Testing balanced versus unbalanced headphone amplifier drive

[amirm]

Hello everyone. I like to investigate whether there are real improvements to be had in using balanced output on amplifiers and corresponding balanced input in headphones. I am seeking information on best way to organize this.

I currently have one headphone amplifier with balanced output: Topping DX7. My son also has the Oppo so I can get at least two pieces of gear for this testing.

Question is, am I testing the analog output of the amplifier in balanced versus unbalanced? Or do I need to measure what is played out of the headphone?

On the later, I have the Sennheiser HD-650. I have seen balanced connections for it but they were pretty expensive. Do they make cheap wiring for it? And at any rate, what do people recommend here?

I have a minidsp headphone testing rig. THe problem with it is that any disturbance of the headphone changes what is captured. So if we are going to AB test, we need to not touch the headphone. Thinking out loud, I am assuming that I can merge the balanced connection to unbalanced at the end of the cable and not disturb the wiring to the headphone.

Anyway, appreciate any and all insight and whether you believe in merits of balanced or lack thereof.

 

[Candlesticks]

I'd like to see the affect of balanced and unbalanced on crosstalk with very low impedance, high sensitivity in ear monitors.

Fiio has a balanced MMCX cable that isn't expensive.

These are cheap, use MMCX, low impedance and quite high sensitivity.

AtomicBob did balanced and unbalanced measurements of the Schiit Jotunheim.
https://www.amazon.com/FiiO-Balanced-Headphone-cable-connectors/dp/B073GQ1BYB

 

[Vincent Kars]

For me it starts with “balanced”
Normally a headphone has a jack hence TRS
This means a common ground for L+R
A “balanced” wiring means not only separate hot L+R but also separate ground L+R
Exactly as we connect speakers to an amp.
What is "balanced" about the 2 wire connection between a speaker and a amp?

 

[SpeedyRodent]

Hello everyone. I like to investigate whether there are real improvements to be had in using balanced output on amplifiers and corresponding balanced input in headphones. I am seeking information on best way to organize this.

I currently have one headphone amplifier with balanced output: Topping DX7. My son also has the Oppo so I can get at least two pieces of gear for this testing.

Question is, am I testing the analog output of the amplifier in balanced versus unbalanced? Or do I need to measure what is played out of the headphone?

On the later, I have the Sennheiser HD-650. I have seen balanced connections for it but they were pretty expensive. Do they make cheap wiring for it? And at any rate, what do people recommend here?

I have a minidsp headphone testing rig. THe problem with it is that any disturbance of the headphone changes what is captured. So if we are going to AB test, we need to not touch the headphone. Thinking out loud, I am assuming that I can merge the balanced connection to unbalanced at the end of the cable and not disturb the wiring to the headphone.

Anyway, appreciate any and all insight and whether you believe in merits of balanced or lack thereof.

From my experience and research, I am doubtful there is any meaningful benefits to running a balanced headphone setup. I tried it myself on a few headphones and it seemed more impractical than anything with the xlr connector.

Just a few comments with the test setup:

Balanced Cable
Cheapest way is to cut the 6.5 mm headphone jack off the end of the existing hd650 cable and solder a 4 pin male xlr connector to the headphone side. Leave enough wire on the other cut end with the 6.5mm jack to solder on a 4 pin female xlr connector. This enables you to switch between balanced and unbalanced.

You could also buy a pre-made balanced cable for the hd650. If you want to switch easily back to unbalanced without disturbing the headphone test rig, you will have to build or buy a xlr to 6.5mm adapter.

https://www.ebay.ca/itm/SENNHEISER-...349462?hash=item4652d43456:g:DggAAOSw7RdZwBb4

https://www.aliexpress.com/item/VE-Basic-HD650-HD600-HD580-HD414-HD420-HD430-HD525-HD545-HD565/32806755404.html?src=google&albslr=222075327&isdl=y&aff_short_key=UneMJZVf&source={ifdyn:dyn}{ifplala}{ifdbmBM&albch=DID}&src=google&albch=shopping&acnt=708-803-3821&isdl=y&albcp=658707750&albag=38901860132&slnk=&trgt=68416666751&plac=&crea=en32806755404&netw=g&device=c&mtctp=&gclid=EAIaIQobChMIsbLrteKU2QIVBbbACh1DKAPZEAYYASABEgIG-fD_BwE

Measurements:
I would say both out of the headphone amplifier out and through headphones would be insightful.

Straight from the amplifier output:
Measurable differences between the balanced and unbalanced outputs. ie. differences in distortion, noise, output impedance, crosstalk, etc

Out of the headphone:
Maybe the impact of the measurable differences at the headphone. ie. distortion+noise, frequency response, impulse, square waves at different frequencies, etc.

