Fixing inverted polarity of a multi-driver IEM.

[Jean-Paul Petrosyan]

Hi!

Recently I have purchased a Truthear Nova IEM and am currently waiting for it to arrive. Looking at these measurements (https://reference-audio-analyzer.pro/report/hp/truthear-nova.php#gsc.tab=0), it seems like it has inverted polarity. If I try to fix this by connecting the cable differently, purchasing a cable that has a different polarity (no way to know about this beforehand and after, without a multimeter at least, which is the main reason why I'm asking about this, as I would very much like to change the stock cable for another one since I find it ugly and hate the paracord sheathing), or reversing the signal polarity at the source side; will this change the frequency response of the IEM? Does it depend on whether all drivers have the same polarity and is there a way to know if that's the case or not?

Thank you very much for your time and attention!

 

[Pasarski]

What do you mean by "it has inverted polarity" and how do you come to that conclusion looking at those measurements?

The polarity must be same for both ears, other than that, it makes no difference.

Edit: I realized you must be looking at the impulse response? It sure seems to be inverted. But as long as both ears have the same polarity (which seems to be the case) it doesn't matter a tiniest bit, if you are just listening to music. If you are using the IEM's for monitoring on the stage, it might matter a bit.

 

[DVDdoug]

You'd have to make an inverting cable yourself. And it can only be done if each ear has it's own cable... You need to swap two pairs of wires... You can't invert with a common ground. If they have a balanced connection you've got 4 wires and it can be done/

But there's really no problem. You can't hear it and there's no guarantee that the recording (or some of the tracks in the recording) weren't inverted or that something in your playback system isn't inverting.

Some electronic circuits are easier to build if they invert and there are no "rules". An amplifier or preamp isn't considered defective if it inverts.

If you invert one speaker of a pair, the out-of-phase soundwaves in the air interfere and the bass gets almost completely canceled. If you invert them both, everything is cool again.

BTW - The tweeter in a 2-way speaker or the midrange in a 3-way is often inverted. That's because the crossover creates phase shifts and flipping the polarity of one driver gets them back in-phase at the crossover frequency where both drivers are operating and the soundwaves can interfere and cancel. That could be the same situation with your multi-driver IEMs. The polarity might be inverted on one of the drivers.

no way to know about this beforehand and after, without a multimeter

A multimeter will allow you to check the polarity-wiring of the cable (if you can get-to both ends) but it can't tell you the polarity of the connections to the drivers.

 

[Jean-Paul Petrosyan]

What do you mean by "it has inverted polarity" and how do you come to that conclusion looking at those measurements?

The polarity must be same for both ears, other than that, it makes no difference.

Edit: I realized you must be looking at the impulse response? It sure seems to be inverted. But as long as both ears have the same polarity (which seems to be the case) it doesn't matter a tiniest bit, if you are just listening to music. If you are using the IEM's for monitoring on the stage, it might matter a bit.

I intend to use it for both music listening and monitoring. I've read that phase inversion can be an issue with voice self monitoring due to bone conduction at the very least.

You'd have to make an inverting cable yourself. And it can only be done if each ear has it's own cable... You need to swap two pairs of wires... You can't invert with a common ground. If they have a balanced connection you've got 4 wires and it can be done/

But there's really no problem. You can't hear it and there's no guarantee that the recording (or some of the tracks in the recording) weren't inverted or that something in your playback system isn't inverting.

Some electronic circuits are easier to build if they invert and there are no "rules". An amplifier or preamp isn't considered defective if it inverts.

If you invert one speaker of a pair, the out-of-phase soundwaves in the air interfere and the bass gets almost completely canceled. If you invert them both, everything is cool again.

BTW - The tweeter in a 2-way speaker or the midrange in a 3-way is often inverted. That's because the crossover creates phase shifts and flipping the polarity of one driver gets them back in-phase at the crossover frequency where both drivers are operating and the soundwaves can interfere and cancel. That could be the same situation with your multi-driver IEMs. The polarity might be inverted on one of the drivers.


A multimeter will allow you to check the polarity-wiring of the cable (if you can get-to both ends) but it can't tell you the polarity of the connections to the drivers.

So, if the polarity of one or a few drivers in an IEM is inverted relative to others... What will happen if I invert the polarity of the signal or plug the cable in a "wrong" way? Could that have an impact on the frequency response?

Regarding the last bit, yes, it will only be possible to know the cable's polarity. But given that we know the polarity of the IEM with it's stock cable, learning the polarity of that stock cable should be enough to predict what polarity the IEM will have with any other cable.

I've read that polarity wiring might vary between cables.

