*Excellent explanation* of 'balanced inputs, 'differential headphone output' and 'balanced topology'. I always wondered what those 4-pin connectors for headphones really were. My headphones get loud enough with no noticeable distortion, so I don't need differential headphone output. Thanks for doing this video, Amir. All the best, Rob
The analogy to the 'care about nuts and bolts in your engine' was not good though. It's common for engines to fail due to bad quality head bolts (especially performance engines) so car people absolutely care about what kinds of studs and bolts are used in engines, not just the performance 😂 The differential output he was talking about is akin to small car amps which operate using push-pull dual amps per channel. That way you can get 14,4 watts (4 ohms) of power from a simple cheap 12V power source.
@@AudioScienceReview Really could not agree more. The Emotiva Differential Reference amplifiers make balance interconnect balance XLR integrations sing!
That's one of the kindest remarks I have read. :) Had to get my Covid shot and a few other things that kept me from producing videos in the last few days.
I have learned more about audio science from this channel in the past week of binging than I have in 7-8 years of internet "research". Thank you Amir, great info, please keep up the awesome work!
Yes, explains balanced/unbalanced characteristics for headphones and equipment that’s easy to understand. Also good to know the amount of benefit in going bal input and output depending on your circumstance.
Thank you! This was wonderfully informative and reaffirmed what I’ve learned from the ASR forums while teaching me/confirming what I’ve heard from my own gear regarding crosstalk and power requirements for my headphones.
This explains why I don't hear a difference between my balanced and unbalanced headphone cables. I'm glad my balanced cable is more practical (supple), so not complete waste of money.
The Topping A90, that I also have and adore, takes advantage of this in some way in its outputs since its 4-pin headphone XLR also puts out about double than its SE, TRS output.
All of this knowledge that is gushing out of this fountain is awe-inspiring and all of that feel good jazz Amir! I'll be super keen to also understand how switching to balanced connections contributes to much better separated instruments and open sound. Like the difference is night and day for me, subjectively. This is what I experienced on some fairly decent IEMs initially when did some quick A/B listenings and it was like having an acid trip for the first time kinda big! :D It changed the way I experience music for good and motivated me enough to switch my home listening to a balanced one too! Again big thanks for the invaluable reviews and measurements of studio monitors and DACs among others. Super enjoying learning something new everyday abour our fav hobby! More power to you! Surely the snakey businesses won't like you, but you're a super hero to the rare good ones and us small potatoes :D
Amir: another great video that illustrates that there are plenty reviewers that don’t know anything. Also please make a video debunking Directional RCA. Thanks
While I don't always agree with your reviews, Amir, I really do appreciate what you are doing here. You're one of the good guys. Keep the good content coming! Cheers!
Hello Amir! Just a request. Make playlists for every category of your channel (tutorials, reviews, live sessions or whatever etc.). Your channel is growing and in time there will be a lot of videos . I'd much appreciate if you would create playlists of this kind! Cheers!
Another and educational video. I have 2 Stereo power amplifiers, both from the same manufacturer. One is balanced, the other is ‘single ended’. The balanced amp weighs 73kg. The single ended amp weighs 39kg. The balanced amp cost almost twice that of the single ended. Other than weight and dimensions they share virtually identical specifications and measurements. Whilst I take on board the virtues of a balanced design, is not the caveat that twice the number of electronics are needed. Hence double the weight and price. Surely, the more complex and crowded the signal path is, the more the opportunity to degrade the original. Causing greater deviation from the creators intent. So, whilst the signal may be cleaner, it is not as pure. Not maintaining it’s fidelity!
some turntables have din, "balanced" outputs now like the pro-ject x series tables, and they sell a matching phono preamp that accepts this connection. since this is a 4-pin connection it isn't the same as an XLR balanced connection, any thoughts on these? would this help prevent grounding issues with a turntable? most do not have this though so not really an option
My concern with the growing popularity of XLR/ balanced in/out for marketing reasons is there are a lot of cheap bad balance transformers. A cheap amp will be single ended in and use a cheap transformer to convert the balanced in. I'd take a good unbalanced in to a cheap balanced any day. And the best balanced in would be transformerless.
Hi Amir, I watched this video quite a while ago, but have now developed a problem that I would like to ask a question and possibly get some advice from you. I have a holo audio May kte DAC, it has XLR or balanced output. The output voltage on balanced/XLR is 5.8 V rms. Unbalanced/RCA is 2.9 V rms this is on PCM 48K NOS. The output voltage when using DSD 128x is Balanced/XLR 2.9 V rms and unbalanced/RCA 1.45 V rms. When using an AES/EBU cable from my CD player to the DAC I'm hearing distortion when in PCM mode. When I switch to an unbalanced/RCA connection there is no audible distortion. And again when using a balance connection in DSD there does not seem to be any distortion that I can hear, making me suspect it's to do with the relatively high output voltage from the DAC in balanced PCM mode. Obviously I would like to use a balanced connection without switching to DSD all the time, but I am at a loss as to know how to do this and avoid the distortion. Do you have any suggestions? Many thanks Shane.
I'm not a electrical engineer specializing in audio. I am a professionally trained engine builder. Trust me, when it comes to performance you better care about the nut and bolts. I recommend APR products.
Sorry if I missed something here.. I thought the 4 pin connector was originally used to go hand in hand with a balanced topology? You brought that up at the beginning of the video, but neglected to discuss it. Correct me if I'm wrong, but a normal stereo amp is basically two mono amps that share a ground. A true balanced amp has four mono amps, two per channel in a push/pull configuration. (Positive and negative side of the circuit.) This is the reason for doubling the power of a given amplifier circuit, but the real purpose is to completely reject any feedback from the ground side. This is supposed to reduce the noise floor to below "silent" and reduce cross talk. Some of the modern designs with "balanced output" may not use a true balanced topology, like you were saying. Some are simply single ended with an internal adaptor and others such as the THX AAA series, use other tricks (feed forward error correction on the case of THX) to achieve the same results. I agree though, it doesn't much matter what happens inside, if you like what comes out. I don't know what happens inside my A90 for example, but I plainly prefer the presentation over my 789, even if the immediate characteristics are the same. The sense of depth and realism are much better, to me.
He Amir. Pardon. Don’t know how to contact you otherwise. What do you think about ceiling treatment? I have one strong reflection coming from there and was thinking about treating it. I will need to treat by back wall since my sofa is right against that wall. I agree with Toole that I don’t want to treat the side walls. One because they are far away in my room and I don’t want to over deaden the room. It’s nice a lively now. And I listen to classical mostly. Will treating a portion of the ceiling you think deaden the room? A lot of bass resonates from there as well from my measurements. Thanks as I haven’t found much on ceiling treatments and theory
Great video. However the DT990 which shows as one of the more "least sensitive" headphones in your graph is an Single Ended headphone, so buying a 'balanced' output amp (more power) doesn't help to counter the sensitivity issue.
