New version of Sennheiser HD 560S (!)

[solderdude]

-3dB is quite an audible change. I measured it a few times with varying positions to be sure. Could be the pads (does not have to be the driver) and will check the next time I pull out the test system as that could shed a bit more light. I only have filtered HD560S so direct comparison is not possible unless I open it up and short the filter.

 

[Robbo99999]

-3dB is quite an audible change. I measured it a few times with varying positions to be sure. Could be the pads (does not have to be the driver) and will check the next time I pull out the test system as that could shed a bit more light. I only have filtered HD560S so direct comparison is not possible unless I open it up and short the filter.

Yes, I see what you mean - you're thinking about swapping the pads of the "new" version over to your "old" HD560s and then remeasuring it to see if it's the pads or the driver - good idea.

In the other thread you said you'd contact Jermo too, maybe he could send you a couple more of the "new" HD560s to measure? (to go some way to help rule out unit to unit variation as the explanation)

 

[solderdude]

I will only contact Jermo when it turns out the driver may have undergone a silent change. If it turns out to be the pads then the mystery is solved and there will be no need to measure other copies nor contact Jermo.

For now this is what I found to be the case and matched reports from random people on the web comparing the 2.
The pad swap thing idea came later (after my initial post) so more will follow.

 

[Robbo99999]

I will only contact Jermo when it turns out the driver may have undergone a silent change. If it turns out to be the pads then the mystery is solved and there will be no need to measure other copies nor contact Jermo.

For now this is what I found to be the case and matched reports from random people on the web comparing the 2.
The pad swap thing idea came later (after my initial post) so more will follow.

Well, in my experience unit to unit variation can be caused by manufacturing variance in pads too, so I don't think you can conclude that it's a sonically new version if it turns out the pads are what is causing the difference (in your future experiment). To me the only way to be sure is to measure a few more units or to have absolute confidence in whatever Jermo tells re any potential changes.

 

[solderdude]

As said.... the next measurements will give clarity in which case more copies are not needed.
When it turns out to be the driver then I will contact Jermo about this.

The one I have here is measurable different from my 3 copies and your 3 copies also measure closely the same.

 

[Robbo99999]

As said.... the next measurements will give clarity in which case more copies are not needed.
When it turns out to be the driver then I will contact Jermo about this.

The one I have here is measurable different from my 3 copies and your 3 copies also measure closely the same.

You didn't understand my post then or chose to ignore it. Your next measurements will not give clarity as to whether it is for sure a new version or just unit to unit variation, but yes it will indicate whether it's the driver or the pads - but as I said there's manufacturing unit to unit variation within the pads, so you can't say for sure "new version due to new pads", not unless you measure a few more units. But, but, given the fact we've both measured 3 units of HD560s and the unit to unit variation is low, then there's a fair chance it's a new version - but don't be sloppy in jumping to conclusions. (Or you could get confirmation from Jermo that there is indeed a different sonic trend now with HD560s).

 

[solderdude]

Oh.. I understood your concerns.
Since you already have 3 of them you could also buy a new copy and see how it compares to your copies in which case you could confirm or debunk my findings.
I just reported what I found.

 

[Robbo99999]

Oh.. I understood your concerns.
Since you already have 3 of them you could also buy a new copy and see how it compares to your copies in which case you could confirm or debunk my findings.
I just reported what I found.

Indeed re me buying a new version & measuring it! With your Jermo contact I reckon you could wangle him to send you at least two units (free of charge) to measure for the community, to validate whatever statement he may or may not make re any new potential version. Ok, good that you understood my concerns.

 

[solderdude]

I only want to bother him if I got some hard evidence. I have never asked for a review headphone. Either headphones are sent in (often Superlux, Sennheiser, OLLO, Drop (in the past)) or I buy them and keep them or owners send them in.
Only a few headphones I bought I did send back but don't like to do that.
I am sure if I asked Sennheiser for review models they would probably send them but for me this is a hobby and not a business and don't like to take advantage of businesses or people.

 

[Robbo99999]

I only want to bother him if I got some hard evidence. I have never asked for a review headphone. Either headphones are sent in (often Superlux, Sennheiser, OLLO, Drop (in the past)) or I buy them and keep them or owners send them in.
Only a few headphones I bought I did send back but don't like to do that.
I am sure if I asked Sennheiser for review models they would probably send them but for me this is a hobby and not a business and don't like to take advantage of businesses or people.