Headphones:
If possible, I would suggest the HD800 for high impedance if you can get your hands on one. The HD650 has a slightly elevated thd+n at lower frequencies.

For low impedance, maybe something like the Hifiman HE560.

Amplifier
The topping appears to be fully balanced. Not sure about the oppo. This test is only useful with a true balanced headphone amplifier rather than a amplifier with different output connectors.

 

[kristoffer]

I haven't tried it myself but I have read a few recommendations that say it is simple to terminate the the stock HD-650 cable[0] with a 4-pin XLR connector. It' about $10 + XLR connector and work.

I recently bought a balanced cable for Hifiman HE-560 from Ebay for $55 (delivered in under 14 days from China to northern Europe). So more expensive than the DIY-option but the seller also has one for HD-650 [1].

[0] https://en-us.sennheiser.com/accessories--hd-650--cable
[1] https://www.ebay.com/itm/Balance-ca...phones-/253273320225?var=&hash=item3af8442b21

[Edit] SpeedyRodent was speedier with the DIY recommendation

 

[RayDunzl]

Normally a headphone has a jack hence TRS
This means a common ground for L+R

My HD650 has a TRS jack, but two cables (4 wires) to the drivers.

So, the common point is at the amp. That shouldn't be an issue.

It's not balanced as in opposing signal polarity (ground opposes the active leg, though), but avoids the pitfall of the varying voltage loss caused by the other channel's current flow that occurs (to some degree) on a common cable.

On the other hand, I usually have it plugged into a 10 foot extension, which is 3 wires, so, duh.

I looked at it, thought about it, and that was it, till now.

 

[Brad]

Don’t forget the 6dB signal gain due to balanced connections (from summing the signals)

 

[Blumlein 88]

For me it starts with “balanced”
Normally a headphone has a jack hence TRS
This means a common ground for L+R
A “balanced” wiring means not only separate hot L+R but also separate ground L+R
Exactly as we connect speakers to an amp.
What is "balanced" about the 2 wire connection between a speaker and a amp?

Some amps will be balanced in the sense you don't have + and gnd. You actually have Signal+ and Signal - as the outputs are opposite phase of each other. Or differential drive if you prefer.

 

[DonH56]

A differential connection does not need a signal ground reference. The shield ground serves as the shield but is not important to the signal as is easily proven by disconnecting the shield at one end. You'll still get the same signal out the other side.

There are also various flavors of quasi-differential circuits that drive and source XLR "balanced" connectors but those are not truly differential and have drawbacks like higher sensitivity to ground and common-mode noise.

 

[Jimster480]

You want to test things like impedance curves and power output.

 

[SpeedyRodent]

A lot of headphone amplifiers are advertised as "balanced" but really only have a xlr connector. As mentioned in some of the comments above, balanced will have differential output. So each channel will have a non-inverting and inverting signal.

Imo the benefits of using differential output in a headphone amplifier application is debatable. Some companies like Benchmark state there are no benefits to balanced output in headphone applications.

In the guitar world, I've seen performers use "DI" boxes to convert to and from balanced/unbalanced signals. Sometimes this helps eliminate grounding loops causing hum.

 

[amirm]

AtomicBob did balanced and unbalanced measurements of the Schiit Jotunheim.

Thanks but there are a million graphs there with cryptic notations. I can't tell if "balanced" means inputs/outputs or headphones. Can you point to two graphs, one with balanced headphone out and another with unbalanced?

 

[Nathan Raymond]

In case it might be helpful, here's some information I had collected for myself previously in my research on the topic. I'm including some choice quotes, but I recommend going to the links to get the full story on each.

On Dec 5, 2015 Tyll Herstens gets into the nitty gritty of whether a balanced headphone amp should even be called balanced and takes some responsibility for the situation:

http://www.superbestaudiofriends.or...iscussion-for-noobs-and-boobs.733/#post-18938

"To some extent, I'm responsible for this word showing up in the world of headphones with the introduction of the Blockhead Many years ago. Bottom line: I don't think we can ever use this word in a technically correct way as it applies to balanced headphone amps. I guess I have to side with Steve E.'s take on the subject...technically...that 'bridged' is the more correct term---but that horse left the barn long ago..."

