• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

New version of Sennheiser HD 560S (!)

I have reason to believe it is not only different pads.
One of the reasons is swapping pads between old and new and the Poyatu pads.

One thing is clear... the sound did change.
The new one sounds good with just a little 5kHz reduction (just like HD58X and HD660S needs)



I would not trust e-bay if I wanted original pads but rather buy them from Sennheiser directly or one of their official representatives.
These might be real, they might not be.
True, but I bought a shorter cable to replace the long cable that came with the original version of the HD 560s, and it appeared to be genuine Sennheiser. When I bought replacement earpads, I got them from an online shop in the UK. They were always listed as out of stock on the Sennheiser site. Thomann sells them, but they don't sell Sennheiser products to U.S. customers.

 
@Robbo99999

What a beautiful dumpster fire. You clearly have some anger issue, so I will not bother explaining anything at this point. I don't pretend at all. I see how you type to others. There is a lot of crying on your keyboard.
Hmm, you're a lot wrong, that's all I have to say.
 
Bought another HD560S (June 2022) version and that one is between the original and the 'darker 2022' version and closer to what Robbo measured.
blue = original HD560S, Magenta = HD560S (2022), green = HD560S (2022).png

Drivers are from the same batch so there seems to be quite some tolerance and it is not clear what version you might end up with.
This one is just slightly less bright than the original.

Driver matching is good for both these HD560S.
FR HD560S new 2.png


The plot thickens... there is clearly also manufacturing tolerance in place and Robbo found one with channel imbalance.
 
Bought another HD560S (June 2022) version and that one is between the original and the 'darker 2022' version and closer to what Robbo measured.
View attachment 320526
Drivers are from the same batch so there seems to be quite some tolerance and it is not clear what version you might end up with.
This one is just slightly less bright than the original.

Driver matching is good for both these HD560S.
View attachment 320527

The plot thickens... there is clearly also manufacturing tolerance in place and Robbo found one with channel imbalance.
Solderdude, you say your 2nd unit of the New Version you measured (magenta line in your graph) is closer to what I measured. I don't think that's the case because the New Version I measured is effectively tilted down by -2.1dB above 1200Hz, which you can see from the EQ I did that transforms new version to old version (you can see the High Shelf Filter at filter position #1):
index.php

Whereas your magenta line in your graph looks a lot closer to the old version. It's a bit hard to tell from your graphs due to the scale I think. What degree of tilt in dB would you describe above 1200Hz for your magenta and green line (the 2 units of new version you've measured)? My new version was -2.1dB above 1200Hz. It's true though what you say that your latest unit you measured (magenta) is closer to the old version though - are the pads harder than the softer pads that we noticed in our New Version headphones?
EDIT: when I say "tilt" in this post I'm just referring to a High Shelf Filter, not a true linear tilt.
 
Last edited:
Both new versions have the same cable, same pads, same netting, same driver manufacturing date but very different serial numbers.

There is always some differences in seatings that can amount to the same differences.
In the first series the delta between the 3 versions I had seem very small.
The newer (cheaper ?) versions seem to have a less tight tolerance given the fact that your and my first versions seem to match each other where the 2 versions you and I got (the new ones) seem to be spread further in tolerance.

It appears that in the early ones the EQ may result in a similar tone but in the newer models that may not be the case.
All we can say now is there seems to be a higher tolerance band and tonality can be between the brighter early ones and 'darker' versions.
So a single EQ is likely not to have the same end result.

Part of this tolerance can be attributed to the pads (as I have already shown) and part of it driver and netting.
When we assume netting and pads will be easy to control in quality the driver remains as another contributor.

For me the consequence is that when I buy 2nd hand HD560S in order to filter modify them I have to create custom filters for each one.
The old ones all measured the same so one filter was enough. Now I need to measure, design and test each one of them before putting them up for sale.
 
Both new versions have the same cable, same pads, same netting, same driver manufacturing date but very different serial numbers.

There is always some differences in seatings that can amount to the same differences.
In the first series the delta between the 3 versions I had seem very small.
The newer (cheaper ?) versions seem to have a less tight tolerance given the fact that your and my first versions seem to match each other where the 2 versions you and I got (the new ones) seem to be spread further in tolerance.

It appears that in the early ones the EQ may result in a similar tone but in the newer models that may not be the case.
All we can say now is there seems to be a higher tolerance band and tonality can be between the brighter early ones and 'darker' versions.
So a single EQ is likely not to have the same end result.