Go on man, you got the contacts there, and Jermo seemed super enthusiastic about his work the integrity of that whole approach (from his vids), so I can't imagine him not wanting to engage (indirectly) with the community by allowing you to measure two of the "new" headphones, and potentially he could elucidate whether or not any sonic changes have been made.......if he does want to say yay or nay re sonic changes then I'm sure he'd be happy to provide a few samples to show measurements to back it up. He was super enthusiastic about his work, very commendable......as an idea link him to the thread & the website here (I'm sure he knows ASR already), engaging with it positively is beneficial. Sennheiser are the company I have the most faith in re their processes, low unit to unit variation, etc, I think they have a lot of integrity in their approach, I'd hope they'd want to help & be open.

 

[isostasy]

There was an explanation of manufacturing tolerances and part matching regarding the HD800(S) in the recent Sennheiser video.

I imagine very small changes along the production line, like the supplier of the paper used in front of the driver changing it slightly, could have big differences on sound. Guessing this is why the screen on the HD580/600/650 changed a few times, rather than it being an optional choice.

 

[solderdude]

post 2022 version

I decided to dig a little further.

From about end 2022 there were some changes to the HD560S.
These are not just cosmetic (headband padding) and the shorter (now 1.8m) cable with a different connector. The sound and measurements also differ slightly and audible.
The differences are not very big in the overall sound and comfort though and it still is a HD560S.

The most obvious differences are the cloth used for the headband.

Top one is the 2022 version, bottom one is the original one. Foam feels the same.

The other change is cable length and connector. The original version came with a 3m cable with 6.3mm connector + short extension cable that converts to a 3.5mm connector.
The new cable is 1.8m and has a threaded 3.5mm connector with a screw-on 6.3mm connector which is much more convenient and will surely be cheaper to manufacture.

The changes, however, did not stop there. The foam in the earpads seems a bit firmer (could be aging of the older pads which Sennheiser pads are notorious for).
Cloth used seems the same. Height and shape seems the same.
Below some frequency response measurements of the 2022 HD560S versus 3 measurements of earlier models (all right channel).

Note: the vertical dB scale in this plot is 1dB/div where I normally use 5dB/div.
This is because the differences are small and are more easily visible this way.
The plots have ‘acoustical smoothing’ applied as well to make differences more clearly as well.

The 2022 version has about 2-3 dB less treble and starts to drop -off a bit above 1kHz.
The most common complaint about the HD560S is that it was too bright (upper mids/lower treble).
For this reason I use it with the filter above which removes this downside.
The interesting thing is that the newer version has almost the same attenuation as what it achieved with the filter. The 2022 HD560S thus is a bit less bright sounding. This is a good thing and it seems Sennheiser has taken these comments to heart and improved on it. They did not bother to inform the public about this or name this a HD560S-II or something. A silent revision it seems (other than the obvious differences).

To verify check if this is caused by the pads the 2022 version was measured with its own pads, the pads from an original HD560S and also with Poyatu replacement pads.

Looks like the bulk of the differences is indeed caused by the similar looking but different feeling pads (foam is firmer on the 2022 pads). About 1-2dB lower treble between 2-7kHz

To check this a little bit further the same pads were also measured on a (filter modified) older HD560S.

The same effect is seen here. A bit lower treble but less of a difference. Between 2-7kHz there is about 1.5dB less treble.

That is part of the puzzle as the difference seems to be twice that of what is achieved with the pads.

Another thing that caught my eye when swapping pads was that in the early version you can see the driver clearly through the dust guard (netting). In the 2022 version this was harder to see.
Below the difference between the 2 dust guard (held against a back-light)

On the left the original dust guard, on the right the 2022 version dust guard. It seems less permeable. One can use different materials in front of a driver to influence treble as shown in this article. Below the effect of the thicker (2022) material. The difference is marginal (about 0.5dB max).

It appears that this does not do much but adds a little combined with the 2022 pads.
This still is 1 dB short of the total difference between the early and 2022 version.
Also there is a tiny bit more bass extension (not an audible amount but measurable) which could suggest that, aside from production tolerances, could indicate a slight change in the driver as well.
This can only be verified by measuring more 2022 and later HD560S but I don’t have these.
Below pictures of the drivers of the old version (left) and 2022 version (production date on the driver is June 2022).
There is a different stamp used. The earlier one had IE00576672 on it the 2022 version just 576672 and the plastic used on the 2022 version seems a bit more opaque.

Whether or not this also means a small change in the driver (apart from the stamping) is hard to tell without testing more of them with certainty. The differences found with the pads and duct guard do not seem to be the only reason for the differences in measured response. Also it is not clear whether or not this is just production tolerances, small changes in the material supply chain nor that it was an intentional change to address the often heard comment about the excess brightness.