 

[NTK]

I intend to use it for both music listening and monitoring. I've read that phase inversion can be an issue with voice self monitoring due to bone conduction at the very least.



So, if the polarity of one or a few drivers in an IEM is inverted relative to others... What will happen if I invert the polarity of the signal or plug the cable in a "wrong" way? Could that have an impact on the frequency response?

Regarding the last bit, yes, it will only be possible to know the cable's polarity. But given that we know the polarity of the IEM with it's stock cable, learning the polarity of that stock cable should be enough to predict what polarity the IEM will have with any other cable.

I've read that polarity wiring might vary between cables.

It is easy enough to test. You can use Audacity and apply "Invert" (under menu "Effect" -> "Special") to invert polarity and listen to the outcome yourself.

 

[Pasarski]

There's some cancellation happening when monitoring one's own voice in inverted polarity. But it's not a huge effect. Easy to test following above suggestion, or talking/singing into a opposite sides of a figure 8 mic (the opposite sides are in opposite polarity)

 

[Jean-Paul Petrosyan]

Thank you, everyone! I'm pretty inexperienced, so it might take a lot of time to get a grasp with experimentation. Is there a way I could relatively easily simulate the phase cancellation "honeycomb filtering" effect to learn how it sounds in order to reliably determine if it's present when self-monitoring (and discern it from the native frequency response of the IEM)?

And apart from that, should I expect the overall polarity change to impact the IEM's frequency response due to a possibly different polarity of the individual drivers, after all? Does that kind of thing happen with speakers? And to what extent the system's behavior should deviate from that when it comes to IEMs?

Please, excuse me if being dense here...

 

[Pasarski]

Test it. Listen with headphones and invert the polarity of the other channel. Sounds weird. If both channels have the same polarity the absolute polarity doesn't matter (when only listening). The cancellation only happens when two identical signals with different polarity (or phase) are combined.

 

[DVDdoug]

Is there a way I could relatively easily simulate the phase cancellation "honeycomb filtering" effect to learn how it sounds in order to reliably determine if it's present when self-monitoring (and discern it from the native frequency response of the IEM)?

You could do a "clean inversion" easily in Audacity but simulating the interaction between bone conduction would be super complicated and you'd have to know the characteristics. But for monitoring while recording it doesn't have to be "perfect" and it never will be because the playback of your recording/performance doesn't involve bone conduction or anything like that.

It's actually comb filtering (without the honey). It sounds pretty bad when done "properly". Well... sometimes it's done on purpose a special effect or a "flanging" effect is a moving-shifting comb filter so maybe it can sound good...

Does that kind of thing happen with speakers?

Worse!!! The wavelength of a 1kHz soundwave is about 1 foot. So at 2kHz the sound hitting your left ear can be 180 degrees out-of-phase with the waves hitting your right ear. Plus there are reflected sounds with all kinds of complex delays and phase relationships.

 

[Jean-Paul Petrosyan]

You could do a "clean inversion" easily in Audacity but simulating the interaction between bone conduction would be super complicated and you'd have to know the characteristics. But for monitoring while recording it doesn't have to be "perfect" and it never will be because the playback of your recording/performance doesn't involve bone conduction or anything like that.

It's actually comb filtering (without the honey). It sounds pretty bad when done "properly". Well... sometimes it's done on purpose a special effect or a "flanging" effect is a moving-shifting comb filter so maybe it can sound good...


Worse!!! The wavelength of a 1kHz soundwave is about 1 foot. So at 2kHz the sound hitting your left ear can be 180 degrees out-of-phase with the waves hitting your right ear. Plus there are reflected sounds with all kinds of complex delays and phase relationships.

I see, thank you!

Yeah, I don't know where the "honey" part came from . I guess my ADHD mind had drifted elsewhere.

These complex acoustic phenomena aren't something inherently bad, though, in context of listening to speakers at least. Right? It's natural for us to perceive sound with all these effects taking place, moreso even than without them... To my understanding.

I played a bit with Phase Inversion option in Neutron MP and I definitely perceive a pretty recognizable and consistent difference. Even tried to do a little blind test with the help of my friend, if not the most scientifically rigorous one for sure...

Could that point to the drivers in Nova having different polarity, maybe? Though Truthear specifically boasts about consistent phase across all the drivers/frequency range.

 

[kemmler3D]

should I expect the overall polarity change to impact the IEM's frequency response due to a possibly different polarity of the individual drivers, after all?

No, the individual drivers' polarity is determined by the crossover / IEM itself, so if you change the polarity upstream of that (using the cable) then it won't affect FR. Basically you're just switching whether the driver goes in or out when the signal coming in goes up or down. But that's the only real change.