I always felt that the term “push/pull” more accurately described what they called a balanced headphone output. Probably not as marketable though. Thanks.
Amir, have you had any further consideration on the LA90 amp? I’m thinking of using it on insensitive ML at 4 ohms. Do you think this will be okay or do you suggest something another product?
LA90 doesn't have enough power to drive power hungry speakers. So personally I use high-power class D amps. If you play at more moderate levels though, performance is excellent.
1st if someone needs power out of single ended. FA-10 is the way to go. 2nd from what I understand it is better to not have a ground for audio... generally speaking. I know iFi has some special ground filter. Problem with the balanced is that some brands will only focus single ended just as a check box, and you may run in to issues if try use multiple inputs. 789 for example, if you connect a DAC and a turntable, whatever source you user RCA for, will get a lower quality. That is why I like to keep everything single ended, no ground connection, filtered usb signal from dedicated streamer to DAC, and everyone is happy.
Hello Amir, thank you for your videos, they are very informative. What are your thoughts on the BACCH software that enhances the spatial characteristics of headphones (and speakers)?
Balanced output usually has 'better' measurements apart from higher voltage; are these not audible improvements? For example channel separation/crosstalk being 70db SE and 100db BAL. Is 70db already too good and beyond that is inaudible?
Thank you. Question (for future videos, perhaps): What are the risks in plugging in/out analog connections without turning off DACs, Amps, headphone dongles, active speakers?
Running Focal headphones balanced is so overkill. But I wouldn't do it any other way. I love that they use 2-pole 2.5mm connectors for ease of cable replacement.
So I got a highly praised balanced amp. Great. What can I plug into it? A cd player perhaps? Is there such a thing as a cd player with balanced out?? Thanks
Funny thing about a ground loop is that when you want to create it, it becomes impossible to do so! I had a clear case of ground loop with a device that I was talking to the company about. Went to replicate it and it disappeared completely. Spent good bit of time on it but it just would not come back. But yes, I like to create a special setup where it occurs where I can better demonstrate such effects.
Amir, I'm sure that there is very few people that would attempt to measure the voltage on audio gear, but isn't the voltage/current for analog signals AC? Also, what would the AC current be on a speaker cable if the max output was 100 watts into 8 ohms? Folks get so hung up on speaker cables and I may be wrong, but the required cable size might be smaller than most folks think they need. Being in the welding biz I know more copper is better, to a point, but my work can be 100s of amps DC at 80-90% duty cycle.
Am I correct in thinking that in a differential (balanced) headphone output that it would necessarily introduce another inverter/amp device that theoretically could negatively affect the measurements, even if below the threshold of hearing? Of course I mean in contrast to single ended outputs which doesn't need a driver to invert the signal. Thanks for this vid, I never thought about the headphone output not really being balanced.
This is a case that Benchmark Media has made. In practice though, I almost always measure slightly better performance in balanced output than single ended due to distortion cancellation in balanced mode. The difference is only a couple of dBs so definitely in the noise.
Hi! Let me see. Some amps have balanced inputs and balanced output connections but same power on balanced and rca so are not truly balanced, just ports to use balanced cables. Even that, there is an improvement using balanced connection even when internal topology is not balanced?
Thank you for this educational video Amir! Would it be possible, from your chart of headphone sensitivity/efficiency, to tell at what point when going up the graph in inefficiency the headphones will start needing one, then two, the three etc. volts or watts of current to run those headphones at a loud, but listenable level? Thanks again!
Thanks! Could you expand on balanced relating to phono cartridges? My intuition suggests the are balanced by nature, but perhaps more prone to induced noise by the tonearm cables not always being fully symmetrical (at least not the normal RCA connectors). And while you're at it, what about a separate ground connection, and shielding effect from grounding the armtube and any screen on the cables? I've run XLR everywhere except tonearm to phono preamp for decades. Everything is gratifyingly quiet, even the record player (though possible to hear hum when cranked up beyond listening level or with an ear against a speaker). Really appreciate you the few other sensible engineering minded sources of information on audio.
Can you review the ibasso dc04? It is a cheap dac/amp dongle with impressive specs (on paper). I'm wondering if it is good enough to feed a balanced headphone amplifier as the wa11 topaz (I know the dac section of the wa11 is not great). Thanks
What is disturbing to me is why things sound different (maybe better) when using xlr balanced interconnects. (should sound the same) maybe it is because there is more dynamics in the way the receiver take the signal from xlr vs rca. I use balanced every time I will use longer runs and have a bunch of cables that go near network and power. Is it the "blancing" or the "more volts" that make it sound punchier ? Or maybe a stronger signal makes the amplifier run with a lower volume pushing noise floor lower ? I certainly do not hear any noise in rca nor in xlr, just the xlr seem objectively better (same with various devices dac and/or amps)
@@IsoMacintosh that was my first reflex, but I blind tested it to a sax player and he chose xlr, without hesitation. But there again it can be placebo because when I tried to make a good A/B the xlr played one db louder.
11:49 why trying to fix some humming on an old interface from 1974, I noticed that when I tried putting something on the output of a passthrough and noticed some of it came through on my speakers very quietly, also the humming seems to be caused by some microphonics in the cables picking up from the transformer (this thing has such a mess of cables since the "pcbs" were actually just masked on with like a stencil or something so they couldn't fit a lot of traces. The humming remained even when I took the transformer out and off to the side on carpet so it seems the vibrations are sent down the wires...). My guess is back in the day every channel mixed together a bit so maybe there was some truth to having better separation since it got rid of a reference ground that was full of noise? For the life of me I can figure out what causes this inconsistent and sometimes consistent feedback of that humming causing it to get amplified. I dont actually mind it when it is pretty quiet, its kinda nice actually, but it constantly changes volume sometimes (usually gets louder then all of a sudden quiet with a slight popping sound) or goes to a level where its audible while actually playing music (or usually movies/video games) through.
Amir, can you confirm/deny that you wear your watch on your left wrist? When you flip the video (you go from right of screen to left) my 'balance' goes haywire! Ta.
The balanced outputs on my dac go through a tangle of wires behind my media console to reach the inputs of my active speakers, so I am happy to have them
Amir is it possible to review the Spatial Audio Labs open baffle speakers ? The X3, X5 seem very good in subjective reviews and sound clips (they are available at the New Record Day channel). There is not much measurement out there for these speakers, yet but they seem very good. NRD is going to perform some measurements on the X5 soon though.
