Looking for some schooling for some audio engineering knowledgeable 'nauts…
About a month ago I received a Vermona '14 synth. This instrument has an “transformer-balanced XLR output” and audio coming out of this instrument sounds wonderful. I don’t have the technical knowledge to describe it adequately, but I’d calling it “present and full”. At the time I chocked it up to the fully analog nature of this synth. The instrument also has a huge amount of headroom – so much so that I never have the volume knob all the way up or I’d be totally clipping my digital mixer.
A few days ago I got a Radial Key-Largo as a sub-mixer for some of my gear. This mixer/interface is also advertised as having “transformer-isolated and balanced XLR main outputs.” This gave me an opportunity to compare the sound of instruments going directly to my digital mixer vs through the Key-Largo and then into the digital mixer. I was surprised to find that there is a noticeable difference when I run them through the Key-Largo – stuff sounds better; “more present and full” is again how I’d describe the difference.
Is this just an “audio placebo” effect, or is there a real benefit to fancy outputs that makes audio signals sound “better”
There’s also obviously some amount of gain being applied to the signals by the Key-Largo so maybe that’s all that’s at play here?
Going unbalanced to balanced gives you a 6db drop in volume, so will always sound ‘less present’ until you compensate for the loss by adding gain somewhere. You’d be surprised how much effect a small drop (or increase) in volume has on the subjective perception of the ‘quality’ of a sound. 6db drop is actually a halving of perceived power. Transformers themselves vary in quality, and are known to have the largest effect on low end frequencies, causing distortion and phase shift in that area.
Transformers also excel at EFI and RFI noise rejection, which can create a situation where you have more signal and less atmospherically absorbed noise in your chain which can make a difference as well.
There is a minuscule amount of latency involved in the transformer process which can take a bit away from transient response which is why you have seen most microphone manufacturers move away from them on higher end models.
On the flip side, one of the most famous microphones of all time the Neumann U87 has a transformer output and many people swear by the “smokiness” that the transformer imbues to the signal.
Good transformers generally aim to be transparent but they are limited by their size. Smaller transformers are less accurate in the low frequency range (in general).
I may need some schooling here but I was not aware that this is always the case. A 6dB drop is substantial and a transformer, being a passive device, can drop the level. Electronic balancing would not necessarily do this…I generally associate the 6 dB drop with going the OTHER direction (from balanced to unbalanced) because you are only using one leg of the signal.
Think about it, you’re using only half the signal that a balanced connection would have. The cold side is grounded if you plug straight in with a TS jack, whether it’s fed on to a transformer or a servo balancing circuit using op-amps. This means there’s only half as much energy, hence the 6db drop, which is merely a measurement of the effect. And as you say, it counts going the other way too. You’re always using just half of the power a fully balanced connection has.
@homealone Yup, we are having a heated agreement. I read your statement wrong- I thought you were saying that BALANCING a signal would drop it 6 dB… I just need to read more carefully.
Just to revisit this statement, I don’t know a lot about the theory other than what I’ve read, but core saturation can color the sound and (depending on application) that may be appealing, or undesirable.
Sorry. I think I’m confusing myself now as well! Balancing a signal will not drop the level, but it will be 6db lower than using the same input with an equivalent balanced signal.
What I was really trying to convey in the first post was that differences in level, even small ones, can create very distinctive perceptual effects, and in my experience are the cause of a lot of ‘wow this sounds clear and full’ type of reactions. I shouldn’t have tried using technical terms to make it feel like an authoritative answer…
Totally…in my experience even a half dB can make a difference in what sound we prefer. I managed a huge converter head to head shoot out at a major Nashville studio once and nearly everybody picked the 16 bit converters over the expensive 24 bit ones because we had messed up the gain staging and it was just a tiny bit louder.
One golden-ears guy (out of about 30 great engineers) called me out on it and he was absolutely correct. I was pretty mortified.
Also, audio memory is really flawed. Unless you are truly doing switched AB type comparisons it is unlikely you are getting an accurate read on “better” or not.
That is exactly how I discovered this effect as well. Did a very unscientific converter shoot-out for myself, and found that I couldn’t notice a difference unless the levels were mismatched even by a tiny amount. When they were almost exactly level it was impossible to say what converter I was hearing. Made me give up chasing ‘the best’ and instead look for companies who offer great long-term customer service as a first requirement.
What you’re hearing has not much to do with the fact that the output is balanced, but rather that it passes through a transformer driven by a discrete line amplifier. And indeed it colour the sound, most often in a pleasant way.
