• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Sennheiser HD 820 Review (headphone)

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
32,916
Likes
110,630
Location
Seattle Area
#1
This is a review and detailed measurements of the Sennheiser HD 820 closed back headphone. It is on kind loan from a member and costs US $2,400.

The 820 is a very large headphone but feels like and fits well on my head:

Sennheiser HD 820 review closed back headphone.jpg


It comes with two different length cables, both of which are quite thick and microphonic. They seem strong enough to hang yourself from them!

The 820 weighs 409 grams and has internal cup dimensions of 62, 80 and 30 (width, height, depth). The driver is angled so the other side is less deep.

The measurements you are about to see are made using a standardized Gras 45C. I searched for any and all measurements I could find online. Alas while a number of them are close to mine, none are using the exact fixture down to coupler and pinna. As you will see, I have confirmed the approximate accuracy of the measurements using Equalization and listening tests. Ultimately headphone measurements are less exact than speakers above a few kilohertz so keep that in mind as you read these tests.

Usually I calibrate the loudness relative to 425 Hz. At this frequency the two channels were wildly out of sync as you will see later so I opted for 1 kHz instead. This can change the filter set you create but net results is the same.

As you see below, the channels did not match even though Sennheiser has a reputation of good matching between drivers. Despite the large cups, slight changes would vary the levels fair bit and at any rate, I could not get the two to match in 15 minutes or so I spent on it.

Sennheiser HD 820 Measurements
There is nothing more important than frequency response of a headphone as each is seemingly different and that difference leaves a very distinct character:

Sennheiser HD 820 Frequency Response Measurements.png


At first glance this looks pretty ugly. We have a dip around 60 Hz and then a giant hump followed with a dip. Thankfully in the all important range of 1 to 3 kHz, we have excellent tracking. Note that if you are comparing my measurements to others, they are syncing the preference point at a different spot so wind up showing this compliance differently. We also have a rather broad notch between 3 and 5 kHz.

Subtracting the response from our preference curve gives us:

Sennheiser HD 820 Relative Frequency Response Measurements.png


I have marked two points of interest there. Let's look at the distortion percentage:

Sennheiser HD 820 Distortion Measurements.png


Starting to the right, it is interesting that we have a distortion in the same 3 to 5 kHz range that there is a dip in the driver response. All the way to the left, there seems to be a sudden rise in distortion at the same dip in bass frequency but distortion level actually doesn't change:

Sennheiser HD 820 THD Distortion Measurements.png


THD% is a ratio of level to distortion so when the level dropped due to that dip, it artificially created a spike in THD%.

Overall, this is a low distortion headphone other than 3 to 5 kHz where our hearing is most sensitive.

Group delay is rather clean other than indicating once again something is wrong around 60 Hz and then where that hump is:

Sennheiser HD 820 Group Delay Measurements.png


We had a running theory that headphones with good spatial effects would have a messy group delay in the lower treble area. Such is not the case here at all.

Impedance is variable and can get quite high:

Sennheiser HD 820 Impedance Measurements.png


So you want to pay attention to how high a voltage a headphone amp produces. Typical phone output is usually low in that department making the 820 not a great fit for them. Efficiency is slightly below average which makes this a worse situation:

Most sensitive closed back headphone review.png


Sennheiser HD 820 Headphone Listening Tests
One look at the frequency response and you think the 820 is going to sound strange but it does not. The good sub-bass response means there is healthy response there. Combined with good spatial qualities, you have fair amount of tail wind behind you to like the sound of the headphone. Still, one is forced to try to correct the response errors to see how much better it can get with equalization:

Sennheiser HD 820 equalization eq filter.png


The three left filters should be self-explanatory and are inverse of the frequency response errors (more or less). Let me put a pin in the right shelf for now.

With this new filter, I not only found the sound more correct, but also more detailed and cleaner. With or without the filters though, I found an emphasis in high frequencies resonances of female vocals. Not lisping mind you. But a bit of extra zing. I went back and listened with my AKG K371 and the same thing was there but not nearly as emphasizes. This bothered me a bit without EQ. With equalization in the 3 to 5 kHz to make up that dip, the effect was hugely magnified and perhaps distorted. I resorted to putting that shelving filter on the right to make up for it but it was no help. I tried many techniques to find the cause of this issue but had to give up and not put that correction in there. It is a shame as filling in that hole improved the spatial effects.

Anyway, with what you see above, this was a very comfortable headphone to wear, with decent amount of isolation and good spatial effects. I am listening to them as I type this.