As noted, I am happy to test it if someone sends one in. You could contact the company as well and see if they send one for testing. Some companies are responsive this way and at any rate, it doesn't take much effort to drop them a note. It is certainly an interesting speaker to measure.
I've read that the best way with balanced is to have both source and amp earthed so the current leakage and voltage potential is equalized through the power cable. The shield in the balanced cable would then create a ground loop so its best to be disconnected at one side(amp). It's also the legal way to do it so not to get shocked if a device would short to chassi. There's been a while since I built my diy dac so please correct any erroneous statements.
As a practical matter, you can never equalize the chassis between two audio devices. Some amount of mains leakage always gets in there causing ground currents to run. In that sense, it is OK if it also flows on the shield of balanced cable. As long as the signal wires are independent of the chassis, you are good. FYI I have a full video on ground loops, etc. including coverage of unbalanced and balanced connections: th-cam.com/video/f2soKg9JAUA/w-d-xo.html
Is it possible to connect the balanced 3 pin xrl male at the rear of D90 to a headamp with 4.4 plug input? Any caution should be taken of the voltage output?
Question for ya: I have always just used the Optical out from my PC to my Emotiva DC-1. To me it sounds like going with USB introduces all kinds of random issues. I like simple/boring. Are there any real world drawbacks going this route? My DC-1 just completely out of the blue started giving me a buzzing noise through the headphone jacks, which I use almost exclusively... Now looking at Topping as a replacement. Emotiva - You had a good run!! I did plug my IEMS directly into my PC and even with a high powered 3080 video card, I have ZERO noise issues, so it's the Emotiva...
So with the Topping A30 Pro (that does balanced in but not balanced out but has XLR+pentaconn) you wouldn't be missing out on a lot? I've been wondering since the review.
That was awesome. I've been switching from thinking I need balanced cables to thinking I don't need them at all. I have a Ifi Zen DAC/Amp which on unbalanced is good for a 300-ohm max headphone. As it so happens, I have Drop Senn 6XX which is 300. Unbalanced, it will go 20 600 ohms. The 6xx get plenty loud without it. I'm not good enough to tell if there is a sound difference even if there was a sound difference. I've also switched from my Dragonfly Black 1.5 to my computer to the Zen and on those, I can tell a difference in the same source. I didn't think I would, but the Zen's output is crystal clear. So much extra behind and within the music.
If the voltage for each cup of a single ended headphone cable oscillates from 0V to +2V instead of -2V to 2V with a "balanced" connection, does that change the way the voice coil or diaphragm of the headphone is moved inside the magnetic field?
No. The equivalent to +2V in a SE connection would be -1V and +1V in a "balanced" config. It's a differential signal. And that differential of 2V between the two wires is all the drivers in your headphone care about. The benefit is mostly the low amount of noise/emi/cross talk between cables. If a signal gets past the cable's shielding it'll affect both wires equally - the difference between -0.7v and +1.3v is effectively still the 2V-signal the amp sent out, so you won't hear any hum or crackling even with bad EMI. A fully "balanced" amp has four channels internally compared to the two inside a single ended counterpart, so on top of twice the max voltage you also get twice the total current capability - assuming identical internals per channel. So you have up to four times the electrical power at your disposal. In the end it's just less bad noise and more power to produce the good noise.
Good point with the headphone output connection. Never got my head around why people seem to think the four-pin output is “balanced” and the TRS is unbalanced. Worth a note is that in “professional grade” interfaces you'll rarely see a four-pin headphone output. There are some “mastering grade” interfaces that do but that's rare too. I think this connection is branded to consumer products and that's where we get the idea of it being “balanced.” But second note here would be that mixing engineers and mastering engineers use headphones as a tool for the really critical listening (finding clicks, pops and such).
You make a good point that pro interfaces have not at all adopted 4-wire connections. Sadly most of the headphone amps they have are anemic with high impedance to boot. So best avoided unless you are stuck some and just want to hear what is coming out.
@@AudioScienceReview Thanks Amir. My experience is that many "pro" interfaces have really good headphone amps, often with plenty of power to run even the toughest cans like a Beyerdynamic 600 ohm or HD800 at 300 ohm. I had a few setups last 20 years or so with plenty of headphones to experience and never really had any issues with quality or power. And I'm really fuzzy about what I listen to : ) I might be wrong but I seems that the 4-pin is by some audiophile enthusiasts considered to be “better” audio quality and therefore to be prefered. I've never used a 4-pin so I wouldn't know but purely out of a technical point the only audible difference should be that the 4-pin would give a bit hotter signal then?
The audible difference between the two topologies for regular wire-length (not for metres and metres) is barely perceivable. The only clear advantage for balanced (BAL) is more power, if you need it. Practically, SE can be as good as BAL depending on the implementation. In the non-professional audio market, BAL units are more expensive, or expensive units use BAL, not because it is so but because it is made so by the marketing dept of the manufacturer when they choose specs for the different models. In theory, there are sonic advatages to BAL but in practical terms and for home use the differences are very-to-extremely small while the cost is usually significantly more for BAL, including accessories (e.g. headphone cables, interconnect cables etc).
I used to be able to hear every time my fridge and my central heating kicked in on my stereo. Audiophiles used to tell me that they put unwanted rubbish on my 240V mains. Since I replace the last single ended signal cable connection with a balanced line there has been total silence (except for the music of course).
Funny story. I bought an amp that had balanced inputs, but my preamp only had RCA out. So I bought an unbalanced to balanced converter (an ARTS model) and, ironically enough, having that powered on did end up introducing a very slight ground loop into the whole system. I just switched back to unbalanced, which meant I had to raise the amp's gain. I found the noise floor (whitish noise, and quiet enough you had to put your ear very close to the speaker to hear it) much less annoying than the hum, which could be heard from a few feet away. I wish balanced connections weren't only put on premium products.
Yeh, anytime you add another supply to the chain, you substantially increase the odds of ground loops. Balanced input provides high immunity but not infinite.
