[Robonaut]

I'm in the market to create a new 2-channel setup, and, thus far, I've only been looking at gear which has XLR connections. I've been an amateur musician for many years now, and I always noticed a substantial improvement everytime I switched from unbalanced to balanced connections.

However, as I've been doing my research for new gear, I've seen it mentioned numerous times that balanced only makes a difference when running long cables (50 feet, or so).

So, what are your thoughts on this? Should I concentrate on gear with XLRs in order to get the best sound quality or is XLR really no better than RCA for short cable runs?

 

[Mr_Sukebe]

It does seem to depend upon the equipment in question. Personally I've found little difference when comparing the connection types, though I've certainly read of others find big improvements.

As I've suggested before on this forum, base your decisions on what sounds good, not whether something has a great specification.

 

[hugz]

There are 3 obvious advantages of a balanced setup
1) eliminate noise in the line. this should really only apply for long connections
2) 4 times the power (double the voltage and double the current)
3) XLRs are just a better plug than RCA

those are all the reasons off the top of my head that I can think of why balanced is better. However, many many people attribute a greatness to balanced setup outside of those reasons. I'm not sure what the justification is how this, but many people with balanced certainly say that it adds a whole lot to the setup

 

[sc53]

Keep in mind that some sources have XLR outputs as well as RCA's but are not really a balanced circuit. You have to audition the gear to tell if the XLR connection is really the best. My Ayre CX-7e does sound better with the balanced connection, but that's into an amp (AX-7e) with a true balanced circuit as well. My Arcam gear (FMJ CD-33 and DV-27A) only has RCA connections, and it sounds fantastic. In most gear, the XLR out sounds louder than the RCA out, but that is easily fixed with your volume control. I wouldn't rule out gear just because it doesn't have XLR plugs.

 

[Ori]

Balanced is prefered in pro-audio because it is safe. You don't risk down time in a studio and you definitely don't want to disrupt a live show! Different priorities!
Not only the connection but the gear too. Many people believe that what's good for a studio is good for home. Wrong!
Speakers for nearfield monitoring are not exactly musical and rack gear is not the last word in musicality either. The advantage of a home hobbyist is that he can afford down time and/or endless tweaking in order to experiment for the best sound. Take advantage of that luxury!

Now to your specific question. My answer is: it depends on the gear. Here is a real case. My friend had a dCS Verdi DAC and a Spectral amp, both offer XLR and RCA jacks and both can invert phase at will. Why do I mention that? Well, if you invert phase in both, the end result is no inversion, so theoretically it's the same as both units not inverting phase. Not true in real life. Turned out that when both invert phase, the XLR connection sounded better, otherwise the RCA sounded better.
This has nothing to do with the phase inversions, just with the internal circuitry in that specific path. For some reason, it worked its way in that manner. I'm sure there was no intent by the engineers to do so...
The only way to find such an anomaly is by trial and error and that goes back to the fundamental question: do you want a safe receipe for the home environment or would you rather invest more of your energy to extract the best your gear can potentialy give?

To me, the latter is a great advantage and it results in a lot more rewarding sound. Now we are talking subjectively... To me, simplicity is the way to achieve the best (a very subjective word!) sound. I favor purity of tone over abundance of detail and analytic dissection of sonic cues. I favor integration of tones as one performance over the last word in deep bass response or a super-flat (as though it's achievable...) frequency response. I'll take some cabinet resonance as a trade-off for smooth transition between drivers. I also much prefer simple amplifier circuits over very complex ones. Unlike what common thinking seems to be, there is no free lunch!
You add circuitry - you will pay for it one way or another. You might get diminishing levels of distortion but you will lose another aspect of audio purity and it all boils down again to personal preferences. The studio guy doesn't have that much freedom, in most cases!
Everything I listed above is very subjective, reflecting my own preferences. Regardless, even if your set of preferences differs from mine, I still believe that the true way for your audio perfection is not the safe solution. Don't succumb to misconceptions

!!! Experiment with balanced, single-ended and other aspects of your system - and you shall be greatly rewarded!!!

 

[sc53]

Ori: SO eloquently stated! And so right.

 

[Robonaut]

Quote:

Originally Posted by Ori /img/forum/go_quote.gif Don't succumb to misconceptions !!! Experiment with balanced, single-ended and other aspects of your system - and you shall be greatly rewarded!!!

Based on this logic, though, wouldn't you want to always get gear with XLRs since it offers you the option of whether to use them or not?

 

[Solude]

If you are looking at gear in the price range to include XLR... get XLR. The biggest advantage will be being able to run your cans balanced which you simply can't otherwise.

 

[hugz]

Quote:

Originally Posted by Robonaut /img/forum/go_quote.gif Based on this logic, though, wouldn't you want to always get gear with XLRs since it offers you the option of whether to use them or not?

