This is a cheap, and relatively simple, to try “tweak” that does not involve modifying your existing components or headphones. It does require an audio player with a DSP/plug-in capability, and/OR balanced DAC/amp/headphone setup and a very easy to make adapter cable*. Before saying anything else, I should say that this idea/practice is something that @johnjen shared with me at a local mini-meet back in January. He’s been playing with this sort of thing for a lot longer than I and any credit really should go to him! I’m writing this now for several reasons: In part, because I was talking to a colleague about this earlier today, they asked me to make them the necessary cable so they could give it a try, and in doing so I figured I might as well write the thing up, including the instructions for making the cable. Another part is down to me mentioning building another phase-reversal cable in the DIY forum last week and only just having gotten around to it. But mostly it’s because my fiancé has taken her mother to visit some friends for a few hours, and I’m all alone in the house with a nice hot soldering iron, a freshly-topped-off glass of Scotch, my music and bugger all else to do for a bit. (*I’ll describe how to make/where to get the necessary cable in the following post. It’s not a bad first-cable-project as the connectors involved are big and easy to work with and there’s lots of leeway for those whose tendency is to solder their thumbs, rather than the components, to still get a working result!). Anyway … what’s this all about? In short, the idea is to more efficiently and completely utilize the full power capacity of an amplifier, and the available rail-to-rail potential difference/swing, by driving left and right channels in opposite phase/polarity within the DAC/amplifier, and then correcting phase/polarity before getting to the headphone (we do it in the headphone cable, with a simple adapter). In theory, driving the channels in offset like this will cause less sag against each rail and draw current from both of them independently, rather than in a common mode. At headphone power levels, the relative power demand differences are not going to be huge, but then we’re not dealing with very much available power to begin with. How do we do this? We invert the polarity/phase of one channel as it goes to either the DAC or the amplifier, and then re-invert that channel (putting it back to its original polarity/phase) as it goes to the headphone. Yes, yes … but how do we do THAT?! There are two straightforward ways to do this, both of which require a balanced headphone connection/cable: If you have a fully balanced DAC/amp/headphone setup, you can use two “polarity/phase reversal cables”, one between the DAC and amp, and the second between the amp and your headphones. Alternatively, you can invert the polarity/phase of one channel in your music player software** and then use a single “polarity/phase reversal cable” between your amp and your headphone. In theory, the second approach gives the same benefits to both the DAC’s output stage and the amp, whereas the first only benefits the amplifier. (**Options and configuration settings for various software/platforms will be discussed in this follow-up post, including entirely free ways to give this a try). But what is polarity/phase reversal? In the simplest terms, it involves swapping the positive and negative connections/signals to the amplifier or DAC. Doing that results in what would have been a positive voltage differential at the output becoming a negative one (and vice-versa). Quite a few DACs allow you to swap absolute phase for both channels with a simple button push. Yggdrasil has this ability for example, though it exists to allow for accommodating recordings mixed out of standard phase and isn’t the same thing as we are talking about here. We only want to invert the polarity/phase on ONE channel (and then switch it back again later in the chain). We can do this in software, which inverts the polarity or phase of ONE channel before the signal reaches the DAC (and this is the preferred way to do it if possible). Players like JRiver Media Center have DSP functions that can do this, and there are plug-ins for players like Foobar2000 and Audirvana+ that can do the same thing. However, if we stopped there, the left and right sides of our headphones would play out of phase with each other. If you did that with a speaker system, you’d wind up with some cancellation effects, most notable in the bass region. With headphones it’ll just do odd things to the sound and imaging. So … what is a “polarity/phase reversal cable”? It’s just a balanced cable that swaps the positive and negative connections on ONE of the two channels it carries. If you were going to wire that, something I’ll describe in detail in a follow-up post, assuming 4-pin DIN connections, you’d wire pins from source to destination as: 1-1, 2-2, 3-4 and 4-3. That would have the effect of inverting the polarity/phase of the right-hand channel. You could do it for the left channel instead, but @johnjen suggested “right is reversed” as a simple way to remember which channel had to be inverted in the player, so I stuck with it. The Net Result? We feed the amplifier (and DAC) an inverted signal, and convert it back to the correct polarity before reaching the headphone driver, putting less stress/or getting better utilization out of the available amplifier power/rail-potential-difference. There should be no effect to the phase of the rendered signal at the headphone. Okay … but how will this change what I HEAR?! This is very hard to say. You might not hear any difference. Indeed, there might not BE any difference in a particular setup. At the power-levels involved in headphone listening the effects are smaller than in speaker systems‡. It is likely that any specific differences will vary significantly from system to system. I would imagine driving IEMs there’d be no discernible effect at all. With less efficient systems or those without gobs of excess power, things might be very different. If you’re expecting bold claims as to what I personally hear, or what you should expect, you’re about to be disappointed, this is, for now, a “see what for yourself” tweak; I’m not even sure how much of what I do/don’t hear is down to expectation bias as I have not spent any serious time attempting to discern a difference in blind listening, though I suspect that’s only part of it. The point is that the tweak has a testable theory behind it (something one could easily setup a measurement case for as well – I might do that myself if I get the time) is easy to experiment with, is cheap and entirely non-destructive/non-invasive. (‡Very easy to do in a speaker system; no special cable is required – just swap the +/- connections on ONE speaker. BUT, make sure you swap the polarity/phase on the same channel either in your software player or before the amplifier input, otherwise your speakers will be out of phase and that’ll kill your bass and generally sound entirely wonky!) Finally … I'll post what my experiences have been with this once some others have a chance to give it a try. Also, expect the usual flurry of edits and corrections resulting from me only seeing the myriad mistakes/typos I have no doubt made in my fermented-grain-induced euphoric haste to type and post this.