I love the stuff Amir please I really do! I’m just saying as long ago as when I was in high school rock band with long microphone XLR cables running to some main board Peavey PA system in an BBC auditorium that’s when I first heard of the stuff about “balanced” and XLR. OK I don’t know if this would be a good request for a new video (it’s more related to music instrument industry) rather than hi-fi industry but many times with guitar equipment a DI boxes things that are going to go into a long run of cables back to the PA system they call something the ground lift circuit. I don’t if it’s worth it but a ground lift circuit explanation I would find interesting!. 👏🏻
The Beyerdynamic DT990 250 Ohm headphones may not be terribly sensitive but I regularly use my iPhone to drive them and they get plenty loud and sound great.. For 160 dollars they're hard to beat at their price point.
both ZEN CAN and ZEN DAC put much more power on balanced headphone outputs - even with Sundara once can hear the difference. However I am not sure if I heard any difference between connecting these two with RCA vs 25cm 4.4 to 4.4mm
I just want to add that YOU CAN HEAR crosstalk from unbalanced with SENSITIVE headphones. If I mute one channel via the ADI-2 Pro I can hear the other side playing quietly WITH WRONG PHASE. However, this is only true with cables that have 3 signals, e.g. common ground. If the cable has 4 wires and only terminates in 3, the problem seems to be gone. Easily replicable and confirmed by everyone I showed it.
You mean with low impedance earphones. Generaly headphone amplifier with balanced output have much better crosstalk but not always, for exemple the Topping a90 with SE output have crosstalk of -106 and with blanced output have crosstalk of -105 .On other end the Topping a50s with SE output have crosstalk of -86 and with blanced output have crosstalk of -128 . These crosstalk are unloaded and be lower with headphoene attached. There is even headphone amplifier with SE output that have excellent crosstalk that you can not hear anything if you mute one channel with low impedance earphones like the tomtop sonata pro, the crosstalk is -111 with 33 ohm load, which is excellent . Unfortunately with most headphone amplifier with SE output their crosstalk is not good as you can get with headphone amplifie with Balanced output .
@@Or44931 What I am trying to say is that the crosstalk issue is not from the amp but from the cable. If mass of L and R are sharing a path, it creates crosstalk - doesn’t matter which amp. But of course using a balanced amp forces you to use a balanced cable where this issue is gone.
But that is not how you listen to music, right? The much louder channel causes your brain to completely shut off the one that is bleeding. Of course with real music the other channel has its own bits to play. And with speaker, we are constantly experiencing 100% crosstalk since both speakers are playing at once.
@@AudioScienceReview I did not do blind test but when I use amplifier with low crosstalk and amplifier with high crosstalk, they both sound excellent to me. I think that if there is a difference, it is very little. A good blind test is needed to verify this.
I tried to test this (as discussed in an ASR thread) by artificially adding -20 dB of each channel's musical content to the opposite channel and then ABX-ing the two files. Couldn't tell the difference (through a KSC75), so I figured hey, maybe crosstalk isn't as important as NwAvGuy said, and we don't really need -50, -60 dB for it to be completely inaudible. But since then I've done another experiment (or that's what it turned out to be) with better headphones, where I changed the crappy stock cable (3-wire, Y-shaped return with long common part) on my Verum 1 with a proper 4-wire single-ended cable from my HE-400i. There was an undeniable difference in stage width, I couldn't believe I ever enjoyed the Verum with its stock cable and called it "medium concert hall" or "holographic". It was a damned tunnel because of the width constriction because of the crosstalk. And I could tell the difference just by switching to a single-ended cable with no common return wire, just joining the 2 return wires at the common terminal on the TRS plug. Went from about -29 dB cable-crosstalk (based on simple voltage divider math, assuming amp-out is 1 ohm) to about -45 dB cable-crosstalk and I could absolutely hear a *stage width* difference, after testing with several songs from different genres. (Only thin Amir is right about in his comments on crossalk is that most people probably don't need a "balanced-out" on their headphone amp to solve their cable-crosstalk problems: they just need a 4-wire SE cable and it's done.) Not sure how this can be turned into a rigorous/scientific test, but my money is now on the outcome that we will find out NwAvGuy was right and you do need a solid -50 dB to not hear the effects of the crosstalk.
I have two questions. First: why power would go up x4 when voltage goes up x2? P=V x I ,or did I miss something ? So 2x voltage should bring power up by 2x factor. Second question why nobody build "balanced" connection using RCA? It should be technically possible. Just use negative channel of RCA that is NOT connected to device ground. Cables etc would be a heck of a lot cheaper.
Your lectures are always worth listening to. Thanks ever so much.
Thanks Amir! Dude you are such a value in the HiFi world. Even if companies don't like you, consumers do! So happy you are on YT now.
Not all consumer.
*Excellent explanation* of 'balanced inputs, 'differential headphone output' and 'balanced topology'. I always wondered what those 4-pin connectors for headphones really were.
My headphones get loud enough with no noticeable distortion, so I don't need differential headphone output. Thanks for doing this video, Amir. All the best, Rob
Wow, I'm always amazed at how much I learn from each of your videos. Thank you!
Thanks Peter. I am glad you are getting value out of them.
so you did not know about BAL vs SE before --- LOL.
The analogy to the 'care about nuts and bolts in your engine' was not good though. It's common for engines to fail due to bad quality head bolts (especially performance engines) so car people absolutely care about what kinds of studs and bolts are used in engines, not just the performance 😂 The differential output he was talking about is akin to small car amps which operate using push-pull dual amps per channel. That way you can get 14,4 watts (4 ohms) of power from a simple cheap 12V power source.
@@AudioScienceReview Really could not agree more.
The Emotiva Differential Reference amplifiers make balance interconnect balance XLR integrations sing!
when you stopped counting sunny days by amount of sun but the new video from Amir ;-)
That's one of the kindest remarks I have read. :) Had to get my Covid shot and a few other things that kept me from producing videos in the last few days.
@@AudioScienceReview keep yourself safe Amir because you are NEEDED here (and there probably as well). ;-)
I have learned more about audio science from this channel in the past week of binging than I have in 7-8 years of internet "research". Thank you Amir, great info, please keep up the awesome work!
One of the most important parts of internet “research” is the search. That is how you found this channel and Amir!
Thank you ASR for this explanation. Lifted a veil tormenting me for years!!
Cheers!!🍺
Yes, explains balanced/unbalanced characteristics for headphones and equipment that’s easy to understand. Also good to know the amount of benefit in going bal input and output depending on your circumstance.
Thank you! This was wonderfully informative and reaffirmed what I’ve learned from the ASR forums while teaching me/confirming what I’ve heard from my own gear regarding crosstalk and power requirements for my headphones.
Love these videos. I can learn more in 20 minutes than I have in months of viewing random forum posts :x
An Audio Science Easter Egg! Thanks Amir. :)
One of the best audio channels out there, like this is top 5 out of thousands for sure.
Nice one. Addressing that ground loop issue could be an interesting topic for a future lesson...cheers and thank you!
Thanks so much Amir. This has been a source of confusion since I took the deep dive into "audiophile land" 4 years ago.
