XLR or RCA?

Longer runs of IC's should use XLR. The influences of shielding, insulation, etc. is completely dependent on the entirety of the cable design. In the end, if your IC length is 2m or less, try them both and see what you like. Also, if you are looking to really hear what advantage balance cabling might offer, your system needs to have balanced circuitry stem to stern. Otherwise, as you understand and accept, there is no substitute for listening and finding what you like, so have at it! And don't forget to enjoy the tunes along the way.

My suggestion to OP is not very technical although I'm sure much of what has been said above is correct.

Borrow two sets of the same cable from The Cable Company, one XLR and one RCA. Listen to both and decide which sounds best to you.

Yes, XLR is more resistant to noise. Shielding plays bigger role in RCA since XLR has inherent noise rejection by usage of twisted pair. Materials still plays role (capacitance, dielectric absorption, metal purity etc.)

On the other hand XLR often requires additional stage to "make" inverted signal. In electrically quiet place RCA can sound better IMHO (but I still like locking connectors). It is also application depended, as you stated. XLR has shield grounded on both ends creating possibility of ground loops in some systems. Also XLR is most likely more expensive - why to spend extra money if you cannot hear the difference?

OK, in theory from this discussion should I take it that, all materials being equal, XLR cables are more resistant to noise than RCA cables? That noise can be both a function of cable length and EMF environment? And furthermore should I assume that more complex construction, at least in terms of shielding, should reap bigger benefits for RCA cables versus XLR cables? But what about wire configuration (I.E. metallurgy, thickness, shape) and dielectric composition? If you believe these are relevant at all, wouldn't they be relevant and affect the "sound" for both types of interconnects?

I understand and accept that many of these variables are highly application dependent (gear, physical location relative to EMF, ears and beliefs of the listener), and there is no substitute for "try before you buy", but I am just looking to see if I have grasped the relevant information in the discussion above...

kn

I used to own the McIntosh MC275 amp. Even though it is not fully balanced, I tried both RCA and XLR cables (Crimson Music Links) between it and my preamp at the time, an Aesthetix Calypso. I preferred the sound with the XLR cables -- I thought my system sounded fuller, with better bass than with the RCA cables. This probably had more to do with the Calypso, which is fully balanced, than the MC275, which is not.

In any case, why not just borrow XLR cables from the Cable Company or buy them from Music Direct, which offers a 30-day money-back guarantee? If you hear a difference and like it, then buy the XLR cables. If not, return them.

Fully balanced signal is what makes live sound reinforcement buzzless and clean (except for electric guitar players with noisy single coil pickups)...with 100 foot (or more) snakes. You can take an unbalanced signal cable from a live musician's instrument pickup, stick that cable into an active "direct box" that changes it into the balanced signal the mix needs, and the result is clear as a bell. Also, the "future" is wireless live mixing (use an iPad).
When I play electric guitars I end up using long runs of unbalanced guitar cables (all the stomp boxes with an input cable and out to the amp) and have to apply small "micro amp" boost in there to unload the pickups and retain a modicum of fidelity. Nothing quite like a modicum of fidelity.

Noise gets in by direct capacitive coupling or electromagnetic pickup. In case of capacitive coupling shield can be very effective. Electromagnetic pickup is a little more complex. Shield is made of aluminum or copper - a non magnetic material and cannot protect against electromagnetic waves, but because of skin effect induced noise currents flow on the outside of the cable - shield. Skin effect works at higher frequencies but at low frequencies cable becomes effective antenna when length of the cable approaches 1/10 of offending signal wavelength. It means that if your cable is very long it will pick-up all sorts of electromagnetic interference in spite of shield. Balanced cable not only allows to reject common mode noise by being balanced but also by twisted wires inside - very effective against noise pickup (capacitive or electromagnetic) by exposing both wires to fields evenly causing same noise currents in both conductors producing cancellation. Keeping cables short is extremely important. My IC is 0.5m XLR.