Not if adding the option of XLR makes the gear more expensive, yet RCA is superior in your setup. Why pay for something you dont use? That money could be spent elsewhere in your setup to actually improve the sound

 

[Ori]

Quote:

Originally Posted by Robonaut /img/forum/go_quote.gif Based on this logic, though, wouldn't you want to always get gear with XLRs since it offers you the option of whether to use them or not?

Great logic...

That would certainly be the case for many people, but then we go back to the free lunch concept. If your gear has "extra" circuits and you happened to not use them, is it optimal? Probably not.
For me, there is an over-riding issue, which has to do with the complexity of the balanced circuits vs. the unbalanced. There are more parts in a balanced circuit and to make them work in sync you have to invest more effort (part matching) or sacrifice another aspect (ah, our good friend, Mr. negative feedback...). To me, fine-tuning a simple circuit is the way to achieve minimal electronic processing. I believe that it's the best way to achieve optimal sound quality in the home environment. I'm not completely against balanced or negative feedback, but only in limited levels, where they turn out to be the optimal solution. Unfortunately these techniques are applied as a standard solution in commercial gear, not as a fine-tuning measure to resolve other system defficiencies.
To my ears, the result is not as pure as what I know is achievable by simpler means. To me, most commercial designs miss the concept that "everything must be made as simple as possible" because they dismiss the very simple solutions as "too simple". In my assessment and to my taste - they are not.
I do understand the constraints in mass-producing a design vs. tweaking a specialty product. What I'm saying is that the home hobbyist has that extra level of freedom to tweak his system in a unique way. It then becomes one of a kind, which is likely the more optimal (and more elegant) solution vs. one shoe fits all... Just my opinion...
As a consequence of this philosophical conjecture, I prefer to start from the simplest starting point. You may eventually eliminate this option as "too simple" and move on to a more complicated solution, but the burden is on you to prove that it's "too simple" before resorting to seemingly a less optimal solution.
Man, it's quite complicated debating simple solutions, isn't it?

 

[DrewWinters]

"Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away."
-- Antoine de Saint-Exuper

 

[cotdt]

for some mysterious reason, when i have both my RCA and XLR outputs wired on my Zhaolu 2.5A (note: stock 2.5 doesn't have XLR outputs, i added them myself), the sound is slightly degraded. it is very slight, and yes i wired them correctly. so i had to make a choice and picked the XLR, and removed the RCAs. I could always use the XLR->RCA convertor for my RCA stuff anyway.

 

[audioengr]

This is actually simple:

XLR has advantages of common-mode noise immunity, such as ground-loops that cause hum and RF that can be picked-up by large loops.

RCA single-ended is usually simpler and easier to design, so it can sound better. Fewer parts involved and less likely to screw-up the design.

The is the real answer. Did the designer know what he was doing or not? They can both be very good.

Steve N.

 

[Ori]

Quote:

Originally Posted by Robonaut /img/forum/go_quote.gif Based on this logic, though, wouldn't you want to always get gear with XLRs since it offers you the option of whether to use them or not?

No. Based on my logic, I might try balanced gear but not use it if the XLR connection is inferior. I'd rather then go with simpler gear which supports single-ended-only connection.

 

[Sovkiller]

I'm gonna quote my firend Phil hope he doesn't mind, of some of his thoughts on the balanced topology and some implementations on his amps:

FAQ Balanced Output

--------------------------------------------------------------------------------

(Q) From various sources, the advantages of having a balanced amp with a differential output seem to be:
1.double slew rate due to nature the inverting and non-inverting signal being of opposite phase

(A) The Reason the slew rate is twice that of a non-balanced output is because slew rate defines how many volts in one time period the amplifier can slew or retrace from the positive peak of the waveform to the negative peak. In order for the slewing rate to be improved the Amplifier must be able to slew more voltage in the same amount of time. This is done ether by making the waveform shorter in Duration (Higher in frequency) or larger in amplitude (More output Voltage) the slew rating of an amplifier is mealy the measure of how high in frequency a given Amplifier can maintain a specified output voltage, If the available output voltage remains unchanged then slew rate improvements must come from making the amplifier faster. If the amplifier cannot be made faster the slew improvements can come from making the available output voltage the amplifier can produce greater. By ether doubling the high frequency performance of the amplifier or doubling the available output voltage the slew rate will be twice that it was before. It is interesting to note that since improving the available output voltage and high frequency performance both result in higher slew rate specs that by improving the available output voltages by twice in addition to improving the high frequency performance by twice with give 4 times the slew rate as ether voltage output or speed improvements alone achieve. The answer to this question is that by virtue of balanced output providing twice the output voltage for a given Amplifier configuration than a similarly configured non-balanced amplifier the slew rate of the balanced amplifier is now twice due to twice the available output voltage.

