the gains are minimal. The problem with the snake oil police, is they don't mind upgrading capacitors and the other parts inside there equipment, but they done understand that, different qualities of wire can improve sound, wire is no different to capacitors, the better the capacitors the better the sound, the better the wires the better the sound, it simple, thanks for commenting & watching
A good quality 50ohms interconnect cable doesn't need to be balanced as long as it's triple shielded low capacitance, I would say use balance if you're going above 25 ft. Speaker cable use what the manufacturer recommends and keep it low capacitance solid core is generally better
Paul, as you are well aware, your initial assumption was correct. Now you have been persuaded by Audio Quest to say that you were in error. Is it a coincidence that PS Audio pushes Audio Quest cables? I think not. It is impossible to detect an audible difference between a short and long cable in a typical lengths used in a home system provided the low level signal cables are as short as possible. Cable from a turntable should be short and well shielded with low capacitance. Unbalanced RCA are perfectly fine for cable lengths in most home systems. Audio quest markets some of the most outrageously expensive cables accompanied by fraudulent technical claimf. Come on Paul, show some integrity.