@ Dlcockrum, Hi, The last part of your post that your white papers said, Again keeping cable lengths short will reduce noise problems, I have been saying this for the last 25 years!, the fact of the matter, I have said this recently to a few members, It does not matter what I tell them, They believe some manufacturer that says other wise, or believe someone they know, It's to the point of, what do I know, right?, LOL!, I use 6ft speaker cables, all XLR interconnects are 1-meter length, no rca cables at all, all componets are true balanced from in-put to out-put that I use, been doing audio for 35 years, this a very long time, again, what do I know!

****************02-23-14: Ferrellms
For long runs, rule of thumb being 50 feet, XLR will pick up less interference (less noisy), but in general there will be no real (measurable), as opposed to perceived, (subjective) difference. *****************

This is nonsense. It is an intentional distortion of the following statements from a whitepaper published by Rane, Inc. (in coordination with the Bill Whitlock of Jensen Transformers, Inc.) used by equipment and cable manufacturers who do not offer differentially-balanced products:

"Since the input stage is not balanced, induced noise on the signal conductors is not rejected. If you must use an unbalanced input, use as short an input cable as possible. This reduces the induced noise. There's a reason it's hard to find and buy unbalanced RCA cables longer than 12 feet. Figure 5i shows both ends of the cable shield connected to units with chassis-grounded shields. If the units are far apart, the chance of the shield currents inducing noise on the signal conductors is greater. Keeping this cable very short reduces the shield current and therefore reduces the noise that is not rejected by the unbalanced input stage. Most systems may require disconnecting one end of the shield for the Figure 5i case. Even a small current in the shield may prove too much for an unbalanced input stage."

and this:

"Fully unbalanced systems do not provide a 3-conductor connector to enable proper use of a shield. In the unlikely event you run across one, use the wiring in the fourth column (Figure 5m-p). Again keeping cable lengths short will reduce noise problems, with or without a shield."

Read the entire whitepaper for yourself:

http://www.rane.com/note151.html

Further accurate information on the benefits of the correct implementation of differentially-balanced circuits can be found on the pdf link at the bottom of the page of this link.

Dave

I used a new pair of balanced cables into a decidedly non balanced but XLR friendly power amp, and they sounded like crap. Seriously...replaced them with much better XLR cables and all is well (although in this case I could be using single ended cables I do prefer the XLRs anyway). I noticed some substantial differences between brands of ACTUALLY balanced cables when mixing sound with headphones from a live feed (TV show production). Even with truly balanced signals I can say that better cables can sound much better. Who knew?

XLR any day for me. In my system, when used between my preamp and mono amps, it eliminates hum which is ever present when i use RCA cables.

Yes, you are right, full benefit is obtained if the system entirely is wired using XLR

My experience has been that if the equipment is fully differential, properly implemented, an XLR cable is a real bonus as the cable is sonically transparent. I've run some very good RCA cables, and in one application where the output of the source and input were fully differential, I could find no meaningful difference between a set of Mogami cables and anything of any price upwards of $1K. And I set that limit of $1K not because I think cables above that can beat the Mogami, it's that I just have no interest in spending more on cables.

And as noted, just because there's an XLR output doesn't mean it's better. I have a Pass Aleph Ono phono stage and I asked the kind folks at Pass which I should use into my Pass INT-150. The answer was- use the RCA

Jmcgrogan2: please direct message me your address, and I will send you a sweet record.

XLR balanced low impedence cable signals can run 100s of feet with ZERO noise.

Other than locking, XLR does not have exposed pins (input side is always female). Touching end of RCA cable going to power amp can damage speakers. Power should be turned off, but mistakes happen, especially when people (audiophiles) keep plugging and unplugging things frequently.

02-23-14: Wolf_garcia writes:

Besides, Neutrik Speakons are made in Liechtenstein, which should be all you need to know.

I've done a lot of IT consulting in Liechtenstein and you are right. Workers there inherit holidays for both Switzerland and Austria and average 14 public holidays a year, and in addition, typically with the whole of August off.

A good life - and it's a very pretty place; but you really want to stay in Feldkirch, just across the border in Austria.

Regards,

For long runs, rule of thumb being 50 feet, XLR will pick up less interference (less noisy), but in general there will be no real (measurable), as opposed to perceived, (subjective) difference. XLR is used by pros as it is can be used in long runs and is more reliable physically.

