What are the technical advantages of a balanced headphone amplifier? - Page 6

seems to be a lot of confusion between what a balanced amp is and what a balanced line is.

A balanced amp uses an individual amlifier stage for each polarity of the signal so a stereo amp is actaully four amps and not two as would be normal (left amp/right amp).

A balanced line is just a balancing mechanism at the output of one stage driving a balanced cable which is then "received' at the other end by a balanced receiver.This is strictly for noise cancellation on long cable runs and would have zero benefit on any headphone cable.Headphones do not have gain but are a passive deivce so can not generate noise.

An amp that uses a Push-Pull topology or a push-pull output is not a balanced amp even if it has a balanced line input but simply a balanced input single ended amp with a push-pull output stage for more power and will still use a common ground reference unlike a true balanced stage which will not and can not have a common ground.
It would self destruct because the "-" signal amp replaces the ground channel in a conventional amp and even using a A/B switch on the output to select between two events will destroy the amp if it is not a true floating ground 4PDT switch.
True balanced also means a single headphone cable is out or you again introduce a short to the "-" side amp

Just a little refinement: There is a possibility of a small amount of noise to appear on the cable going to the headphones, but the output impedance of the amp is so low that effectively shorts out any noise signal on the headphone cables. This is why it's better to have long cable runs on a headphone extension as opposed to extending the amp with long interconnects. (Though that means you have to have long arms (or a remote) to adjust the volume control.)

If ever an amp needs to be close to hand it is the headphone amp maybe better named a "personal amp" which it really is.Why ? No idea.
It just feels right being there and not across the room.Might have something to do with you listen to loudspeakers from a distance so the amp also seems right at a distance but headphones being more directly connected to the human means a direct tactile connection to the amp is also iimportant to me.More chairside than rack/shelf mount plus never once have I tripped over a headphone cable and yanked the amp to the floor with a "local amp" but have more than once with a headphone extension when after yanking the cans then running to answer the door tripped on the cable which caused the somewhat lightweight amp to be riped from the rack and in the ensuing swearing at volume by the time I answere the door no one is there which p*sses me off even more !

So I use balanced lines for all runs over ten feet with both single ended RCA and balanced XLR/Phone "combo jack" inputs but cheap it is not.Adds a C-Note or better to the cost of every headphone amp just in parts

No doubt. The problem of extending a headphone amp away from your stereo is one of the Achillies Heels of headphone listening.

Is this an insiders leak, Headroom soon to build sources with headphone amps included?

I coud be reading this wrong (and I am sure Tyll will correct me if I am ),it is not that the local CDP or other Source to Headphone Amp is the problem but more what way to go in transmitting to the actual listener who more often than not wants a comfy chair,a side table and a reading lamp while listening and not be nudged right up to the equipment rack as if it is about the equipment rather than music in comfort.
With the average home headphone cable coming in at 3-6 ft the only options are presently :

1-A Headphone Extension Cable.This could be a recipe' for disaster unless you live alone or lock out family members who may walk by but is the best electronically sound methos.you also lose the ability to adjust volume in small amounts unless you either get up and walk over to the headphone amp or use a remote control which means usually comprimised sonics.Usually because there are some good ones beginning to show up in the analog domain and if you are already in the digital domain easier to add this feature without doing any sonic damage.

2-Local amp (side table) with long lines from the system to the amp input.Usually not a real good idea on two levels : current drive and noise pickup.This can also be a comprimise unless you attack it with money due to no part of the Driver/Balancing Mechanism/Long Interconnects being cheap.I save some loot by doing my interfaces "in house" via the dIY route but even there the cost added is over $100 stereo not including the interconnects.

Those with a computer based system are right there anyway so this "fix" does not apply but for those who listen off the main rig and in style a very real problem with no "correct" solution.Options,solutions,but no single best for all solution.

OK! I think I got it!

So my Balanced Line Out may not be a true balanced line out, but as it is, its main purpose is to be used for XLR cables to extend way out to a power amp/receiver with XLR inputs. We see this all the time at concerts, churches, clubs, etc.