This explains why I don't hear a difference between my balanced and unbalanced headphone cables. I'm glad my balanced cable is more practical (supple), so not complete waste of money.
unlike me: on balanced output of my amp there is much more voltage for my difficult to drive headphones
The Topping A90, that I also have and adore, takes advantage of this in some way in its outputs since its 4-pin headphone XLR also puts out about double than its SE, TRS output.
Most useful discussion on the topic I have seen. Thanx
The best in the world! He helped me immensely in my hi end low price shopping.
Thanks for taking the time to educate us on the topic.
All of this knowledge that is gushing out of this fountain is awe-inspiring and all of that feel good jazz Amir!
I'll be super keen to also understand how switching to balanced connections contributes to much better separated instruments and open sound. Like the difference is night and day for me, subjectively. This is what I experienced on some fairly decent IEMs initially when did some quick A/B listenings and it was like having an acid trip for the first time kinda big! :D
It changed the way I experience music for good and motivated me enough to switch my home listening to a balanced one too! Again big thanks for the invaluable reviews and measurements of studio monitors and DACs among others. Super enjoying learning something new everyday abour our fav hobby! More power to you!
Surely the snakey businesses won't like you, but you're a super hero to the rare good ones and us small potatoes :D
Great video. Very helpful. I’ve been in the hobby for about 10 years but I learned a fair bit from this
Amir: another great video that illustrates that there are plenty reviewers that don’t know anything. Also please make a video debunking Directional RCA. Thanks
unmeasurable ;-)
@@zyghom I know, but it would make for a cool debunk video
@@IsmaelMartinezPR just a matter of the audience and Amir's time ;-)
That's a great topic.
The best explanation of this topic I have come across
While I don't always agree with your reviews, Amir, I really do appreciate what you are doing here. You're one of the good guys. Keep the good content coming! Cheers!
I've sent this video to all my friends complaining about "too much separation on 'balanced' HP outs" :D
thanks for this! it was on my request list!
Hello Amir! Just a request. Make playlists for every category of your channel (tutorials, reviews, live sessions or whatever etc.). Your channel is growing and in time there will be a lot of videos .
I'd much appreciate if you would create playlists of this kind! Cheers!
Looking forward to your tests debunking balanced channel separation. Especially in a balanced tube amp.
Thanks for your concise succinct information on this often misunderstood subject.
Read a lot of reviews of Amir, nice to see him live in a video! very interesting video
Thanks for the great content !
You are very kind Shawn.
Wow!! Just found this and instantly subscribed. Love your website/trusted honesty
and your expertise is top rank. Thanks for doing this!!!!!!!!
Thank you so much Charles. I am glad you found the information useful.
Curious to know what Amir consideres to be "way up" volume, I'm assuming extremely loud.
Another and educational video. I have 2 Stereo power amplifiers, both from the same manufacturer. One is balanced, the other is ‘single ended’. The balanced amp weighs 73kg. The single ended amp weighs 39kg. The balanced amp cost almost twice that of the single ended. Other than weight and dimensions they share virtually identical specifications and measurements. Whilst I take on board the virtues of a balanced design, is not the caveat that twice the number of electronics are needed. Hence double the weight and price. Surely, the more complex and crowded the signal path is, the more the opportunity to degrade the original. Causing greater deviation from the creators intent. So, whilst the signal may be cleaner, it is not as pure. Not maintaining it’s fidelity!
Thanks for a short "clear" explanation.
Mr. Pete---------->
aging hippie
some turntables have din, "balanced" outputs now like the pro-ject x series tables, and they sell a matching phono preamp that accepts this connection. since this is a 4-pin connection it isn't the same as an XLR balanced connection, any thoughts on these? would this help prevent grounding issues with a turntable? most do not have this though so not really an option
Best video explaining Balanced audio. thanks
This video was so informative. It addressed all my doubts. Great work! and Thank you!
My concern with the growing popularity of XLR/ balanced in/out for marketing reasons is there are a lot of cheap bad balance transformers. A cheap amp will be single ended in and use a cheap transformer to convert the balanced in. I'd take a good unbalanced in to a cheap balanced any day. And the best balanced in would be transformerless.
Hi Amir, I watched this video quite a while ago, but have now developed a problem that I would like to ask a question and possibly get some advice from you.
I have a holo audio May kte DAC, it has XLR or balanced output.
The output voltage on balanced/XLR is 5.8 V rms.
Unbalanced/RCA is 2.9 V rms this is on PCM 48K NOS.
The output voltage when using DSD 128x is
Balanced/XLR 2.9 V rms and unbalanced/RCA 1.45 V rms.
When using an AES/EBU cable from my CD player to the DAC I'm hearing distortion when in PCM mode. When I switch to an unbalanced/RCA connection there is no audible distortion.
And again when using a balance connection in DSD there does not seem to be any distortion that I can hear, making me suspect it's to do with the relatively high output voltage from the DAC in balanced PCM mode. Obviously I would like to use a balanced connection without switching to DSD all the time, but I am at a loss as to know how to do this and avoid the distortion.
Do you have any suggestions?
Many thanks Shane.
I'm not a electrical engineer specializing in audio. I am a professionally trained engine builder. Trust me, when it comes to performance you better care about the nut and bolts. I recommend APR products.
I'm a big fan of your work! Thanks for being an asset to the audio community
Good info on the 4 pin XLR.
Thank you for making such great content. They are really useful.
Sorry if I missed something here.. I thought the 4 pin connector was originally used to go hand in hand with a balanced topology? You brought that up at the beginning of the video, but neglected to discuss it. Correct me if I'm wrong, but a normal stereo amp is basically two mono amps that share a ground. A true balanced amp has four mono amps, two per channel in a push/pull configuration. (Positive and negative side of the circuit.) This is the reason for doubling the power of a given amplifier circuit, but the real purpose is to completely reject any feedback from the ground side. This is supposed to reduce the noise floor to below "silent" and reduce cross talk. Some of the modern designs with "balanced output" may not use a true balanced topology, like you were saying. Some are simply single ended with an internal adaptor and others such as the THX AAA series, use other tricks (feed forward error correction on the case of THX) to achieve the same results. I agree though, it doesn't much matter what happens inside, if you like what comes out. I don't know what happens inside my A90 for example, but I plainly prefer the presentation over my 789, even if the immediate characteristics are the same. The sense of depth and realism are much better, to me.
Great video! Thank you!
Thanks again for another clear and relatively easy to follow presentation. You answered questions I had not thought to ask.:-)
the ground pin in the XLR connector is also connected to power/chassis ground by some means. It's still prone to ground loop.
Learnt so much, thank you Amir.
Excellent explanation, amazing teaching skills.