I'm mystified that RCA and phone plugs are still around, but I have to accept it and use them daily...there may be an argument for RCAs taking up less space, but after many years of using XLR mic and balanced ins and outs of live sound mixers and recording stuff I'm an athletic supporter of XLRs, even when not part of a balanced system. When I'm declared Ruler of the Universe I'll demand Neutrik Speakon connectors and XLR plugs be universally used all the time. Besides, Neutrik Speakons are made in Liechtenstein, which should be all you need to know.

I think this basically sums the difference between rca and xlr and why xlr was developed.

Unbalanced Cables

Standard Jack Unbalanced Cable.

Unbalanced audio cables are cables that have 2 wires that serve as conductors; 1 carries both the signal and the ground. They are less expensive, but they tend to allow the introduction of noise into the audio signal. This noise is caused by interference from the electrical fields of nearby electronic devices. For this reason, unbalanced cables should not be used for audio applications that require them to be long. The maximum suggested length for unbalanced cables is 25 to 30 feet.

Balanced Cables

Balanced audio cables have 3 wires that serve as conductors--negative, positive and ground. The positive and negative wires both carry the audio signal but with opposite polarities. This means that any noise from outside interferences (such as from other electrical devices) will be cancelled out. Balanced cables are suitable for audio applications that require long lengths (more than 25 to 30 feet).

XLR Cables

XLR Female Connectors

XLR cables are balanced cables. They have circular connectors with 3 pins--positive, negative and ground. Essentially, they are the same as a TRS cable (which stands for Tip Ring Sleeve). The only difference is the type of connector (TRS cables look like standard-jack cables). "Pin 1" (the pin on the left side when facing the male-connector, or the hole on the right side when facing the female connector) is always the ground signal.

How XLR Cables Work

When two balanced devices are connected, the XLR cable delivers a positive audio signal, a negative audio signal and a ground signal. The positive audio signal and the negative audio signal are identical, except that they are inverted. If electrical interference is encountered (such as from nearby electronic devices), that noise enters both the positive audio signal and the negative audio signal. When the positive and negative signals reach the balanced device at the end of the chain, one of the two audio signals is inverted. Now the two audio signals are identical in every way. However, the noise that was on the audio signal has now been inverted. When the inverted noise is combined with the original noise, they cancel each other out. All that is left is the original audio signal.

I compared the 2 side by side when I had a Meridian G08 to an Octave V70se. What I found was the xlr was louder as it should be since the average voltage output is typically 4V where the rca is typically 2V. For that reason I decided on the rca because I didn't care for the higher gain. Others prefer the higher gain. I'm glad I decided on the rca because when I replaced the Meridian with a Naim music server it only has rca connectors. On the other hand if I needed cables that are 20ft or longer I would highly reconsider the xlr and equipment that has a truly balanced designed.

Just my opinion

Elegal, I found more of a difference using an upgraded cable than between using XLR vs RCA connections. Using the same upgraded cable into my C2300, I preferred the XLR. As you know if you are using the phono stage on the C2300, you need to use RCA

Bob, differential amplifier can suppress even order harmonics (associated with warm sound). Voltage of each output can be seen as:

Vp=a1*Vi+a2*(Vi^2)+a3*(Vi^3)+a4*(Vi^4)+a5*(Vi^5)+ .....

Vn=a1*(-Vi)+a2*(-Vi^2)+a3*(-Vi^3)+a4*(-Vi^4)+a5*(-Vi^5)+ ....

where "a2, a4 etc" are even order coefficients. Since polarity of the input signal will be always positive (always raised to even power) for even order harmonics they appear same on each output and will be removed (assuming identical gains).

Vp-Vn=2a1*Vi+2a3*Vi^3+2a5*Vi^5....

The other problem might appear when gains are not exactly even because it can convert common mode noise/interference into normal mode signal.

I just changed out some antique 25 foot RCAs running from my C50 to my MC302 with 5 meter XLRs and dropped the noise floor substantially. Immediately noticeable. Can't give you a critique of the sound quality yet as I haven't been able to do any serious listening, but so far I'd say I made the right move. Even though you are all tube, I strongly suspect you will find the same thing. Dropping the noise floor always allows more fully resolved fine detail.

