Dan Clark Expanse Headphone Review

[oscar_dziki]

I see many reviewers complaining about the lack of dynamics. Do they really lack punch and dynamics or do people just underestimate the power requirements for these? I would only use low impedance headphones like those with a speaker amp to make sure they get enough current for their maximum potential

Dynamics - no. Punch -maybe a little. I like to check the punch and dynamics with Artificial Red by Alice in chains. Dynamics is flawless, for percussion on this track I need to add about 5 db around 150 Hz and it's perfect as well. Not sure if I would like to keep it that way for other tracks though.

 

[juliangst]

As an idiot I am gonna make a guess for fun. I would imagine a lot of the SUBJECTIVE 'dynamics' and 'punch' is due to the recording itself and as these are accurate headphones I assume? Then there is no artificially inflated much of anything and as such the tricks/tweaks many headphone/speaker manufacturers make which stray from accuracy in favour of making somebodies ear perk up are simply not there - because it's not there in the recording.

A lot of those audiophile terms are mostly bullshit but those reviewers say that they are less dynamic than other headphones, and I don't really know what they are referring to.

 

[solderdude]

When one of their favorite headphones has a (mild) lift between 1kHz and 4kHz for instance they will sound more 'forward/clearer/dynamic' and when one compares that to the Expanse it could be perceived as less 'dynamic'. In most cases people compare a headphone to something else (speakers or headphones) and may regard that tonality as 'reference' because they like that. All headphones (not just Expanse) that do not have that tonal character may well be perceived as 'less dynamic'.
It thus could be a mistaken identity of 'dynamics'.

 

[Rthomas]

A lot of those audiophile terms are mostly bullshit but those reviewers say that they are less dynamic than other headphones, and I don't really know what they are referring to.

I wanted to test this for myself so I bought an Audeze LCD-5 and a DCA Stealth and used both with an RME ADI 2 DAC.

Both headphones were more than loud enough for me from the RME.

The DCA Stealth have a lot more bass than the LCD5 according to the measurements but on my head, the Audezes just sounded more powerful and punchy. Much closer to what speakers sound like. The DCA bass sounded ''gentle'' in comparison.

I don't wear glasses and have very short hair. The LCD-5 has a pretty strong clamp but the DCA has almost zero clamp.

My guess is that this is the reason for the difference. Also DCA headphones use a lot of damping in front of the driver the Audeze has almost nothing.

You can call my findings subjective BS but when multiple people report similar findings there may be a small chance that there is something going on.

My advice is to try both preferably at home at normal volumes.

Crank up any of these excellent headphones to 100db and they're going to sound impressive. My question is how do they sound at normal listening levels.

The Audezes were engaging and fun, the Stealth not as much so I returned the Stealth. I can afford to keep both but there was no chance the Stealth would get any head time so it was a waste to keepi it although it is gorgeous and a piece of art IMHO.

 

[oscar_dziki]

When one of their favorite headphones has a (mild) lift between 1kHz and 4kHz for instance they will sound more 'forward/clearer/dynamic' and when one compares that to the Expanse it could be perceived as less 'dynamic'. In most cases people compare a headphone to something else (speakers or headphones) and may regard that tonality as 'reference' because they like that. All headphones (not just Expanse) that do not have that tonal character may well be perceived as 'less dynamic'.
It thus could be a mistaken identity of 'dynamics'.

Sure but I just found out that this bass hump around 150 hz is too little for my liking

 

[Jimbob54]

Artificial Red by Alice in chains

Its by Mad Season, Layne's side project

 

[solderdude]

Sure but I just found out that this bass hump around 150 hz is too little for my liking

That is the 'punch' aspect, the 'dynamics' part is not in the same frequency band. There are quite a lot of headphones around with some extra 'bass' in that frequency band.
For some it translates as 'muddy' or 'fat' bass when it is becoming too much.

 

[oscar_dziki]

Its by Mad Season, Layne's side project

sorry, my bad. Of course, it is.

 

[oscar_dziki]

That is the 'punch' aspect, the 'dynamics' part is not in the same frequency band. There are quite a lot of headphones around with some extra 'bass' in that frequency band.
For some it translates as 'muddy' or 'fat' bass when it is becoming too much.

Yes, that's what I wrote here:

 

[Music1969]

Note that measurements above 7 to 8 kHz are generally not accurate

Hi @amirm this is an interesting comment

Can you expand on it?

Is it a comment about your GRAS 45 rig or all headphone measurements?

Can you also share your thoughts on this comment @Dan Clark ?

cc @solderdude

 

[Resolve]

Hi @amirm this is an interesting comment

Can you expand on it?