For monitoring your own vocals, I can see why this is a concern, but typically it's a non-issue as others have mentioned.

 

[Jean-Paul Petrosyan]

No, the individual drivers' polarity is determined by the crossover / IEM itself, so if you change the polarity upstream of that (using the cable) then it won't affect FR. Basically you're just switching whether the driver goes in or out when the signal coming in goes up or down. But that's the only real change.

For monitoring your own vocals, I can see why this is a concern, but typically it's a non-issue as others have mentioned.

Thank you for the explanation!

After listening to a few test samples showcasing the difference in absolute polarity, which is pretty apparent to me, on my phone and then on a TV, I believe that Nova has correct polarity. My Xenns Tea Pro, however, seems to have inverted polarity. That is assuming that my iBasso DX180's polarity is not inverted... (what a can of worms to discover). As well as Moondrop Variations that I previously owned. The difference can be distinguished from the frequency response with experience, but I had my IEMs equalized to the same target curve for testing.

Polarity difference doesn't seem like something that will make otherwise good set bad and vice versa, but going by my subjective impressions it shouldn't be discarded as completely irrelevant and unimportant.

I guess I'll have to go learn how to determine the polarity of amplifier outputs now.

I watched a video from Audio University channel on the difference b/w polarity and phase and it is stated that polarity and phase inversion are not effectively equivalent as the latter may not affect all frequencies tye same, while the former always will. Is that so?

This means that there's no simple and convenient way to compensate for reversed polarity via playback software features and I feel bothered by this, I should stick to IEMs with 2-pin connection and seek out cables without the heat shrink earhooks so that they could be plugged in reverse.

It blows to have OCD...

 

[Lion♡]

I’m not sure whether nova simply isn’t included in the measurements(from BRIR Gallery) taken by other Korean users or if I just can’t find it, but compared to Hexa, it looks like this.

 

[Soandso]

We should consider that manufactured polarity switching in a multi-driver iem might be deliberately selective. Polarity switching during multiple music track mixing is used to drop certain peak sound crests. If we consider how different iem drivers are targeted frequency related then those individual drivers are in a sense carrying different musical tracks.

In iems certain selective driver's with inverted polarity would allow the the interplay of the multiple drivers' composite amplitude (peak) to include altered signal input in the mixed music heard. This would compensate for our natural human hearing of different frequencies at different decibel levels. The benefit would be doing away with any need to reactively resort to turning up the volume, thus avoid introducing device artifacts and/or clipping.

As for our detection of loud speaker polarity inversion: it is not humanly noticeable (according to B.Katz in 2007's "Mastering Audio - the Art and the Science"). What may be detected is revealed by the speaker itself having non-linear properties and/or the design limitations of driver magnets.

As others pointed out polarity should not be conflated with phase. Which has more to do with sound aligning in terms of time and thus some potential of sounds bleeding into one another.

 

[RogerSmith]

We should consider that manufactured polarity switching in a multi-driver iem might be deliberately selective. Polarity switching during multiple music track mixing is used to drop certain peak sound crests. If we consider how different iem drivers are targeted frequency related then those individual drivers are in a sense carrying different musical tracks.

In iems certain selective driver's with inverted polarity would allow the the interplay of the multiple drivers' composite amplitude (peak) to include altered signal input in the mixed music heard. This would compensate for our natural human hearing of different frequencies at different decibel levels. The benefit would be doing away with any need to reactively resort to turning up the volume, thus avoid introducing device artifacts and/or clipping.

As for our detection of loud speaker polarity inversion: it is not humanly noticeable (according to B.Katz in 2007's "Mastering Audio - the Art and the Science"). What may be detected is revealed by the speaker itself having non-linear properties and/or the design limitations of driver magnets.

As others pointed out polarity should not be conflated with phase. Which has more to do with sound aligning in terms of time and thus some potential of sounds bleeding into one another.

Inverted polarity is audible. At least if you are a trained listener.

 

[RogerSmith]

Hi!

Recently I have purchased a Truthear Nova IEM and am currently waiting for it to arrive. Looking at these measurements (https://reference-audio-analyzer.pro/report/hp/truthear-nova.php#gsc.tab=0), it seems like it has inverted polarity. If I try to fix this by connecting the cable differently, purchasing a cable that has a different polarity (no way to know about this beforehand and after, without a multimeter at least, which is the main reason why I'm asking about this, as I would very much like to change the stock cable for another one since I find it ugly and hate the paracord sheathing), or reversing the signal polarity at the source side; will this change the frequency response of the IEM? Does it depend on whether all drivers have the same polarity and is there a way to know if that's the case or not?