I have a Calypso tube preamp. It has fully balanced differential inputs and outputs. When I use the differential inputs and outputs I loose much of the tube character of my preamp. I have a tube preamp to get the euphonic tube sound to music. Tubes improve the emotional impact and add a wonderful organic character to the music. The differential inputs and outputs seem to be clearer and possibly more technically accurate versus the single ended ones. I guess that is due to the noise and distortion canceling effect of true differential circuitry. I use the single ended inputs for music and use the differential inputs for HT. I gather this only applies with true differential circuits and probably won’t apply to Macintosh. If so, then I would think the XLR connections would be best whenever they are available. Of course, the only true answer is your own ears. This is just a guide to know what kind of things to listen for.

Bob

My Cary SLP 05 sounds better with RCA, so that's what I use.

if preamp and amp are balanced you'll get most of most benefit and cheap xlr wire will sound much better tnan expencive rca

No worries Al, I am sure that you are right. I would not disagree with you. I am simply repeating what I have been told, it could very well be misinformation, as this hobby is filled with it.

To be honest, I have gone both routes, running a fully differential system with all XLR cables, and running all single ended with RCA's. I cannot say that I have had a noise issue with either. I think I may have convinced myself the balanced was quieter, when I had it, but now that I'm back running SE again, and have no noise issues, I'm convinced I was just buying the hype. ;)
There may have been more measured noise, if I had the proper instruments to measure. If there was any difference in noise level, it was not audible, to me anyway.

I do agree with Mofi that the XLR connectors make for a superior connection though.

Elegal...no need to send LP's, that's why we're here....unless you have extras laying around that you are just looking to give away. ;^)

There is a minor benefit in using XLR even though the component is not internally balanced....the big benefit is when one does it right.

Thank you MOFI; AL; and JMC. I feel like I should send you guys some LP's for your help

I would like to add to what Al says above in that the XLR connector itself is much better than most of the RCA connectors. IMHO, better than even the locking RCAs. Another potential added benefit to using balanced cables.

This is one of the rare occasions when I must very respectfully disagree with my learned A'gon colleague and friend Jmcgrogan2, although the opinion he expresses is shared by many audiophiles.

Elegal 02-17-14
I then read that xlr cables only provide a real benefit if you are using balanced cables throughout your entire system, which I presume means from my turntable into my preamp. Is this true?

My answer is that it is not necessarily true. Regardless of whether the components in the system are internally balanced or not, and regardless of whether all of the interconnections in the system are balanced or not, a balanced interface between any given pair of components is inherently less susceptible to noise that may be introduced as a result of ground loop or other effects. And in referring to "noise" I am referring not just to noise that may be audible as such, but also to the possibility that noise that may be introduced at frequencies which are too high to be audible may have audible consequences, by "intermodulating" with signal within the components.

On the other hand, it is not uncommon for the XLR interface circuits of components that are not internally balanced to be sonically inferior to the RCA interface circuits of those components.

The bottom line, IMO: Which interface would be best to use in your particular case cannot be predicted with any certainty. Hopefully someone will chime in who has experience making that comparison with the same or similar McIntosh components, as other opinions stand a very good chance of not being applicable. IMO, of course :-)

Regards,
-- Al

The real benefit of using XLR cables is in the fact that it allows you to utilize fully differential equipment designs, which will lower the noise floor. That being said, just because you are using XLR cables does not mean that your equipment is a fully differential design, it could be a single ended design that simply has XLR inputs and/or outputs on the unit for flexibility. If the unit is not a fully differential design, you will not receive the benefits of running XLR cables.

Also, yes, to receive the benefits of running fully differential, you should run all the way from source through amplifier to maintain this 'advantage'. Any break in the chain, anywhere in the chain, and using XLR interconnects becomes rather pointless.

According to the McIntosh website, neither the C2300 nor MC275 are fully differential designs. The XLR inputs and outputs are simply cosmetic, therefore should hold no advantage.