Is it a comment about your GRAS 45 rig or all headphone measurements?

Can you also share your thoughts on this comment @Dan Clark ?

cc @solderdude

It's most likely a comment on the rig being rated for accuracy up to 8khz. The thing is, positional variance above that makes accuracy less important in that region anyway.

 

[Music1969]

It's most likely a comment on the rig being rated for accuracy up to 8khz. The thing is, positional variance above that makes accuracy less important in that region anyway.

For EQ'ing purpose, does this mean that for >8 kHz , best to never use any 'peak' filter and better to just use a 'high shelf' to personal taste ?

And is B&K 5128 better > 8 kHz ?

 

[solderdude]

It is said the 5128 is more 'correct' up to 10kHz and maybe even higher.
As Resolve already mentioned the plots vary a lot from the squiggly you see on the screen due to positioning there anyway.
It can show the general 'trend' of tonal balance which is an important aspect. One should just realize that above 6kHz or so one should not look at high Q 'wiggles' in headphone measurements and assume those are really there. They were on that particular fixture, with that particular headphone, on that (or averaged of a few) seatings but the 'general' tonal balance below 5kHz or so is much more determining of 'good sound'.

Just look at various measurements on various rigs. Then see if you can spot some commonalities in that region and EQ the treble part to that (or experiment to see what you like).
You can use shelves or peak filters with a Q up to 3 max in that region.

You can safely forget that headphone measurements are accurate in anyway above a few kHz.
In the 4-7kHz range there is 'sibilance' when the peak is too high.
You can EQ there if you hear sibilance.
In the 7-12kHz range there is 'sharpness' to sounds. When bothered by this simply experiment with filters in that range.
In the 12kHz + range there is 'air'. One can shelf there (but steeply otherwise you get sharpness.

As the range above 5kHz is highly 'personal' anyway, just like bass below 150Hz, you should EQ to taste. One has to use various recordings and see if you can get an 'average' setting that works for you with most music.

The potential 'danger' is that you may deviate from EQ designed by computers and experts in the field. In the end nothing as personal as headphones and taste.
So if you enjoy EQ that differs from what experts claim is the 'only correct one because... bla bla reason' consider that YOU have to enjoy music on that headphone and not the HATS and software that determines what a specific copy of that headphone measures 'correct' on that fixture under the given circumstances.

Headphone measurements are nowhere near the accuracy that speakers can be measured with and far, far, far removed from the accuracy with which electronics can be measured.
Simply too much individual and test fixture as well as acoustical properties of headphones and conditions it is measured in for headphone measurements to be considered accurate. Alas... the 5128 may be 'more accurate' and maybe even reproduce measurements better but is still flawed because of all the variables that are present.

 

[oscar_dziki]

As an idiot I am gonna make a guess for fun. I would imagine a lot of the SUBJECTIVE 'dynamics' and 'punch' is due to the recording itself and as these are accurate headphones I assume? Then there is no artificially inflated much of anything and as such the tricks/tweaks many headphone/speaker manufacturers make which stray from accuracy in favour of making somebodies ear perk up are simply not there - because it's not there in the recording.

I still stand 100% by my first (edit - second) post mind.

I would agree with you, but right now I think there is something wrong with the measurements of on ears headphones or with the Harman target curve, because to my ears Crinacle zero and expanses while having very similar FR are different sounding - crinacle sounding correct, and Expanses lacking in 100-200 Hz department. And I highly doubt that it is the case of a bad recording because I could hardly find a song that has a properly sounding kick drum. I heard drum sets many times, unamplified I know how they should sound. And it boggles my mind, why it sounds correct on Crinacles, and weaker on Expanses while Expanses have this region louder on paper. Expanses while having good sub-bass, they have no kick without EQ. Today I will travel 200 km to check if that is also the problem with the Stealth. And I will switch to Stealths or make a return altogether depending on the outcome of this investigation.

 

[Robbo99999]

It is said the 5128 is more 'correct' up to 10kHz and maybe even higher.
As Resolve already mentioned the plots vary a lot from the squiggly you see on the screen due to positioning there anyway.
It can show the general 'trend' of tonal balance which is an important aspect. One should just realize that above 6kHz or so one should not look at high Q 'wiggles' in headphone measurements and assume those are really there. They were on that particular fixture, with that particular headphone, on that (or averaged of a few) seatings but the 'general' tonal balance below 5kHz or so is much more determining of 'good sound'.

Just look at various measurements on various rigs. Then see if you can spot some commonalities in that region and EQ the treble part to that (or experiment to see what you like).
You can use shelves or peak filters with a Q up to 3 max in that region.