In any case the 2022 version without a filter comes very close to the earlier version with a filter and are hard to distinguish. Even when tested blind this was difficult to hear. Due to a slightly different clamping force and pad softness a true ‘blind’ test was not possible.
It is clear that there are audible differences between the early and post 2022 HD560S and that the brightness of the 2022 version is audibly less.

Phase and distortion measurements were still almost unchanged.

 

[IAtaman]

It is clear that there are audible differences between the early and post 2022 HD560S and that the brightness of the 2022 version is audibly less.

That sounds like a positive conclusion. I wish manufacturers were a bit more open with the variations / versions of their products and figuring out what is going on did not require users to try and reverse engineer them.

 

[solderdude]

Yes, this is quite common.... not being open about all (even small) changes.
In the end it is still a HD560S.
Funnily enough you can buy the same headphone but with 2 cables and different logo as HD400Pro so they are selling the same headphone with 2 different designators.

 

[Qaz]

In the end it is still a HD560S.

as ones before 2022 with a filter mod.

 

[solderdude]

Yes, closer to that arguably. The 2022 HD560S is still a very similar to an early HD560S, but with a little less 'brightness' which is a good thing in the end.
Some will like the tonal change and others may say it isn't a HD560S any more as it doesn't sound exactly the same.
The overall tonal balance and rest of the sound is still the same.
It also means the familiar digital EQ based on early models needs some adjusting.

I am going to try and lower the brightness a tad more using a filter by lowering around 5kHz a bit more (similar to HD58X/HD660S) in the coming weeks. Just to see if it can be improved a bit more passively.

 

[Qaz]

@solderdude

Do you take into account wearing off pads over time, resulting in reduced amount off treble? Maybe it's better to have a tad more treble energy at the beginning. Digital EQ seems to be a better choice here and faster to check without doing any mod.

 

[thewas]

In any case the 2022 version without a filter comes very close to the earlier version with a filter and are hard to distinguish.

That is a nice evolution if such a filter isn't necessary anymore, maybe they read your reviews?

 

[solderdude]

That is a nice evolution if such a filter isn't necessary anymore, maybe they read your reviews?

I have been in contact with Jermo and he has a filter I sent him.
This was already before it was launched.

@solderdude

Do you take into account wearing off pads over time, resulting in reduced amount off treble? Maybe it's better to have a tad more treble energy at the beginning. Digital EQ seems to be a better choice here and faster to check without doing any mod.

Yes, pad wear will lower the treble. Below the effect of pad compression (which is not the same as pad wear but gives an idea)

The thing is that the new version had new pads and lower treble.

The filter is not a modification and can be external. When digital EQ is not an option the filter is a viable option.
Digital EQ can obviously correct more than just the treble and is more flexible and can be adjusted over time.
The question is where that EQ is based on (fixture type and placements) and to what target.
Of course one can always only apply a simple parametric filter to do the same but only when digital parametric EQ is available.

Digital EQ is found everywhere on the web based on many measurements. It all depends on the presence of digital manipulation.
There are people that can't or won't do DSP and for them modifications or a filter could be welcome.

It happens to be so that the FR of the new HD560S comes close to the filtered original and is why it was mentioned.

 

[Robbo99999]

post 2022 version

I decided to dig a little further.

From about end 2022 there were some changes to the HD560S.
These are not just cosmetic (headband padding) and the shorter (now 1.8m) cable with a different connector. The sound and measurements also differ slightly and audible.
The differences are not very big in the overall sound and comfort though and it still is a HD560S.

The most obvious differences are the cloth used for the headband.

Top one is the 2022 version, bottom one is the original one. Foam feels the same.

The other change is cable length and connector. The original version came with a 3m cable with 6.3mm connector + short extension cable that converts to a 3.5mm connector.
The new cable is 1.8m and has a threaded 3.5mm connector with a screw-on 6.3mm connector which is much more convenient and will surely be cheaper to manufacture.

The changes, however, did not stop there. The foam in the earpads seems a bit firmer (could be aging of the older pads which Sennheiser pads are notorious for).
Cloth used seems the same. Height and shape seems the same.
Below some frequency response measurements of the 2022 HD560S versus 3 measurements of earlier models (all right channel).

Note: the vertical dB scale in this plot is 1dB/div where I normally use 5dB/div.
This is because the differences are small and are more easily visible this way.
The plots have ‘acoustical smoothing’ applied as well to make differences more clearly as well.