Part of this tolerance can be attributed to the pads (as I have already shown) and part of it driver and netting.
When we assume netting and pads will be easy to control in quality the driver remains as another contributor.

For me the consequence is that when I buy 2nd hand HD560S in order to filter modify them I have to create custom filters for each one.
The old ones all measured the same so one filter was enough. Now I need to measure, design and test each one of them before putting them up for sale.
So you're saying that the pad softness between your New Version Unit #1 & #2 are the same, as well as the netting being the same "density", yet they measure differently, whereas in your earlier work you show that pad softness was the biggest contributory factory to frequency change - that doesn't quite add up to what you're showing here with the magenta line - don't you think so?

What do you mean by "seatings"? Are you telling me that you're only doing one measurement of each unit & then posting that up - you're not doing an average of multiple reseats? If that's the case then that's your methodology at fault, you can only accurately compare different units as long as you measure each one thoroughly through multiple reseats on the rig.

(You're starting to worry me, with regards to the last 2 paragraphs of points I've brought up)

It's true that we've both seen tighter tolerances between the old version that we've both measured, most certainly in terms of channel balance with regards to my units, but it lucked out (perhaps) that the New Version Unit #1 of mine that had the bad channel balance also had both channels average out at practically identical frequency response as the average of New Version Unit #2 that I've got. It is pointing towards the Old Version having tighter tolerances though, within the limited number that we've analysed.

EDIT: and why didn't you answer my questions in my previous post, this is supposed to be a conversation right and you're a smart individual? (with regards to the "tilt" points that I brought up, and the fact that you said that your latest measurement matched what I had measured, which I couldn't see - details in my previous post, not retyping here).

EDIT #2: apologies for my shortness (negativity), I suppose I'm a bit annoyed that it's possible we don't have a clear cut answer on what the New Version actually is when I thought we had it nailed down, but it's kinda compounded by the "points that don't add up" and "revelations" that I talk about in the first 2 paragraphs of this post in relation to your work.
 
Last edited:
I suppose I'm a bit annoyed that it's possible we don't have a clear cut answer on what the New Version actually is when I thought we had it nailed down, but it's kinda compounded by the "points that don't add up" and "revelations" that I talk about in the first 2 paragraphs of this post in relation to your work.

Basically it would appear as the newer versions seem to have different pads (confirmed by swapping pads), different cable (no consequence), different netting (almost no consequence) and as it seems some variability in the drivers too and judging from your copy, occasionally poor QC.
3 of the 4 measured new versions appear to have good channel matching.
That's all that is certain.
There simply seems to be a larger production tolerance than before.

I merely report things how I find it and tend to draw conclusions based on knowledge I have at that point in time.

As in regards to the pads.
You could verify this yourself but seem reluctant to swap pads. They are removable you know.
The HD560S I had here are all boxed up to ship tomorrow so can't do more measurements until the next one comes up.

So... basically all I can conclude now is that there seems to be quite some variance in FR so, unlike the older ones, there does not seem to be a single 'correct' pre-chewed EQ possible.

I usually tend to not react to certain questions, annoying I know...;)
 
@solderdude , any chance you could send all your HD560s measurement data to Oratory? A few days ago I just put him in the picture that simply we've just measured a consistent trend on the New Version vs Old Version, with the view that he can decide if thinks it warrants providing a delineation in his measurements between Old Version and New Version (ie offering two different EQ's). I've sent him all my info, and he mentioned you, feel free to send him your data on reddit. It's just in interest of the community. Currently he is thinking that there is too much overlap in the confidence levels. Intuitively I think they're different though based on my own measurements as there is very high correlation within the two different groupings, but your latest measurement does throw a spanner in the works. He did mention he'd be interested in your data too as that could add to what I gave him. Me, personally, I'm gonna still recommend people with the New Version add my EQ that turns New Version into Old Version before applying the Oratory EQ - until I'm convinced otherwise.
 
Basically it would appear as the newer versions seem to have different pads (confirmed by swapping pads), different cable (no consequence), different netting (almost no consequence) and as it seems some variability in the drivers too and judging from your copy, occasionally poor QC.
3 of the 4 measured new versions appear to have good channel matching.
That's all that is certain.
There simply seems to be a larger production tolerance than before.

I merely report things how I find it and tend to draw conclusions based on knowledge I have at that point in time.

As in regards to the pads.
You could verify this yourself but seem reluctant to swap pads. They are removable you know.
The HD560S I had here are all boxed up to ship tomorrow so can't do more measurements until the next one comes up.