Thank you very much for your time and attention!

The Nova has a 2-pin cable. You can just buy another 2-pin cable without these curved ear hooks and plug it in rotated by 180 degrees.

 

[Soandso]

Inverted polarity is audible …

Might what is detected audibly be nuances the below screen shot describes as "synthetic signals" and summarized from B.Katz (#14 above citation) I described as equipment artifacts?

Screen shot text citation =

 

[Jean-Paul Petrosyan]

We should consider that manufactured polarity switching in a multi-driver iem might be deliberately selective. Polarity switching during multiple music track mixing is used to drop certain peak sound crests. If we consider how different iem drivers are targeted frequency related then those individual drivers are in a sense carrying different musical tracks.

In iems certain selective driver's with inverted polarity would allow the the interplay of the multiple drivers' composite amplitude (peak) to include altered signal input in the mixed music heard. This would compensate for our natural human hearing of different frequencies at different decibel levels. The benefit would be doing away with any need to reactively resort to turning up the volume, thus avoid introducing device artifacts and/or clipping.

As for our detection of loud speaker polarity inversion: it is not humanly noticeable (according to B.Katz in 2007's "Mastering Audio - the Art and the Science"). What may be detected is revealed by the speaker itself having non-linear properties and/or the design limitations of driver magnets.

As others pointed out polarity should not be conflated with phase. Which has more to do with sound aligning in terms of time and thus some potential of sounds bleeding into one another.

Thank you for you input! Alas, I don't think I was able to understand it, though. My knowledge is too limited...

I don't know if there's a way to identify what part of the audio chain shows the audible issues... But if I was able to understand at least something from your post, it may bear no correlation to the integrity of the original audio signal itself?

Anyways, here's an interesting thread on this matter with audio samples for listening tests. I found it pretty curious and it seems to confirm with my personal perceptual experience which might make me biased, of course... ‍).

https://www.reddit.com/r/headphones/comments/12gvhcn/why_polarityabsolute_phase_is_audible_with_audio

The Nova has a 2-pin cable. You can just buy another 2-pin cable without these curved ear hooks and plug it in rotated by 180 degrees.

Yep, it seems like I'll have to do just that. Though that criteria limits the pool of potential options quite significantly. I'd like to have a cable with interchangeable connectors and I haven't seen one without the earhooks yet.

Maybe I should try to come up with a way to safely remove the hear shrink that form the ear hooks on any cable... I've tried that before and it has proven to be tricky.

 

[DVDdoug]

Really, if you think this is a problem maybe just get different in-ears. Different in-ears will sound different anyway (a bigger difference than inverting polarity) and you should be able to find something that sounds better, especially since you don't seem to be satisfied with what you have...

You may want to avoid multi-driver designs because it's likely (or at least possible) that at least one of the drivers is inverted, as above with speakers to keep them in-phase at the crossover frequency.

I'd like to have a cable with interchangeable connectors and I haven't seen one without the earhooks yet.

If you know a little electronics and how to solder you could make a little box with a 4PDT switch to optionally invert the polarity. Then you could "correct" any recordings that you think are inverted. (From what I've read it should be about 50/50.) This does require 4 wires (2 separate connections for each driver) so it won't work with the standard TRS headphone connection. But... In some recordings the guitar might be inverted but not the vocal, etc.

 

[RogerSmith]

Thank you for you input! Alas, I don't think I was able to understand it, though. My knowledge is too limited...

I don't know if there's a way to identify what part of the audio chain shows the audible issues... But if I was able to understand at least something from your post, it may bear no correlation to the integrity of the original audio signal itself?

Anyways, here's an interesting thread on this matter with audio samples for listening tests. I found it pretty curious and it seems to confirm with my personal perceptual experience which might make me biased, of course... ‍).

https://www.reddit.com/r/headphones/comments/12gvhcn/why_polarityabsolute_phase_is_audible_with_audio

Yep, it seems like I'll have to do just that. Though that criteria limits the pool of potential options quite significantly. I'd like to have a cable with interchangeable connectors and I haven't seen one without the earhooks yet.

Maybe I should try to come up with a way to safely remove the hear shrink that form the ear hooks on any cable... I've tried that before and it has proven to be tricky.

Honestly, I have the Nova and albeit it might have inverted polarity, I did not hear that, I think the issue is really tiny and only at very special music material it becomes audible and then only if you are really trained. I did not manage to hear it. So I would say you don't need to bother.