I am working on a new build and I have a little trouble deciding what to do.
I usually build regular 2-channel builds with a good old fashioned ground and have had great success doing so.
However I have also been looking into a 2 channel dual mono build so I could have separate ground connections for the left and right transducers.
But what I am really considering is doing a "balanced" build this time. Sure it wont be really balanced in the traditional sense, more like a bridged amp for each channel.
In best case I could get 40dB common mode noise rejection if I use 1% gain resistors for each channel and up to 60dB best case with 0.1% resistors. Basically the "balanced"/bridged channels have to be as closely matched as possible to get any worthwhile common mode noise rejection. But then again, if there is not much noise to begin with getting more common mode noise rejection is sort of useless. The other issue is twice the output impedance in "balanced"/bridged builds.
Any other potential issues with "balanced"/bridged builds?
Thoughts?
I usually build regular 2-channel builds with a good old fashioned ground and have had great success doing so.
However I have also been looking into a 2 channel dual mono build so I could have separate ground connections for the left and right transducers.
But what I am really considering is doing a "balanced" build this time. Sure it wont be really balanced in the traditional sense, more like a bridged amp for each channel.
In best case I could get 40dB common mode noise rejection if I use 1% gain resistors for each channel and up to 60dB best case with 0.1% resistors. Basically the "balanced"/bridged channels have to be as closely matched as possible to get any worthwhile common mode noise rejection. But then again, if there is not much noise to begin with getting more common mode noise rejection is sort of useless. The other issue is twice the output impedance in "balanced"/bridged builds.
Any other potential issues with "balanced"/bridged builds?
Thoughts?
Last edited:
1st question is an amp for which model headphone - they vary by orders of magnitude in Z, sensitivity, required I,V
"doubling V swing" is pretty pointless driving a sensitive IEM that reaches 120 dB SPL @ 200 mVrms - and look really silly with dual XLR "balanced" connectors at the end of the cable...
"doubling V swing" is pretty pointless driving a sensitive IEM that reaches 120 dB SPL @ 200 mVrms - and look really silly with dual XLR "balanced" connectors at the end of the cable...
I work with electronics for a living(I'm an electroncis technician) so I am not really after any technical help.
It is a Class A headphone amp of my own design.
Intended to drive my 96dB/1mW, 250 Ohm Beyerdynamic DT880 Pro headphones.
I am not after more voltage swing as such, what I am interested in is knowing the potential for improvements in sound quality by going balanced or the potential negative issues by doing so.
I have never tried "balanced"/bridged headphone amplifiers before and as such I really do not know if there is any real world benefit to it compared to SE.
From what I could find about the issue so far, it seems that balanced headphone amplifiers is more about marketing than any real world benefit.
Benefits :
Higher voltage swing, common mode noise rejection(how much depends on match between + and - amps), the abilty to connect balanced DACs directly.
Cons :
Higher output impedance, Higher distortion(I read that somewhere), Expensive.
That is the technical Benefits/cons of "balanced"/bridged headphone amplifiers as I see it, but what about the sound?
Last edited:
You won't get any common-mode rejection with bridged amps. Need a differential input for that. Bridged channels will just amplify and pass along any common-mode noise and you'd have to rely on your headphones for any common-mode rejection.
In any case, there's no earthly reason for a "balanced" headphone amp to sound any better than a single-ended headphone amp.
se
In any case, there's no earthly reason for a "balanced" headphone amp to sound any better than a single-ended headphone amp.
se
Actually, I know I won't get any common mode rejection from the bridged amps as such, but it should increase the common mode rejection across the transducers. Should have wrote what I meant a little more clearly.
The reasoning for this is that you should have the same noise on the + and - signal in a bridged amp but out of phase and that should give you better common mode rejction at the transducers. A standard SE build would also have common mode rejection across the transducers, but it does not work that well since you instead of 1xsignal+noise and 1xsignal+noise out of phase only have 1 x signal+noise and ground.
So while a "balanced"/bridged amp will not have any inherent common mode noise rejection it should improve the common mode noise rejection at the transducers, since one of the signals+noise is out of phase compared to the other signal+noise. In the end it comes down to how closely the amp are matched as well as how close the input signal to each amp is matched.
As for the "balanced"/bridged builds not sounding any better, well they might sound different at the most but if they sound better, I really do not know.
I am most likely going with a SE build as I have done previously, it just works and sounds great as well.
Last edited:
What's the point of getting better common-mode rejection at the transducers if you're just going to amplify any common-mode noise that's at the input of the amp? Why not reject any common-mode noise at the input and get rid of it there instead of relying on the transducers to do it after it's been amplified?
se
se
Well, if you have a simple SE build, any noise the input cabling picks up will get amplified and end up in your headphones as, well, noise.