He Amir. Pardon. Don’t know how to contact you otherwise. What do you think about ceiling treatment? I have one strong reflection coming from there and was thinking about treating it. I will need to treat by back wall since my sofa is right against that wall. I agree with Toole that I don’t want to treat the side walls. One because they are far away in my room and I don’t want to over deaden the room. It’s nice a lively now. And I listen to classical mostly.
Will treating a portion of the ceiling you think deaden the room? A lot of bass resonates from there as well from my measurements.
Thanks as I haven’t found much on ceiling treatments and theory
finally trully pro audio channel, thank you!
Great video. However the DT990 which shows as one of the more "least sensitive" headphones in your graph is an Single Ended headphone, so buying a 'balanced' output amp (more power) doesn't help to counter the sensitivity issue.
Excellent video, and NOW I understand it!
I always felt that the term “push/pull” more accurately described what they called a balanced headphone output. Probably not as marketable though. Thanks.
Great explanations, much appreciated.
Amir, have you had any further consideration on the LA90 amp? I’m thinking of using it on insensitive ML at 4 ohms. Do you think this will be okay or do you suggest something another product?
LA90 doesn't have enough power to drive power hungry speakers. So personally I use high-power class D amps. If you play at more moderate levels though, performance is excellent.
@@AudioScienceReview In your tests, the amps held up at 4 ohm, and I am actually thinking of using them as monoblocks out of spec. Thoughts?
1st if someone needs power out of single ended. FA-10 is the way to go.
2nd from what I understand it is better to not have a ground for audio... generally speaking. I know iFi has some special ground filter.
Problem with the balanced is that some brands will only focus single ended just as a check box, and you may run in to issues if try use multiple inputs. 789 for example, if you connect a DAC and a turntable, whatever source you user RCA for, will get a lower quality. That is why I like to keep everything single ended, no ground connection, filtered usb signal from dedicated streamer to DAC, and everyone is happy.
Hello Amir, thank you for your videos, they are very informative. What are your thoughts on the BACCH software that enhances the spatial characteristics of headphones (and speakers)?
you should do a video on when and why you need a pre-amp
Balanced output usually has 'better' measurements apart from higher voltage; are these not audible improvements? For example channel separation/crosstalk being 70db SE and 100db BAL. Is 70db already too good and beyond that is inaudible?
Thank you. Question (for future videos, perhaps):
What are the risks in plugging in/out analog connections without turning off DACs, Amps, headphone dongles, active speakers?
Running Focal headphones balanced is so overkill. But I wouldn't do it any other way. I love that they use 2-pole 2.5mm connectors for ease of cable replacement.
So I got a highly praised balanced amp. Great. What can I plug into it? A cd player perhaps? Is there such a thing as a cd player with balanced out?? Thanks
That was a great explanation, thank you much. I learned a lot.
maybe you could try and replicate a ground loop, and show how balanced reduces/removes ground loops
Funny thing about a ground loop is that when you want to create it, it becomes impossible to do so! I had a clear case of ground loop with a device that I was talking to the company about. Went to replicate it and it disappeared completely. Spent good bit of time on it but it just would not come back. But yes, I like to create a special setup where it occurs where I can better demonstrate such effects.
Amir, I'm sure that there is very few people that would attempt to measure the voltage on audio gear, but isn't the voltage/current for analog signals AC? Also, what would the AC current be on a speaker cable if the max output was 100 watts into 8 ohms? Folks get so hung up on speaker cables and I may be wrong, but the required cable size might be smaller than most folks think they need. Being in the welding biz I know more copper is better, to a point, but my work can be 100s of amps DC at 80-90% duty cycle.
Am I correct in thinking that in a differential (balanced) headphone output that it would necessarily introduce another inverter/amp device that theoretically could negatively affect the measurements, even if below the threshold of hearing? Of course I mean in contrast to single ended outputs which doesn't need a driver to invert the signal. Thanks for this vid, I never thought about the headphone output not really being balanced.
This is a case that Benchmark Media has made. In practice though, I almost always measure slightly better performance in balanced output than single ended due to distortion cancellation in balanced mode. The difference is only a couple of dBs so definitely in the noise.
Hi! Let me see. Some amps have balanced inputs and balanced output connections but same power on balanced and rca so are not truly balanced, just ports to use balanced cables. Even that, there is an improvement using balanced connection even when internal topology is not balanced?
Thank you for this educational video Amir! Would it be possible, from your chart of headphone sensitivity/efficiency, to tell at what point when going up the graph in inefficiency the headphones will start needing one, then two, the three etc. volts or watts of current to run those headphones at a loud, but listenable level? Thanks again!
Thanks!
Could you expand on balanced relating to phono cartridges? My intuition suggests the are balanced by nature, but perhaps more prone to induced noise by the tonearm cables not always being fully symmetrical (at least not the normal RCA connectors). And while you're at it, what about a separate ground connection, and shielding effect from grounding the armtube and any screen on the cables?
I've run XLR everywhere except tonearm to phono preamp for decades. Everything is gratifyingly quiet, even the record player (though possible to hear hum when cranked up beyond listening level or with an ear against a speaker).
Really appreciate you the few other sensible engineering minded sources of information on audio.
I need to run 3.5mm from my TV to stereo RCA on my powered speakers. How can ensure audio integrity/quality with the adaptor connection?
Can you review the ibasso dc04? It is a cheap dac/amp dongle with impressive specs (on paper). I'm wondering if it is good enough to feed a balanced headphone amplifier as the wa11 topaz (I know the dac section of the wa11 is not great). Thanks
What is disturbing to me is why things sound different (maybe better) when using xlr balanced interconnects. (should sound the same) maybe it is because there is more dynamics in the way the receiver take the signal from xlr vs rca. I use balanced every time I will use longer runs and have a bunch of cables that go near network and power. Is it the "blancing" or the "more volts" that make it sound punchier ? Or maybe a stronger signal makes the amplifier run with a lower volume pushing noise floor lower ? I certainly do not hear any noise in rca nor in xlr, just the xlr seem objectively better (same with various devices dac and/or amps)
probably placebo
@@IsoMacintosh that was my first reflex, but I blind tested it to a sax player and he chose xlr, without hesitation. But there again it can be placebo because when I tried to make a good A/B the xlr played one db louder.
11:49 why trying to fix some humming on an old interface from 1974, I noticed that when I tried putting something on the output of a passthrough and noticed some of it came through on my speakers very quietly, also the humming seems to be caused by some microphonics in the cables picking up from the transformer (this thing has such a mess of cables since the "pcbs" were actually just masked on with like a stencil or something so they couldn't fit a lot of traces. The humming remained even when I took the transformer out and off to the side on carpet so it seems the vibrations are sent down the wires...). My guess is back in the day every channel mixed together a bit so maybe there was some truth to having better separation since it got rid of a reference ground that was full of noise?