Conclusions
Hard to imagine what was behind the motivation to produce a headphone with this kind of objective frequency response. Then again, I appreciated the deep bass response out of the box. Pulling down that hump seemingly reduced distortion anyway so I did not mind it. The 3 to 5 kHz remains a puzzle that I could not solve. Usually I get to the bottom of the "story" behind a headphone by the time I am done but here, I declare defeat and leave it at that. :)

Overall, based on post EQ and subjective results, I can recommend the Sennheiser HD 820. Mind you, I personally would not at all pay $2,400 for it. Performance ratings are price independent for the most part so there you are.
------------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 

Attachments

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,698
Likes
18,303
Location
The Neverlands
#2
This headphone measures 'better' with seal somewhat broken.



The dip around 60Hz is what Tyll called 'pad bounce' and the few times I looked into this showed you can change the depth and frequency of the dip when changing pads or compressing them. That's why it also is so prominent in the GD.
The pad likes to vibrate/absorb/delay frequencies around 60Hz. When the seal is changed you can also see the effect and frequency change.
The HD800(S) have very different pads so doesn't have this effect.
Ignore the dip between 1kHz and 5kHz it is caused by the lack of pinna gain.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
32,916
Likes
110,630
Location
Seattle Area
Thread Starter #3
This headphone measures 'better' with seal somewhat broken.
Oh, it have been that way when I had my son test it. He said it didn't have as much sub-bass.
 

sweetchaos

Major Contributor
Forum Donor
The Curator
Joined
Nov 29, 2019
Messages
1,702
Likes
3,924
Location
BC, Canada
#4
To import this PEQ profile into 'Equalizer APO', use:
Preamp: -4.8 dB
Filter 1: ON PK Fc 67 Hz Gain 4.0 dB Q 4.0
Filter 2: ON PK Fc 139 Hz Gain -5.0 dB Q 3.0
Filter 3: ON PK Fc 300 Hz Gain 5.0 dB Q 2.0
Filter 4: ON HS Fc 11000 Hz Gain -4.0 dB Q 1.0
Otherwise, see my PEQ guide.
..................................................................................................................
For those who don't have PEQ-capable app, and want to use GEQs instead.
Preamp: -3.5db
32 -0.9
64 3.6
125 -5.0
250 4.0
500 0.6
1000 -0.4
2000 -0.1
4000 0.0
8000 -1.2
16000 -5.0
Preamp: -5.3db
20 -0.7
25 -0.1
32 -0.2
40 -0.3
50 -0.6
63 3.7
80 0.9
100 -0.2
125 -3.9
160 -2.8
200 0.5
250 2.4
315 4.6
400 1.1
500 0.2
630 0.1
800 -0.0
1000 -0.1
1250 -0.1
1600 -0.1
2000 -0.1
2500 -0.1
3150 -0.1
4000 -0.2
5000 -0.2
6300 -0.4
8000 -0.9
10000 -1.8
12500 -3.4
16000 -5.6
20000 -7.5
If you want to import into "Wavelet" (Android App):
GraphicEQ: 20 -4.9; 21 -4.9; 22 -4.9; 23 -4.9; 24 -4.9; 26 -4.9; 27 -4.9; 29 -4.9; 30 -4.9; 32 -4.9; 34 -4.9; 36 -4.8; 38 -4.8; 40 -4.8; 43 -4.7; 45 -4.7; 48 -4.5; 50 -4.4; 53 -4.1; 56 -3.7; 59 -3.1; 63 -1.8; 66 -1.2; 70 -1.6; 74 -2.7; 78 -3.6; 83 -4.3; 87 -4.6; 92 -5.0; 97 -5.3; 103 -5.7; 109 -6.1; 115 -6.7; 121 -7.5; 128 -8.5; 136 -9.4; 143 -9.2; 151 -8.3; 160 -7.2; 169 -6.2; 178 -5.6; 188 -5.0; 199 -4.5; 210 -4.0; 222 -3.4; 235 -2.8; 248 -2.1; 262 -1.4; 277 -0.7; 292 -0.2; 309 -0.2; 326 -0.6; 345 -1.3; 364 -2.0; 385 -2.6; 406 -3.0; 429 -3.4; 453 -3.7; 479 -4.0; 506 -4.1; 534 -4.3; 565 -4.4; 596 -4.5; 630 -4.5; 665 -4.6; 703 -4.7; 743 -4.7; 784 -4.7; 829 -4.8; 875 -4.8; 924 -4.8; 977 -4.8; 1032 -4.8; 1090 -4.9; 1151 -4.9; 1216 -4.9; 1284 -4.9; 1357 -4.9; 1433 -4.9; 1514 -4.9; 1599 -4.9; 1689 -4.9; 1784 -4.9; 1885 -4.9; 1991 -5.0; 2103 -5.0; 2221 -5.0; 2347 -5.0; 2479 -5.0; 2618 -5.0; 2766 -5.0; 2921 -5.0; 3086 -5.0; 3260 -5.0; 3443 -5.0; 3637 -5.0; 3842 -5.1; 4058 -5.1; 4287 -5.1; 4528 -5.1; 4783 -5.1; 5052 -5.2; 5337 -5.2; 5637 -5.3; 5955 -5.3; 6290 -5.4; 6644 -5.5; 7018 -5.6; 7414 -5.7; 7831 -5.8; 8272 -6.0; 8738 -6.1; 9230 -6.3; 9749 -6.5; 10298 -6.7; 10878 -6.9; 11490 -7.1; 12137 -7.4; 12821 -7.6; 13543 -7.7; 14305 -7.9; 15110 -8.1; 15961 -8.2; 16860 -8.4; 17809 -8.5; 18812 -8.6; 19871 -9.0
Otherwise, see my GEQ guide.
 