In other words, no real benefit for headphones, but if I ever wanted to hook my Melos up to an amp across a room with low noise floor, I should use the XLR line outs eh? Cool!

Sounds right to me, Rick.

We're getting a little off topic, but there is another way now to ship audio around the house and that's the Apple Airport Express. If your satisfied with shipping audio out of your computer, this little device will alow you to get your audio pretty much anywhere in your house where you have a power outlet.

Nooooo! Though shipping a digital signal around with an AirportExpress (as I mentioned above) would let you take the digital out to a headphone amp with built in DAC. ANd you know where you can get one of those.

We have toyed with the idea of wireless digital transmission, but nothing is on the drawing board. (Not that I'd leak that if it were.)

It would seem so "wabbit"

Yes.Noise does not know squat about polarity so will be common to both parts of a balanced signal.Any noise picked up between the "transmitting" end and the "receiving" end will be canelled out by the process of converting the signal back to single ended.Simple difference amp theory (only the difference is amplified not the common mode signal).
That takes care of the balancing parts.For very long runs you still need to have enough "drive" to get the signal down the line to the other end without attenuating the highs too much and that means current to overcome the resistance of the line itself.That is where a cable driver comes in which usually a headphone amp makes a good one of.

Possibly a bit off topic but worth commenting on just the same.The move to USB is coming fast as is computer as server for ALL A/V needs.Central control,wifi and hardwired transmission,computer as entertainment center with the hard drive being in control of all.I won't be giving up the anlog rig anytime soon but to deny where home audio is headed would be to be an ostrich.Would not surprise me to see headphones with built in wireless interfaces in the next couple of years totally taking over the mid-fi market (but THAT is fodder for another thread ).

Total elimination of Distortion has not been done yet by any presently available Technology, Moreover since the amplifiers output impedance is twice that of the same amp operated as a bridge amplifier then all else being equal Distortion will actually be greater in the Balanced configuration because of the higher output impedance. Adding to the Distortion issue is the fact the Amp now sees a load that is half the impedance than the same load and amp operated unbalanced again more Distortion is the result.



Tyll! my friend please consider that the reason the slew rate is twice in the Balanced mode is because of the increased output voltage swing however the actual slew rate under load can be and often times is allot less than twice and can be actually less than an unbalanced configuration. Getting rid of the common Ground with a separate ground from the standard 3 conductor stereo phone Jack will have the same effect regardless of what configuration the amp is operated in, This is because the DC resistance of the standard 3 conductor headphone cable is allot greater than the output impedance of your amp. It is nice that a balanced configuration can eliminate this however requires a aftermarket cable set up for a 4 conductor cable. I have found most stock Headphone cables have about 2-3 ohms of resistance this will completely swamp any benefit gained from a balanced configuration, IMHO this improvement is available simply by a cable upgrade as opposed to a new amp.

No it will not.

Oh how Very true the primary reason for balanced operation in the Pro market is because of the requirement to run cable long distances in the presence of Noise from Lighting and such.


The extra drive and punch of balanced amp vs the same amp in the SE mode is realized in headphone amps are from the higher available output swing. As Tyll pointed out unlike most Loudspeaker amps most Head amps have an overkill of output current.

In conclusion a balanced system can be made providing that at no time is the Audio inverted and as such will require taking the balanced output from the DA chip to completely identical Amps and use a precision 4 stage volume control or Digital volume control. And each amp must be matched in gain using 0.01% better parts or the common mode rejection will be degraded in addition to crossover distortion if the gains differ by any amount and unless you hand match this as is typically done in test instrumentation. The question should not be what topology should I get but should be what can I spend and what sounds best at that price point. So many times folks concern them self’s with technical stuff that’s so hard to understand it can take away from the pleasure of music. IMHO a home system needs not to be balanced in order to sound good. A balanced Amp often times will sound different than the same amp operated SE however it may just be different and not actually any better. The same thing can also be told of Slew Rate This spec is completely unimportant for most Headlamp applications since the amp will never see slew rates above 1 volt per uSec except in the presence of high frequency signals like the leakage from your digital sources then some overkill hear would be a good thing but even then 10 Volts per uSec is plenty good. Also consider that slew rate of an Amplifier is more often than not limited by the voltage gain stage and more specifically the input stage operating current and the capacitance of the following 2ed stage and or Compensation capacitors. It defines the frequency at which the output voltage of the amp will change from a sign wave to a triangle. If one uses two amps in Balanced or Bridge mode then the input stage sees half the voltage for a given output voltage and thus allow this frequency at which the waveform will change from a sign to triangle and thus twice the slew rate there is no magic hear. The fact that Amps having high slew rates may sound better than ones with less of a slewing rate is attributed to the changes to the circuit required to obtain the higher Slew Rate like using a Low capacitance 2ed gain stage device or operating the input stage at more current to drive the capacitance.