For the life of me I can figure out what causes this inconsistent and sometimes consistent feedback of that humming causing it to get amplified. I dont actually mind it when it is pretty quiet, its kinda nice actually, but it constantly changes volume sometimes (usually gets louder then all of a sudden quiet with a slight popping sound) or goes to a level where its audible while actually playing music (or usually movies/video games) through.
Amir, can you confirm/deny that you wear your watch on your left wrist?
When you flip the video (you go from right of screen to left) my 'balance' goes haywire!
Ta.
The balanced outputs on my dac go through a tangle of wires behind my media console to reach the inputs of my active speakers, so I am happy to have them
Amir is it possible to review the Spatial Audio Labs open baffle speakers ? The X3, X5 seem very good in subjective reviews and sound clips (they are available at the New Record Day channel). There is not much measurement out there for these speakers, yet but they seem very good. NRD is going to perform some measurements on the X5 soon though.
I think someone ought to ship equipment to Amir.
On a different topic, anomalous name you've got 😄
@@giriprasadkotte9876 Haha, yeah I know what you mean lol.
As noted, I am happy to test it if someone sends one in. You could contact the company as well and see if they send one for testing. Some companies are responsive this way and at any rate, it doesn't take much effort to drop them a note. It is certainly an interesting speaker to measure.
Thanks, will keep you posted.
Learned a lot here, thanks! :)
I've read that the best way with balanced is to have both source and amp earthed so the current leakage and voltage potential is equalized through the power cable. The shield in the balanced cable would then create a ground loop so its best to be disconnected at one side(amp). It's also the legal way to do it so not to get shocked if a device would short to chassi. There's been a while since I built my diy dac so please correct any erroneous statements.
As a practical matter, you can never equalize the chassis between two audio devices. Some amount of mains leakage always gets in there causing ground currents to run. In that sense, it is OK if it also flows on the shield of balanced cable. As long as the signal wires are independent of the chassis, you are good.
FYI I have a full video on ground loops, etc. including coverage of unbalanced and balanced connections: th-cam.com/video/f2soKg9JAUA/w-d-xo.html
If my integrated amp will accept up to 6volts xlr ...will i overdrive my amp using an XLR input of 5.8volts?
Is it possible to connect the balanced 3 pin xrl male at the rear of D90 to a headamp with 4.4 plug input? Any caution should be taken of the voltage output?
Question for ya: I have always just used the Optical out from my PC to my Emotiva DC-1. To me it sounds like going with USB introduces all kinds of random issues. I like simple/boring. Are there any real world drawbacks going this route? My DC-1 just completely out of the blue started giving me a buzzing noise through the headphone jacks, which I use almost exclusively... Now looking at Topping as a replacement. Emotiva - You had a good run!! I did plug my IEMS directly into my PC and even with a high powered 3080 video card, I have ZERO noise issues, so it's the Emotiva...
So with the Topping A30 Pro (that does balanced in but not balanced out but has XLR+pentaconn) you wouldn't be missing out on a lot? I've been wondering since the review.
Definitely not missing anything. A30 Pro has tons of power so has no need to balanced output to spit out even more.
That was awesome. I've been switching from thinking I need balanced cables to thinking I don't need them at all. I have a Ifi Zen DAC/Amp which on unbalanced is good for a 300-ohm max headphone. As it so happens, I have Drop Senn 6XX which is 300. Unbalanced, it will go 20 600 ohms. The 6xx get plenty loud without it. I'm not good enough to tell if there is a sound difference even if there was a sound difference. I've also switched from my Dragonfly Black 1.5 to my computer to the Zen and on those, I can tell a difference in the same source. I didn't think I would, but the Zen's output is crystal clear. So much extra behind and within the music.
Hi Amir, could you do a vid on AES vs I2S vs USB audio, as usual keep up the excellent content!
So if I have a balanced output on my Schitt freya plus preamp and the rca input of my amp does is help to use a cable that has rca and xlr?
If the voltage for each cup of a single ended headphone cable oscillates from 0V to +2V instead of -2V to 2V with a "balanced" connection, does that change the way the voice coil or diaphragm of the headphone is moved inside the magnetic field?
No. The equivalent to +2V in a SE connection would be -1V and +1V in a "balanced" config. It's a differential signal. And that differential of 2V between the two wires is all the drivers in your headphone care about. The benefit is mostly the low amount of noise/emi/cross talk between cables. If a signal gets past the cable's shielding it'll affect both wires equally - the difference between -0.7v and +1.3v is effectively still the 2V-signal the amp sent out, so you won't hear any hum or crackling even with bad EMI.
A fully "balanced" amp has four channels internally compared to the two inside a single ended counterpart, so on top of twice the max voltage you also get twice the total current capability - assuming identical internals per channel. So you have up to four times the electrical power at your disposal.
In the end it's just less bad noise and more power to produce the good noise.
Good point with the headphone output connection.
Never got my head around why people seem to think the four-pin output is “balanced” and the TRS is unbalanced.
Worth a note is that in “professional grade” interfaces you'll rarely see a four-pin headphone output.
There are some “mastering grade” interfaces that do but that's rare too.
I think this connection is branded to consumer products and that's where we get the idea of it being “balanced.”
But second note here would be that mixing engineers and mastering engineers use headphones as a tool for the really critical listening (finding clicks, pops and such).
@Douglas Blake But Amir just explained in the video that the four-pin is not balanced in that sense or did I miss something about this?
You make a good point that pro interfaces have not at all adopted 4-wire connections. Sadly most of the headphone amps they have are anemic with high impedance to boot. So best avoided unless you are stuck some and just want to hear what is coming out.
@Douglas Blake Thanks Douglas, I take it that Amir's comment was correct then : )
@@AudioScienceReview Thanks Amir. My experience is that many "pro" interfaces have really good headphone amps, often with plenty of power to run even the toughest cans like a Beyerdynamic 600 ohm or HD800 at 300 ohm.
I had a few setups last 20 years or so with plenty of headphones to experience and never really had any issues with quality or power. And I'm really fuzzy about what I listen to : )
I might be wrong but I seems that the 4-pin is by some audiophile enthusiasts considered to be “better” audio quality and therefore to be prefered.
I've never used a 4-pin so I wouldn't know but purely out of a technical point the only audible difference should be that the 4-pin would give a bit hotter signal then?