Last edited:

GWolfman

Active Member
Forum Donor
Joined
Nov 3, 2020
Messages
170
Likes
195
#6
Thanks for the review.

I've never worn anything so expensive on my head!
 

GWolfman

Active Member
Forum Donor
Joined
Nov 3, 2020
Messages
170
Likes
195
#7
@amirm You talk a lot about spatial effects. What would you say are your top 2-3 that create this effect (as far as you recall)? Also, can you recommend a top 1-2 that are affordable (LOL)?
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,698
Likes
18,303
Location
The Neverlands
#8
"Loose thougt": A loudspeaker with this FR, would get the headless panther treatment... Why can't headphones have FR as good as speakers? Is it the diaphragm working full range?
You can get really close to speaker alike FR but is very hard to do with closed headphones.
The reason is the small restricted space and the fact that to get speaker like response in such a small 'sound chamber' with its unique response and target (the target is not flat like speakers in an-echoic conditions) requires playing around with pads, angling, driver size, membrane etc. to create that. The problem here is that each changed aspect also affects parts of the FR you did not want changed. And thus one can only do so much.
Multiple drivers doesn't really help here either.
 

don'ttrustauthority

Active Member
Forum Donor
Joined
Feb 10, 2021
Messages
208
Likes
129
#9
Would spatial effects be related to the 'speed' of the headphone, or is that covered in your measurements?
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,698
Likes
18,303
Location
The Neverlands
#10
@amirm You talk a lot about spatial effects. What would you say are your top 2-3 that create this effect (as far as you recall)? Also, can you recommend a top 1-2 that are affordable (LOL)?
The HD820, to me, had very similar spatial accuracy (would not call it an effect) as the HD800(S) and has 'fixed' treble response to boot.
It doesn't measure nor sound that well in the lows. The lows are 'impressive' but a bit 'disconnected' from the rest (dip at 300Hz) but otherwise just as detailed and fine as HD800S.

It sounds a LOT better than how it measures (in the lows), certainly with some corrective EQ, and kind of succeeded in making a closed version of the HD800(S) with improved treble but weren't able to get the f'ed up bass right. Still a good sounding closed headphone overall but there are closed alternatives here and there. Just not with the exact same positive traits of the HD800(S).

The spatial properties are from the 10% angled large drivers further away from the ear with lots of 'room' around the ear. Not many headphones do this. I don't think it will be easy to prove 'spatial' effects of headphones in a measurement. I don't see it in FP nor proper measurement rigs.
It is one of those things that may need a different test signal to show it.
 

renaudrenaud

Addicted to Fun and Learning
Joined
Apr 20, 2019
Messages
827
Likes
1,504
Location
Paris
#11
Conclusions
Hard to imagine what was behind the motivation to produce a headphone with this kind of objective frequency response.


Maybe if all headphones adhere precisely to the Harman curve, they will sound more or less the same.

With this FR, you produce something really different, and because it's expansive, it becomes a benchmark.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
32,916
Likes
110,630
Location
Seattle Area
Thread Starter #13

ZolaIII

Addicted to Fun and Learning
Joined
Jul 28, 2019
Messages
939
Likes
537
#14
You can get really close to speaker alike FR but is very hard to do with closed headphones.
The reason is the small restricted space and the fact that to get speaker like response in such a small 'sound chamber' with its unique response and target (the target is not flat like speakers in an-echoic conditions) requires playing around with pads, angling, driver size, membrane etc. to create that. The problem here is that each changed aspect also affects parts of the FR you did not want changed. And thus one can only do so much.
Multiple drivers doesn't really help here either.
Yet Pioneer did it 45 years ago and to my better knowledge nobody ever again.
 