On a balanced build any noise picked up by the input cabling should be identical on both signal wires. That noise would be amplified evenly by your "balanced"/bridged amp and then rejected at the transducers. This all depends on the matching of each amplifier as well as any potential attenuation at the input being equal on both signals.
Correct me if I am wrong.
But if you do not have any noise issues, SE should be more than good enough.
balanced audio interface between source and amp is a possibly independent issue from “balanced”/bridged headphone amplification/driving
I believe most professional equipment using balanced signal interface converts to single ended internally
many “balanced” headphone amps are in fact just 2 identical, separate amps for the +,- balanced signal inputs – which means any common mode on the input is amplified, eats into the output Vswing
while a “proper” balanced receiver actually removes the common mode V
dynamic headphone's drivers basically don't care about anything but the difference V at their terminals so there is rejection of the (amplified) common mode in their audio output
rewiring headphones for “balanced” cables, bridged drive is somewhat faddish but separating the R,L driver gnds does offer technical improvement in common impedance crosstalk over TRS common gnd
but you don't need the bridged drive to get most of the advantage of the separated R,L driver connectors – just attention to the amp's internal gnding fine details
bridged drive can give even harmonic distortion cancellation – where it is reproducible enough between the +,- amplifier devices
since even harmonics are often considered less audibly objectional the benefit is questionable
and a proper balanced receiver input, driving a single ended amp can be technically way better than estimates of audibilty
the technical differences with a “balanced”/bridged drive amp can easily be in implementation details rather than fundamental topology advantages
I believe most professional equipment using balanced signal interface converts to single ended internally
many “balanced” headphone amps are in fact just 2 identical, separate amps for the +,- balanced signal inputs – which means any common mode on the input is amplified, eats into the output Vswing
while a “proper” balanced receiver actually removes the common mode V
dynamic headphone's drivers basically don't care about anything but the difference V at their terminals so there is rejection of the (amplified) common mode in their audio output
rewiring headphones for “balanced” cables, bridged drive is somewhat faddish but separating the R,L driver gnds does offer technical improvement in common impedance crosstalk over TRS common gnd
but you don't need the bridged drive to get most of the advantage of the separated R,L driver connectors – just attention to the amp's internal gnding fine details
bridged drive can give even harmonic distortion cancellation – where it is reproducible enough between the +,- amplifier devices
since even harmonics are often considered less audibly objectional the benefit is questionable
and a proper balanced receiver input, driving a single ended amp can be technically way better than estimates of audibilty
the technical differences with a “balanced”/bridged drive amp can easily be in implementation details rather than fundamental topology advantages
Last edited:
That depends on your design. You could, for instance, add a good quality 1:1 transformer at the input. That would offer you outstanding common-mode rejection even if you feed it from an unbalanced source (and it can just as easily accept a balanced source). In addition, you get ground isolation which can help reduce noise due to ground loops.
se
That is an option.
But when you do not have issues with noise(I don't), well, then there is nothing to fix.
But it is a simple solution for those that do have noise issues.
That is an option.
But when you do not have issues with noise(I don't), well, then there is nothing to fix.
But it is a simple solution for those that do have noise issues.
Even without any problems to "fix," I've always been fond of using an input transformer on my stuff.
se
That would help if you had a limited supply voltage to deal with like a battery supply, but with an AC supply it's easy enough to choose whatever voltage you need.
se
Hi Neutrality,
I find your thinking convincing, probably because it mirrors my own. I started building dual mono (discrete then LME49600 buffered), feeding them a balanced signal, and feeding to each side of modified cabling across HD600 cans. This result bettered the Creek OBH-11 I had been using. I tweaked the power supplies but the big leap came when I built 'the Wire' BAL-BAL.
I wish I could have skipped a few of the intermediate steps and gone straight to 'the wire'. If the HD600's were less of a hog the improvements might have been less apparent, dunno. I heard opc plans a rerun. recommended.
I find your thinking convincing, probably because it mirrors my own. I started building dual mono (discrete then LME49600 buffered), feeding them a balanced signal, and feeding to each side of modified cabling across HD600 cans. This result bettered the Creek OBH-11 I had been using. I tweaked the power supplies but the big leap came when I built 'the Wire' BAL-BAL.
I wish I could have skipped a few of the intermediate steps and gone straight to 'the wire'. If the HD600's were less of a hog the improvements might have been less apparent, dunno. I heard opc plans a rerun. recommended.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.