The audible difference between the two topologies for regular wire-length (not for metres and metres) is barely perceivable. The only clear advantage for balanced (BAL) is more power, if you need it. Practically, SE can be as good as BAL depending on the implementation. In the non-professional audio market, BAL units are more expensive, or expensive units use BAL, not because it is so but because it is made so by the marketing dept of the manufacturer when they choose specs for the different models. In theory, there are sonic advatages to BAL but in practical terms and for home use the differences are very-to-extremely small while the cost is usually significantly more for BAL, including accessories (e.g. headphone cables, interconnect cables etc).
I used to be able to hear every time my fridge and my central heating kicked in on my stereo.
Audiophiles used to tell me that they put unwanted rubbish on my 240V mains.
Since I replace the last single ended signal cable connection with a balanced line there has been total silence (except for the music of course).
Funny story. I bought an amp that had balanced inputs, but my preamp only had RCA out. So I bought an unbalanced to balanced converter (an ARTS model) and, ironically enough, having that powered on did end up introducing a very slight ground loop into the whole system. I just switched back to unbalanced, which meant I had to raise the amp's gain. I found the noise floor (whitish noise, and quiet enough you had to put your ear very close to the speaker to hear it) much less annoying than the hum, which could be heard from a few feet away. I wish balanced connections weren't only put on premium products.
Yeh, anytime you add another supply to the chain, you substantially increase the odds of ground loops. Balanced input provides high immunity but not infinite.
Very nice videos Amir! One small remark, your audio levels are kind of low, the peak levels seldomly reach -10dBFS.
I love the stuff Amir please I really do! I’m just saying as long ago as when I was in high school rock band with long microphone XLR cables running to some main board Peavey PA system in an BBC auditorium that’s when I first heard of the stuff about “balanced” and XLR. OK I don’t know if this would be a good request for a new video (it’s more related to music instrument industry) rather than hi-fi industry but many times with guitar equipment a DI boxes things that are going to go into a long run of cables back to the PA system they call something the ground lift circuit. I don’t if it’s worth it but a ground lift circuit explanation I would find interesting!. 👏🏻
The Beyerdynamic DT990 250 Ohm headphones may not be terribly sensitive but I regularly use my iPhone to drive them and they get plenty loud and sound great.. For 160 dollars they're hard to beat at their price point.
Galvanic isolation is still needed on the USB input to avoid electrical noise from the computer.
Wouldn't just doubling the output voltage of the amplifier (unbalanced) be a lot easier than implementing the balanced headphone drive?
I don't know that it is easier but yes, that is another approach that is indeed done.
Great video. Now I don't feel so bad about spending a little extra to get a 4.4mm to 4.4mm cable for my Zen stack.
both ZEN CAN and ZEN DAC put much more power on balanced headphone outputs - even with Sundara once can hear the difference. However I am not sure if I heard any difference between connecting these two with RCA vs 25cm 4.4 to 4.4mm
It would be nice to have also current measurements, low impedance headphones do care about current more than watts.
I just want to add that YOU CAN HEAR crosstalk from unbalanced with SENSITIVE headphones.
If I mute one channel via the ADI-2 Pro I can hear the other side playing quietly WITH WRONG PHASE.
However, this is only true with cables that have 3 signals, e.g. common ground. If the cable has 4 wires and only terminates in 3, the problem seems to be gone.
Easily replicable and confirmed by everyone I showed it.
You mean with low impedance earphones. Generaly headphone amplifier with balanced output have much better crosstalk but not always, for exemple the Topping a90 with SE output have crosstalk of -106 and with blanced output have crosstalk of -105 .On other end the Topping a50s with SE output have crosstalk of -86 and with blanced output have crosstalk of -128 . These crosstalk are unloaded and be lower with headphoene attached. There is even headphone amplifier with SE output that have excellent crosstalk that you can not hear anything if you mute one channel with low impedance earphones like the tomtop sonata pro, the crosstalk is -111 with 33 ohm load, which is excellent . Unfortunately with most headphone amplifier with SE output their crosstalk is not good as you can get with headphone amplifie with Balanced output .
@@Or44931 What I am trying to say is that the crosstalk issue is not from the amp but from the cable. If mass of L and R are sharing a path, it creates crosstalk - doesn’t matter which amp. But of course using a balanced amp forces you to use a balanced cable where this issue is gone.
But that is not how you listen to music, right? The much louder channel causes your brain to completely shut off the one that is bleeding. Of course with real music the other channel has its own bits to play. And with speaker, we are constantly experiencing 100% crosstalk since both speakers are playing at once.
@@AudioScienceReview I did not do blind test but when I use amplifier with low crosstalk and amplifier with high crosstalk, they both sound excellent to me. I think that if there is a difference, it is very little. A good blind test is needed to verify this.
I tried to test this (as discussed in an ASR thread) by artificially adding -20 dB of each channel's musical content to the opposite channel and then ABX-ing the two files. Couldn't tell the difference (through a KSC75), so I figured hey, maybe crosstalk isn't as important as NwAvGuy said, and we don't really need -50, -60 dB for it to be completely inaudible.
But since then I've done another experiment (or that's what it turned out to be) with better headphones, where I changed the crappy stock cable (3-wire, Y-shaped return with long common part) on my Verum 1 with a proper 4-wire single-ended cable from my HE-400i. There was an undeniable difference in stage width, I couldn't believe I ever enjoyed the Verum with its stock cable and called it "medium concert hall" or "holographic". It was a damned tunnel because of the width constriction because of the crosstalk. And I could tell the difference just by switching to a single-ended cable with no common return wire, just joining the 2 return wires at the common terminal on the TRS plug. Went from about -29 dB cable-crosstalk (based on simple voltage divider math, assuming amp-out is 1 ohm) to about -45 dB cable-crosstalk and I could absolutely hear a *stage width* difference, after testing with several songs from different genres. (Only thin Amir is right about in his comments on crossalk is that most people probably don't need a "balanced-out" on their headphone amp to solve their cable-crosstalk problems: they just need a 4-wire SE cable and it's done.)
Not sure how this can be turned into a rigorous/scientific test, but my money is now on the outcome that we will find out NwAvGuy was right and you do need a solid -50 dB to not hear the effects of the crosstalk.
I have two questions. First: why power would go up x4 when voltage goes up x2? P=V x I ,or did I miss something ? So 2x voltage should bring power up by 2x factor. Second question why nobody build "balanced" connection using RCA? It should be technically possible. Just use negative channel of RCA that is NOT connected to device ground. Cables etc would be a heck of a lot cheaper.
Because I also goes up when V goes up (and R remains constant). That's why you can see it better in P = V*I = V*(V/R) = V^2 / R.