So... basically all I can conclude now is that there seems to be quite some variance in FR so, unlike the older ones, there does not seem to be a single 'correct' pre-chewed EQ possible.

I usually tend to not react to certain questions, annoying I know...;)
So you're inferring that the Unit #2 of the New Version you measured (magenta line in the graph) does indeed have the softer pads that we know that the New Version has, yet it measures nearly the same as the Old Version - so you're saying the drivers are the ones offsetting that difference, because of course you showed a week or so ago that the main element of difference to the frequency response was due to the pads? Which indicates that your magenta line is an outlier unit or perhaps for some reason Sennheiser tried to match drivers with pads to almost target the old version frequency response, which doesn't really make sense as an idea because all the other 3 units of New Version that we've measured are broadly showing the same downwards tilt above 1200Hz in relation to the Old Version, and prior attributed by you to the pad softness change. So it feels like the magenta line New Version Unit #2 is a bit of an outlier.

EDIT: I'm still not happy that you're ignoring important points & questions that I'm bringing up in both of my prior posts, because it undermines the quality of the conversation. I can accept it, but it's a slur on you, but I'm not gonna get angry about it.

EDIT#2: yeah, I'm a bit reluctant to mess around swapping pads, probably because I've got all the measurements & EQ's done based on how they are at the moment - reseating pads might change that, and I don't want to damage anything either, but I'm a practical guy and have worked on cars and indeed soldering my old K702 headphones back together so I'm confident I can do it, just don't want the time & stress of it at the moment, lol. I may do that in the future as a little project to see if it correlates with yours - but I was convinced by your pad change work anyway in relation to the main difference being the softer pads, so didn't feel the need to validate & add to that work of yours.
 
Last edited:
I am not inferring anything just reporting. I really can't say which is an outlier nor what the actual production tolerance is.
All are packed up so can't do any more pad swapping of finding out difference between new 1 and 2.

You are in the perfect position to verify my 'different pad hypothesis' but seem reluctant.
In the name of science you could independently verify or debunk my findings.

Oratory can't do much with my raw REW data I reckon.
There are 3 different ways he can contact me if he wants my REW files, I will gladly send them but won't do that unsolicited.
It is a lot of work as I normally don't do just one measurement but a whole bunch of different measurements so would have to compile a REW file with only relevant and properly named data.

I am not on Reddit, only here, on my own website and a (not active) member on 'the cesspool server' just for reading what's in there.
 
Last edited:
I am not inferring anything either just reporting. I really can't say which is an outlier nor what the actual production tolerance is.
All are packed up so can't do any more pad swapping of finding out difference between new 1 and 2.

You are in the perfect position to verify my 'different pad hypothesis' but seem reluctant.
In the name of science you could independently verify or debunk my findings.

Oratory can't do much with my raw REW data I reckon.
There are 3 different ways he can contact me if he wants my REW files, I will gladly send them but won't do that unsolicited.
It is a lot of work as I normally don't do just one measurement but a whole bunch of different measurements so would have to compile a REW file with only relevant and properly named data.

I am not on Reddit, only here, on my own website and a (not active) member on 'the cesspool server' just for reading what's in there.
"Pad swapping" is not the contentious point here, the way I see it, I'm confident in the work you already did on the pad swapping, to me it's totally obvious that a softer pad will change the frequency response, & you showed that! I don't really know why you can't understand my point on the magenta line being the outlier given you said it has the same softer pad yet measures very very close to the Old Version - which suggests it's a driver difference for that particular unit to offset the soft pad - which as I said flys in the face (contradicts) what is happening with the other 3 units of New Version we've measured - hence "outlier" albeit based on 4 units of New Version measured, which is where you might be struggling to say that it's an outlier when only 4 units measured, but still. I'll let Oratory know that you'd rather be contacted directly by him.
 
I simply can't say it is an outlier or production tolerance in the driver.

Given your measurement of the one you returned it appears as though production tolerance in the drivers is a thing and judging from the channel matching on the good ones it does appear as though Sennheiser does have a bunch of measured drivers they might 'grade' and then assemble headphone from graded batches.
At least that's what makes the most sense.
Some HD560S thus may be brighter and others may be less bright.
One thing is certain part of this seems to be caused by the pads but what you call an outlier may well be a combination of the 'brighter' driver combined with the new pads and the first one I measured may well be a less 'bright' driver with the new pads.
I can't say with confidence that the 2 I measured are the least bright and brightest drivers and that they represent the tolerance of the drivers. There could be even brighter and less bright ones and even outliers beyond that.
 