On Dec 6, 2017 Bob Katz had this to say:

https://www.innerfidelity.com/content/katzs-corner-episode-21-how-insensitive-part-2

"You've probably heard this before, but it bears repeating: There is no such thing as a balanced headphone, because a headphone is not an active device, it has no innate reference to ground. A better term for this type circuit would be "symmetrical push-pull". I am agnostic about balanced versus unbalanced headphone amps. If an amplifier can deliver the sonics at the required levels, then it should not matter if it's balanced or unbalanced. A balanced amp can produce 6 dB more level with the same power supply as the unbalanced. But the unbalanced amp's designer can use a larger power supply and achieve equal level. Sure, it's nice to say that an amp like the Mjolnir is fully symmetrical from input to output, but does that translate to a sonic advantage or just an engineering goal? I would never dare to claim an amplifier's sonic superiority is due to its balanced topology. The variables cannot be isolated..."

His Episode 22 which will have the measurements from his amp roundup are still pending (he's skipped to 'Episode 23' for now).

The CEO of Lake People / Vioelectric put together this paper with some schematics, where he says:

http://violectric-usa.com/download/Balanced Headphone Amplifier Explained.pdf

"'Balanced' amps are also called Push-Pull Amps or BTL amps (Bridge-Terminated-Load)."

If you go looking, there are lots of websites out there were people document mods to headphones to get a 'balanced' cable to them, such as this:

https://robrobinette.com/HeadphoneMods.htm

Also worth noting is that only audiophiles throw around the term 'fully balanced amp', which isn't something an electronics engineer would recognize as having any meaning. Example in use:

http://www.6moons.com/audioreviews/balancedheadphones/one.html

Here's an example of the type of unsubstantiated here-say floating around on the topic as well:

http://www.instructables.com/id/Balanced-Headphones-for-About-130/

"Balanced mode delivers 4x the power to your voice coils, cuts your signal-to-noise ratio by 4, doubles the signal slew rate, and eliminates channel-to-channel crosstalk. The result is a precipitous drop in your noise floor, almost unbelievable increase in audio detail, much fuller and precisely defined soundstage and imaging, and an overall warmer, richer and fuller sound reaching your ears."

Hah! And for those into trivia, Moon Audio (which has as one line of their business headphone cables) lists quite a few of the balanced connector types:

https://www.moon-audio.com/single-ended-versus-balanced-connection

• Dual 3pin XLR connectors
• Single 4 pin XLR
• RSA/ALO Male Connector
• 2.5mm TRRS
• 3.5mm TRRS mini
• iBasso balanced
• JHAudio JH3a connector

However they missed the Japan Electronics and Information Technology Industries Association 4.4mm balanced connection standard that was introduced in 2014:

https://www.headfonia.com/picture-sunday-4-4mm-balanced-effect-audio-ares-ii/

Anyway, lots of cable options out there. Since my only balanced amp is in my Onkyo Rubato DAP which has a 2.5mm TRRS, that's what I've been purchasing with regards to cables. Obviously you can convert to XLR with something like this:

https://www.ebay.com/itm/4-pin-XLR-...FI-Cable-headphone-Audio-Adapter/202182913527

Aftermarket balanced cables for Hifiman HE-400i (and similar types of headphones that have 2.5mm TRS sockets on the cups) start at around $43:

https://www.ebay.com/itm/2-5mm-Trrs...400S-HE-400I-HE560-HE-350-HE1000/202122585731

The Fiio RC-MMCXB cable goes for about $31:

http://www.fiio.net/en/products/58

I made the mistake of thinking that the Fiio cable would work with my Onkyo ES-CTI300 headphones (which have MMCX connectors on the cups) but they had a loose fit. Cable worked fine on a couple of IEMs I have though. However for $15.50 I was able to get the "19 core 2.5mm balanced" MMCX cable here which was advertised as compatible with the ES-CTI300 and sure enough it works great:

https://www.ebay.com/itm/MMCX-Silve...Cable-3-5mm-Audio-2-5mm-Balanced/262966678773

I use a Fiio L26 adapter to use my headphones with balanced cables on an unbalanced source. The nice aspect about 2.5mm TRRS balanced becoming somewhat common and downmarket is that prices tend to be lower than for the XLR stuff. The Fiio F5 IEMs come with both an unbalanced and balanced MMCX cable (along with a full suite of accessories) for $60 street. The Pioneer SE-MHR5 over-ear headphones went on special for $80 during the 2017 holidays and come with both a balanced (2.5mm TRRS) and unbalanced cable.

I'm really looking forward to a review of balanced vs. unbalanced amps. If you're ever looking for some headphones to torture-test amps, the ones I know of offhand with the worst impedance variations are the Audio Technica ATH-R70x (peaks at over 900 Ohm):

https://www.innerfidelity.com/images/AudioTechnicaATHR70x.pdf

And in the 20-year vintage category, the Aurex HR-V9 gets over 4000 Ohm:

https://www.innerfidelity.com/images/AurexHRV9.pdf

 

[amirm]

Nice post. Indeed I had no idea what people were talking about when I first heard the term "balanced" with respect to headphones. We don't call speakers balanced so what the heck was this about with headphones???