sweetchaos

Major Contributor
Forum Donor
The Curator
Joined
Nov 29, 2019
Messages
1,702
Likes
3,924
Location
BC, Canada
#15
Amir vs Oratory1990:
amir oratory.gif

Amir vs Jaakkopasanen (measured by Oratory1990):
amir jaakko.gif

Oratory1990 vs Jaakkopasanen (measured by Oratory1990):
oratory jaakko.gif

PEQ profile by Amir (from my post #4 above):
Preamp: -4.8 dB
Filter 1: ON PK Fc 67 Hz Gain 4.0 dB Q 4.0
Filter 2: ON PK Fc 139 Hz Gain -5.0 dB Q 3.0
Filter 3: ON PK Fc 300 Hz Gain 5.0 dB Q 2.0
Filter 4: ON HS Fc 11000 Hz Gain -4.0 dB Q 1.0

PEQ profile by Oratory1990:
Preamp: -5.5 dB
Filter 1: ON PK Fc 74 Hz Gain 6.8 dB Q 2.8
Filter 2: ON PK Fc 158 Hz Gain -8.4 dB Q 1.1
Filter 3: ON PK Fc 170 Hz Gain -8.4 dB Q 0.17
Filter 4: ON PK Fc 268 Hz Gain 2.3 dB Q 3.0
Filter 5: ON PK Fc 270 Hz Gain 14.5 dB Q 0.8
Filter 6: ON PK Fc 1000 Hz Gain -2.3 dB Q 2.0
Filter 7: ON PK Fc 2850 Hz Gain -4.5 dB Q 3.5
Filter 8: ON PK Fc 4000 Hz Gain 5.5 dB Q 0.7
Filter 9: ON PK Fc 5130 Hz Gain -7.5 dB Q 3.8
Filter 10: ON PK Fc 7450 Hz Gain -2.9 dB Q 4.0

PEQ profile by Jaakkopasanen (measured by Oratory1990):
Preamp: -6.6 dB
Filter 1: ON PK Fc 27 Hz Gain -5.0 dB Q 1.08
Filter 2: ON PK Fc 46 Hz Gain -4.1 dB Q 2.33
Filter 3: ON PK Fc 141 Hz Gain -9.5 dB Q 1.43
Filter 4: ON PK Fc 291 Hz Gain 8.2 dB Q 2.39
Filter 5: ON PK Fc 3742 Hz Gain 5.1 dB Q 3.83
Filter 6: ON PK Fc 75 Hz Gain 2.6 dB Q 6.57
Filter 7: ON PK Fc 980 Hz Gain -3.1 dB Q 2.12
Filter 8: ON PK Fc 1860 Hz Gain 1.9 dB Q 2.70
Filter 9: ON PK Fc 5109 Hz Gain -3.0 dB Q 6.65
Filter 10: ON PK Fc 9677 Hz Gain 3.1 dB Q 1.36
 

alont

Member
Forum Donor
Joined
Sep 1, 2018
Messages
40
Likes
83
#16
I gotta say that I'm surprised by the panther and overall conclusions. The measurements paint a fairly horrible picture - look at those incredibly uneven lows and mids! My subjective impressions weren't any better - tonally, these headphones were a disaster to my ears.
 

thewas

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
2,488
Likes
5,395
#17
I gotta say that I'm surprised by the panther and overall conclusions. The measurements paint a fairly horrible picture - look at those incredibly uneven lows and mids! My subjective impressions weren't any better - tonally, these headphones were a disaster to my ears.
I agree, $2400 for a really not great measuring headphone, any electronic device with such a poor combination would deservedly get a broken panther...
 

MayaTlab

Active Member
Joined
Aug 15, 2020
Messages
159
Likes
218
#19
This headphone measures 'better' with seal somewhat broken.



The dip around 60Hz is what Tyll called 'pad bounce' and the few times I looked into this showed you can change the depth and frequency of the dip when changing pads or compressing them. That's why it also is so prominent in the GD.
The pad likes to vibrate/absorb/delay frequencies around 60Hz. When the seal is changed you can also see the effect and frequency change.
The HD800(S) have very different pads so doesn't have this effect.
If pad bounce is related to pad compression, wouldn't it logically then proceed that, since pads often are unevenly compressed when mounted around our ears, the pad bounce effect is distributed across frequencies and nullified ?
Could it explain that Rtings, which measure FR on real humans at lower frequencies, or my own on head measurements, never really show that pad bounce, or at least not to the same extent ?
 

Similar threads

Top Bottom