You can safely forget that headphone measurements are accurate in anyway above a few kHz.
In the 4-7kHz range there is 'sibilance' when the peak is too high.
You can EQ there if you hear sibilance.
In the 7-12kHz range there is 'sharpness' to sounds. When bothered by this simply experiment with filters in that range.
In the 12kHz + range there is 'air'. One can shelf there (but steeply otherwise you get sharpness.

As the range above 5kHz is highly 'personal' anyway, just like bass below 150Hz, you should EQ to taste. One has to use various recordings and see if you can get an 'average' setting that works for you with most music.

The potential 'danger' is that you may deviate from EQ designed by computers and experts in the field. In the end nothing as personal as headphones and taste.
So if you enjoy EQ that differs from what experts claim is the 'only correct one because... bla bla reason' consider that YOU have to enjoy music on that headphone and not the HATS and software that determines what a specific copy of that headphone measures 'correct' on that fixture under the given circumstances.

Headphone measurements are nowhere near the accuracy that speakers can be measured with and far, far, far removed from the accuracy with which electronics can be measured.
Simply too much individual and test fixture as well as acoustical properties of headphones and conditions it is measured in for headphone measurements to be considered accurate. Alas... the 5128 may be 'more accurate' and maybe even reproduce measurements better but is still flawed because of all the variables that are present.

To add my own points, I think I agree with you mostly. The one difference is that I think headphone measurements really can be quite accurate (in terms of pertaining to the rig) above a "few kHz" as long as you do quite a few measurements and reseating the headphone each time, and then averaging them out. I've probably posted these graphs already in this forum, but to take my miniDSP EARS rig for example I measured all 3 units of my HD560s headphone on there, measured each one 10 times (reseating the headphone each time) and then averaged them to create an average frequency response line for each unit, and you can see that the 3 units correlate all the way to 20Hz pretty darn accurately in the following graph:

So I'm not saying that this is how an individual will perceive the headphone each time they wear it as there are differences between individuals which could cause different peaks & troughs especially as you go further up the frequency range and would vary also with how the headphone was placed on your head for that particular listening session. So I do agree with you re not using sharp filters up there, and even more so if it's in relation to a unit that you haven't measured (ie if you're using a measurement of your headphone model off the internet) - due to uncertainties re unit to unit variation. But I suppose what I am saying is that it's possible to accurately characterise a headphone in terms how it measures on that rig by doing multiple measurements & averaging them - and from my results in that graph you can see that it's possible to do even up to 20kHz as the averages of each of the 3 units are still close together above 10kHz (which I found surprising & especially considering I never expected 3 units of a headphone model to measure so close to each other through the whole frequency range).

 

[solderdude]

I suppose what I am saying is that it's possible to accurately characterise a headphone in terms how it measures on that rig by doing multiple measurements & averaging them

It is possible to get a decent squiggly on paper by averaging and that it says something about the tonal balance and it is 'accurate' to that fixture.
It is also quite possible to get a measurement to reproduce quite well and 'follow' that average quite well when seatings are not extreme.
But that was not my point.

My point is that a specific model and copy of a headphone measured on various fixtures and on actual ears differs above a few kHz and it is impossible to tell which one is the most accurate to 'the average' of those copies and on one's ears.

So yes, measurements on a specific fixture can reproduce well but it does not mean someone else can get an accurate EQ on that measurement.

 

[Zenairis]

Well I decided to sell my Expanse. Hoping to get at least $3100 for it. I decided I like the Stealth more personal preference wise.

 

[Robbo99999]

It is possible to get a decent squiggly on paper by averaging and that it says something about the tonal balance and it is 'accurate' to that fixture.
It is also quite possible to get a measurement to reproduce quite well and 'follow' that average quite well when seatings are not extreme.
But that was not my point.

My point is that a specific model and copy of a headphone measured on various fixtures and on actual ears differs above a few kHz and it is impossible to tell which one is the most accurate to 'the average' of those copies and on one's ears.

So yes, measurements on a specific fixture can reproduce well but it does not mean someone else can get an accurate EQ on that measurement.

Well, yes, I think we agree.

 

[Robbo99999]

Well I decided to sell my Expanse. Hoping to get at least $3100 for it. I decided I like the Stealth more personal preference wise.

What were the differences for you?

 

[Phoney]

In the 12kHz + range there is 'air'. One can shelf there (but steeply otherwise you get sharpness.

What do you mean by steeply? Should I use a higher Q than 0.71 at 10khz and above?