I simply can't say it is an outlier or production tolerance in the driver.

Given your measurement of the one you returned it appears as though production tolerance in the drivers is a thing and judging from the channel matching on the good ones it does appear as though Sennheiser does have a bunch of measured drivers they might 'grade' and then assemble headphone from graded batches.
At least that's what makes the most sense.
Some HD560S thus may be brighter and others may be less bright.
One thing is certain part of this seems to be caused by the pads but what you call an outlier may well be a combination of the 'brighter' driver combined with the new pads and the first one I measured may well be a less 'bright' driver with the new pads.
I can't say with confidence that the 2 I measured are the least bright and brightest drivers and that they represent the tolerance of the drivers. There could be even brighter and less bright ones and even outliers beyond that.
Well I think we can agree that the New Version does have more of a downwards tilt than the Old Version - even the brightest unit of the New Version you measured (magenta line in your graph) had more downwards tilt than the Old Version. I think it's plausible they may be grading & matching drivers like you say, and perhaps they have a few overly bright drivers that they want to get rid of which they deem acceptable to combine with the new softer pad because the end result is that it's close to the Old Version, even if it may be a fair way off from the majority of the New Version units (conjecture based on the 4 units we've measured). If I had to wager, I'd say that most units of the New Version are gonna be measuring closer to other 3 units that we've measured (ie my 2 units of New Version) and your darker unit (green line in your graph). I'm still gonna recommend people use my EQ that changes New Version to Old Version before using an Oratory EQ, but this new information suggests that it might not hit the nail on the head for each New Version unit out there.

EDIT: kudos to you for measuring another unit too though, we've measured 6 units of Old Version & 4 units of New Version altogether between us (each of us measuring 3 Old & 2 New).
 
Last edited:
Maybe more to follow... depending on how many filter-modified HD560S I can sell.

Yes, they appear to be tuned slightly less bright overall. This is a good thing IMO.

Grading drivers was already done ever since the HD650. It is quite easy to do that in the QC of the drivers after they come from the assembly line. Could even be automated.
This way it is fairly easy to create HPs with matching drivers. From what I have seen from Sennheiser they are quite good at it.
That's also why it isn't wise to replace a driver on a headphone. The new one may not match the original.
 
Maybe more to follow... depending on how many filter-modified HD560S I can sell.

Yes, they appear to be tuned slightly less bright overall. This is a good thing IMO.

Grading drivers was already done ever since the HD650. It is quite easy to do that in the QC of the drivers after they come from the assembly line. Could even be automated.
This way it is fairly easy to create HPs with matching drivers. From what I have seen from Sennheiser they are quite good at it.
That's also why it isn't wise to replace a driver on a headphone. The new one may not match the original.
Ah, that's how you're getting your New Version headphones, good!

I doubt I'll measure more units as I currently have 3 units of Old Version & 1 unit of New Version, after having sent back 1 unit of New Version, which is how I've managed to measure 2 units of New Version......but I am pleased I bought & kept that unit of New Version, because it does sound better at stock than the Old Version, and I use it in situations where no EQ is possible. As a "standalone headphone" without doing quick direct comparisons in same listening session to EQ'd headphones then it sounds really good and hard to fault.
 
Last edited:
I have the old 560s and I am quite happy with it. But I cannot ignore this new "normal" price of $220. The 6xx can now be bought for cheaper. If I was to buy one today and use EQ, it would be the 6xx.
 
Where is this $220 price? German Amazon sells it at €157

If HD 600 is to go by for 6XX performance, then it has worse distortion than 560S.
 
Where is this $220 price? German Amazon sells it at €157

If HD 600 is to go by for 6XX performance, then it has worse distortion than 560S.
US Amazon right now is $220. BestBuy and Sennheiser store are selling for $230.
 
I have the old 560s and I am quite happy with it. But I cannot ignore this new "normal" price of $220. The 6xx can now be bought for cheaper. If I was to buy one today and use EQ, it would be the 6xx.
I think the HD560s is the better headphone, I have both (well the HD560s new version & old version as well as the HD600). It's true though that it seems HD560s is quite expensive over there in the USA, whereas at the moment here in the UK it's at £149 on Amazon (normal price is £169). Unusual for the USA to have worse deals on electronics, but in the case of the HD560s it does seem to be true!
 
I connected hd560s to balanced connection (K7 AKM4493), damn i love the sound again
 
Back
Top Bottom