Wow quite a lot of info but very important comments there I think. Most of the technicalities are over my head, but I wanted to just comment on a couple of things that I find particularly interesting:

I had read about this, that balanced configurations while fighting noise etc. paradoxically might possibly be more prone to distortion.

That I find very enlightening, because it somewhat makes a case for a cable upgrade possibly having a larger effect than an amplifier upgrade in certain cases.

Also resonates with what I have always read about the reason balanced circuits started to be used in audio.

I wonder how the headphone out of the Creek 5350 SE might sound with appropriate source and high end headphones. That amp features a slew rate of 50 V per uSec. From what you indicate it seems way overkill. But does that number possibly apply only to its speaker outs, and not to its headphone out? Though it might never put 50 V in its headphone out, I guess its possible that it featured the equivalent slew rate of let's say 1 V per each 50th of a uSec.

Another question, is the slew rate of SS amps in general faster than the slew rate of Tube amps, let's say on average? Or maybe we really can't generalize between tube vs. ss as far as slew rates go?


Ok now I'll shut up and will let the experts continue their exchange of thoughts and ideas.

Btw thanks for this exchange guys!

Obviously, this topic is heating up as more people are considering Balance operation for their headphones. So, I think I'll try to respond to some of ppl's comments. Mind you, there are many here who are far more qualified to discuss this in technical terms. I'll try to wade into these deep waters, but I may have to get Joe W. (our engineer) to respond if I start drowning.

Very few headphone amps (none that I'm aware of, maybe it's some of the tube amps that have output transformers) operate convertable in bridged or balanced mode. So my comments are primarily pointed at single ended amps and balanced amps that are using the the same topology as the single ended design but doubling up on the electronics for the inverted channel. In this case the output impedance for each channel remains the same but because they are effectively in parallel, the output impedance of the effectively halved as you now have two amps driveing one driver coil.

But not in the above described situation where you are using both an inverted and non-inverted ouput amp stage. In that case your slew rate double becauses you get the slew rate of the non-inverted channel going one direction plus the inverted channel going the other direction.

I agree here completely. A mindless rush to balanced forces people to replace their cable and give the credit of improved sound to the amp. But cable replacement on a headphone for an unbalanced amp is deffinately a better step towards sound quality improvements in terms of price/performance.

This is a very good point: Componant matching is much more critical in a balanced design; you not only have to double the electronics but you have to make sure that the +/- signal chains are critically matched with each other----and that will cost you money. Also, as ppl so rightly notes, in order to make this all worth while the source signal MUST be generating the +/- phases correctly. If you have a balanced DAC it must have FOUR DACs, one for each phase. Therefor a move to balanced must be done with the ENTIRE audio chain in mind. There is no "cheap" way to get balanced and make it worth while. This is a game for VERY healthy wallets.

As ppl says, GREAT reproduction can be had in single ended headphone amplification without the expense of exotic headphone re-wiring and sources.

I can't speak for other manufacturers, but I will say that in the manner we do balanced it does sound different AND better. We do use the particularly brute force method of just doubling up amp stages and not a bridging/balanced type amp, but it is clearly better. I think that's why there is so much interest in balanced operation, because, when properly done, you can get tremendous gains in performance, but only once you get up into the rather stratospheric price ranges.

Just a quick question: Does the DAC in the balanced Max have four Dacs? Because if yes, that would sound like a very good deal for "only" $500.