(Q) 2.(in theory) four times the power due to half impedance seen by each amp (per channel) leading to twice the current and twice the voltage due to the signals being of opposite phase

(A) this is actually one of the disadvantages of balanced operation is that since the amp is seeing half the load impedance that a non balanced amp would for the same transducer the lower impedance limit the amplifier can safely drive is now also twice what is is in the unbalanced configuration and thus the load driving ability of the Amp is now reduced by half. It is a misconception to believe balanced amplifiers have twice the output current capability because reality is that the current demands placed on the amp are greater. This causes current limiting and compression in the amp because its power supply and output current capabilities remain unchanged where as the demands placed upon it have just got worse by twice. This is why a typical loudspeaker amp will have a low impedance load tolerance of twice what it would operated unbalanced, thus Amplifiers able to be used with Loudspeakers of 4 ohms minimum are now restricted to loudspeakers with twice that impedance or 8 ohms.

(Q) 3.push-pull output supposedly has more authority on the headphone driver membranes

(A) this is because of more available output voltage results in more power and thus more power adds authority and dynamics however for the headphone amplifier obtaining any required amount of power without resorting to balanced output is obtainable with higher voltage power supply Operating voltages. The Headcode operates on supply voltages that are twice that of the highest voltages used in op-Amp based Designs where as the op-amp is limited to supply voltages of 24-36 volts rail to rail or +/- 12 to +/- 18 Volts the Headcode operates on +/- 36 volts (72 volts rail to rail) and as such will provide twice the output voltages of an Op-Amp based Design operating on its highest safe voltages of +/- 18 Volts and are typically operated on +/- 12 to 15 volts.

(Q) 4.no common ground for better common-mode-rejection and less channel crosstalk (although I have read that the PPA counteracts this with an active ground)

(A) This is one advantage of balanced operation that has some merit and to address this I introduced the ground channel in the PPA. Under development is the next generation of this same technology first introduced to the High End Headphone Community in the PPA, LaRocco Audio’s TRIAD topology continues with improvements to the TRI-Mode of operation of which was first introduced in the PPA. With the addition of one complete additional power supply including power transformer and one complete amplifier channel known previously as the ground channel in such amps as the PPA, the Headcode Dual mono can become the Headcode TRIAD. Once the TRIAD topology is refined to the High standards set by the Headcode Larry at LaRocco Audio will offer TRIAD upgrades. The Headcode Dual mono however is unchained from most of the Limitations imposed upon the PPA such as a single power supply for the entire Amplifier to accommodate portability as the PPA is a portable amp. It was primarily the desire to have an Amplifier that will tolerate poor quality wall adaptors that resulted in the Technology used in the PPA. The Headcode by virtue of two completely separate power supplies including the power transformer result in elimination of any inter-channel reactions.

(Q) 5.6dB more dynamical range (picked that up from rane.com, I’ll have to check on the exact reason but I suppose it has to do with the increased voltage)

(A) This is true since more voltage is available the dynamic range is also improved. However obtaining more voltage dose not requires a balanced output but is obtained in a more linear fashion with more power supply voltage.

(Q) [Just a quick check whether I got the terminologies right: Differential simply means that there a two signals being opposite of phase (+2V and –2V for instance). But then again, the “differential amp” is defined as a amp that takes a negative and positive input signal and one output that responds to the difference between these signal. So does that mean that “differential” only refers to the input? Bridge-tie load means that the speakers/headphones are driven from two outputs via push-pull and with a inverting and non-inverting signal (e.g. +2V and –2V). And balanced is just a general term applying to both driving issues as well as balanced lines.]

(A) Yes just as the term is used to describe the differential of your automobile. If your vehicle is raised upon a lift while in a service station you cold rotate one rear tire and watch the other one turn the opposite direction on rear wheel drive vehicle.

(Q) Now, I understand that these kinds of shortlists only cover the theory but not the actual implementation. Still, balanced mode seems to be everybody’s darling at the moment especially because most audiophiles understand enough electronics to grasp the basic concepts like “quadruple power”, “twice slew rate” and “no common ground”, and to be honest, they do sound quite convincing in theory, especially the slew rate part. I wonder whether truly balanced operation of headphone amps also has its disadvantages seeing that many state of the art amps like the RP010 or Angstrom amps are not balanced either. And finally, why did you decide to use a non-balanced single-ended output for the Headcode?

(A) If you consider that each channel of the Headcode is comprised of true fully symmetrical complementary Circuits from the input to the output and are only tied together at the input and output and otherwise contain completely separate however symmetrical circuitry. In Actual operation by virtue of a fully complementary PNP NPN relationship formed between semiconductor devices of each half of the internally balanced circuit the Headcode has symmetrical rising and falling slewing rates and similar transfer function of both the positive and negative half of the circuit. In conventional balanced output configurations the operation of two independent amplifiers in both the inverting mode and the other in the non-inverting mode results in each half of the waveform being processed by Amplifiers with completely different slew rates and frequency response. Even ideal matched dual IC op amps containing two perfectly matched Amplifiers on one chip are subject to this if one Amplifier of the dual is operated in the inverting mode and the other in the non-inverting mode the one operated in the Inverting mode will typically have higher slew rates and bandwidth in addition the spectral content of the distortion spectrum will also be different than the one operating in the non inverting mode. This is thus processing each half of the waveform differently in the ultra critical time domain.