 

[Wombat]

I thought I would sneak this in: https://audiosciencereview.com/forum/index.php?threads/balanced-line-technology.2329/

 

[DonH56]

Phono cartridges can also be considered "balanced". For them, and headphones, and speakers, I think it is more about marketing than engineering. Actually, I think that about a lot of things these days. For headphones it may make some sense as a balanced connection potentially provides better shielding for the cable, which then attaches to a long loop of wire around a magnet or spread out on a panel, where I suspect any benefit is lost...

 

[Dismayed]

Benchmark has an opinion on balanced headphones.

“But, this does not mean that balanced and unbalanced headphone amplifiers will sound the same! In many cases, voltage-balanced headphone amplifiers will produce more noise and more distortion than single-ended amplifiers of an identical design. The reason for this is that two separate output amplifiers are required in a voltage-balanced amplifier, and each must drive one half of the transducer's load impedance. The output noise will double because there are two amplifiers instead of one. The damping factor will degrade by a factor of 2 because both amplifiers contribute to the source impedance of the balanced amplifier (output impedance is doubled). Distortion will usually increase because each amplifier is required to drive half of the impedance that would be seen by an unbalanced amplifier.”

https://benchmarkmedia.com/blogs/application_notes/audio-myth-balanced-headphone-outputs-are-better

 

[mindbomb]

I think the main practical difference is going to be crosstalk. But I think what people want to see is thd+n x mw graphs so they can see the difference in power.

 

[sergeauckland]

Phono cartridges can also be considered "balanced". For them, and headphones, and speakers, I think it is more about marketing than engineering. Actually, I think that about a lot of things these days. For headphones it may make some sense as a balanced connection potentially provides better shielding for the cable, which then attaches to a long loop of wire around a magnet or spread out on a panel, where I suspect any benefit is lost...

MC cartridges are very close to being balanced. Balanced doesn't just mean two wires, it means that all the parasitic impedances to ground are as close to identical as possible such that the common-mode rejection is high. MC cartridges and microphones will be very close to the ideal, MM cartridges won't be, as the fixed coils are normally encased in a mu-metal box, grounded to the pickup headshell or to one channel, and so inherently unbalanced.

Regarding headphones, it's a total nonsense, and just marketing. The ONLY benefit of driving the two headphone elements individually rather than sharing a common return (ground) path is the tiny amount of crosstalk that results from the common impedance. This can be calculated from knowing the resistance of the common connection and the relative resistance of the driven elements. It is utterly negligible in any practical scenario.

For headphone extensions, I've made up a few using either three core mains cable or screened balanced cable, and whatever the additional shared impedance, the effect is still negligible for the 10m extensions I've made.

I feel this is just another marketing exercise and a way of creating interest in buyers who don't understand what any of this really means. It allows the manufacturer to charge a premium price for something that is of no benefit. It's of no detriment either, but technically pointless.

S

 

[DonH56]

Benchmark has an opinion on balanced headphones.

“But, this does not mean that balanced and unbalanced headphone amplifiers will sound the same! In many cases, voltage-balanced headphone amplifiers will produce more noise and more distortion than single-ended amplifiers of an identical design. The reason for this is that two separate output amplifiers are required in a voltage-balanced amplifier, and each must drive one half of the transducer's load impedance. The output noise will double because there are two amplifiers instead of one. The damping factor will degrade by a factor of 2 because both amplifiers contribute to the source impedance of the balanced amplifier (output impedance is doubled). Distortion will usually increase because each amplifier is required to drive half of the impedance that would be seen by an unbalanced amplifier.”

https://benchmarkmedia.com/blogs/application_notes/audio-myth-balanced-headphone-outputs-are-better

Some of his does not fit the normal engineering view, something I find a bit surprising from Benchmark.

Differential may have an additional amplifier stage depending upon how it is implemented, but signal and noise are both doubled. Noise is uncorrelated, however, so after doing the math the signal increases by 6 dB and noise by only 3 dB so you actually gain 3 dB in SNR. What they may be describing is that, due to the effective increase in gain, you turn down the level for the same volume and thus the equivalent noise floor is higher.

The damping factor, drive, and distortion comments I agree with and are equally applicable to bridging amplifiers, They represent drawbacks of operating in bridged mode. You can design an amplifier for that mode, natch, and one could argue if any of it even really matters for headphones. But, I do not know what a typical headphone's impedance looks like, nor what sort of output impedance most headphones amplifiers exhibit.