The 2022 version has about 2-3 dB less treble and starts to drop -off a bit above 1kHz.
The most common complaint about the HD560S is that it was too bright (upper mids/lower treble).
For this reason I use it with the filter above which removes this downside.
The interesting thing is that the newer version has almost the same attenuation as what it achieved with the filter. The 2022 HD560S thus is a bit less bright sounding. This is a good thing and it seems Sennheiser has taken these comments to heart and improved on it. They did not bother to inform the public about this or name this a HD560S-II or something. A silent revision it seems (other than the obvious differences).

To verify check if this is caused by the pads the 2022 version was measured with its own pads, the pads from an original HD560S and also with Poyatu replacement pads.

Looks like the bulk of the differences is indeed caused by the similar looking but different feeling pads (foam is firmer on the 2022 pads). About 1-2dB lower treble between 2-7kHz

To check this a little bit further the same pads were also measured on a (filter modified) older HD560S.

The same effect is seen here. A bit lower treble but less of a difference. Between 2-7kHz there is about 1.5dB less treble.

That is part of the puzzle as the difference seems to be twice that of what is achieved with the pads.

Another thing that caught my eye when swapping pads was that in the early version you can see the driver clearly through the dust guard (netting). In the 2022 version this was harder to see.
Below the difference between the 2 dust guard (held against a back-light)

On the left the original dust guard, on the right the 2022 version dust guard. It seems less permeable. One can use different materials in front of a driver to influence treble as shown in this article. Below the effect of the thicker (2022) material. The difference is marginal (about 0.5dB max).

It appears that this does not do much but adds a little combined with the 2022 pads.
This still is 1 dB short of the total difference between the early and 2022 version.
Also there is a tiny bit more bass extension (not an audible amount but measurable) which could suggest that, aside from production tolerances, could indicate a slight change in the driver as well.
This can only be verified by measuring more 2022 and later HD560S but I don’t have these.
Below pictures of the drivers of the old version (left) and 2022 version (production date on the driver is June 2022).
There is a different stamp used. The earlier one had IE00576672 on it the 2022 version just 576672 and the plastic used on the 2022 version seems a bit more opaque.

Whether or not this also means a small change in the driver (apart from the stamping) is hard to tell without testing more of them with certainty. The differences found with the pads and duct guard do not seem to be the only reason for the differences in measured response. Also it is not clear whether or not this is just production tolerances, small changes in the material supply chain nor that it was an intentional change to address the often heard comment about the excess brightness.

In any case the 2022 version without a filter comes very close to the earlier version with a filter and are hard to distinguish. Even when tested blind this was difficult to hear. Due to a slightly different clamping force and pad softness a true ‘blind’ test was not possible.
It is clear that there are audible differences between the early and post 2022 HD560S and that the brightness of the 2022 version is audibly less.

Phase and distortion measurements were still almost unchanged.

Interesting, thanks for all that testing. One thing that bugged me initially when I was reading your post was you calling it the "2022 version", because I bought all three of my HD560s in both 2021 & 2022 and they all measure the same when I measured mine, so I thought no way you can be calling it the "2022 version", but you're basing it on the fact the driver in your "new version" is dated June 2022.....and when did the supposed different pads come about, that was 2023 early wasn't it.....so it seems more of a thing that would affect units bought by the consumer in 2023 rather than 2022, so I want to point that out as calling it the "2022 version" doesn't really marry with people's experience of buying an HD560s in 2022, although possibly if they buy it in 2023 & from now on.

I'm still not totally happy about how this has all panned out because your contact at Sennheiser said "only the cable (shorter and with threaded 3.5mm + 6.3 adapter) and fabric for the headband has changed. No changes to the driver nor to the pads or chassis", which is from your post a short while back:

New version of Sennheiser HD 560S (!)

A new version of Sennheiser HD 560S has arrived. The big question is, does it sound (or measure) differently ? In a post on Reddit, a user has posted the following:

www.audiosciencereview.com

So I really think it would be helpful if you approach your contact at Sennheiser again and ask him what's going on, because it still could just be a unit to unit variation issue that you're seeing that has coincided with some potential small changes in source/spec of pads/dust guard - it's possible these changes in spec (if indeed they have occurred) are not supposed to have this affect on the frequency response, especially given the dude indirectly told you that they've done nothing to change the sound (frequency response) of the headphone. I don't think we've ruled out unit to unit variation until we get confirmation from your contact. I'm not happy with the conflicting information, it casts doubt and gives more sway to unit to unit variation being the cause of what you're seeing during your testing, even though I understand your observations. Could you contact him with your info & see what he has to say?