CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS
Lobby => Headphone Measurements => Topic started by: Marvey on December 09, 2011, 09:32:23 PM
-
Sennheiser HD558 (modded) Frequency Response and CSD Waterfall Plots.
The mods consist of the single foam strip (right behind the driver) removed from inside the housing. I have my doubts the changes are significant (moar bass, etc.) I re-applied the strips and felt the stock sounded slightly more midrangy. That was about it. Maybe I'll measure the stock headphones one day. I'm not really that interested.
These are good headphones though, especially for the price.
-
Im guessing Anax doesn't like these?
-
Im guessing Anax doesn't like these?
Why? I wouldn't have kept them. I'm not done w/ them either. I need some time alone w/ a Dremel.
-
I remembered you saying you're sensitive to 4-5k and the FR has a little "love bump" at 5k.
-
Actually it's the 5-6khz for me. Plus the relative difference before the peak at 4Khz isn't dramatic like say a Pro900 from 4-6khz w/ a 20dB delta. I think I'm more sensitive to the instantaneous change or peak rather than a mild to moderate ringing w/o the larger delta in the FR. Plus the love bumps seem to roll rather smoothly compared to other jagged ridges. Just a guess.
I was actually modding the 558 for gaming originally. Now I'm thinking if I can make the SA5K work somehow w/ a subwoofer and preferably some boom mic apparatus. It's just so efficient from a soundcard. Adding power does little to nothing for the dynamics IME. It also needs a bit more padding on the headband or adjustment to the suspension since the metal bands still do too much work IMO.
-
Actually it's the 5-6khz for me. Plus the relative difference before the peak at 4Khz isn't dramatic like say a Pro900 from 4-6khz w/ a 20dB delta. I think I'm more sensitive to the instantaneous change or peak rather than a mild to moderate ringing w/o the larger delta in the FR. Plus the love bumps seem to roll rather smoothly compared to other jagged ridges. Just a guess.
I was actually modding the 558 for gaming originally. Now I'm thinking if I can make the SA5K work somehow w/ a subwoofer and preferably some boom mic apparatus. It's just so efficient from a soundcard. Adding power does little to nothing for the dynamics IME. It also needs a bit more padding on the headband or adjustment to the suspension since the metal bands still do too much work IMO.
Anax is OK with 5-6k. He doesn't like 6k to 7k.
-
Can we add "love bump" to a list of technical terms somewhere?
-
Actually it's the 5-6khz for me. Plus the relative difference before the peak at 4Khz isn't dramatic like say a Pro900 from 4-6khz w/ a 20dB delta. I think I'm more sensitive to the instantaneous change or peak rather than a mild to moderate ringing w/o the larger delta in the FR. Plus the love bumps seem to roll rather smoothly compared to other jagged ridges. Just a guess.
I was actually modding the 558 for gaming originally. Now I'm thinking if I can make the SA5K work somehow w/ a subwoofer and preferably some boom mic apparatus. It's just so efficient from a soundcard. Adding power does little to nothing for the dynamics IME. It also needs a bit more padding on the headband or adjustment to the suspension since the metal bands still do too much work IMO.
Anax is OK with 5-6k. He doesn't like 6k to 7k.
Well, the Sony EX1000 would beg to differ. I was the first to call them out on that spike at CES a year ago.
http://www.innerfidelity.com/images/SonyMDREX1000.pdf (http://www.innerfidelity.com/images/SonyMDREX1000.pdf)
-
It's probably the 8k spike. Can't tell since no CSDs with Tyll's graphs though.
-
i always liked the sennheiser HD5xx line-up i remember hearing a pair of HD555 when i was a hardcore gamer and use to work at small computer shop as a technician.
-
Just wanted to share some brief impressions and measurements. I have to agree these headphones seem a little midrangy but not too offending to my ears. It is also not the last word in detail retrieval, but really like the tonal balance of these cans and they are fairly pleasant to my ears.
I bought these refurbished for $100, and have no mods AFAIK. The distortion plots are in dB instead of %. I tried to follow Marv's anechoic plate approach when measuring these cans using what I've gathered here and there.
In advance apologies for the zombie thread revival :)p5
-
Open them up and take the foam out. See how that sounds.
-
Just did that... Better!!! headbang
Just from brief impressions, seems a little more open and more bass. Measurements bellow...
Seems to have gone down a little at 3 kHz, and 2.5 kHz ring seems to have decreased. Very impressed on how closely these match with the HD558s Marv measured. Decent driver matching and low product variability.
-
Awesome. I now appoint you as my successor.
-
:)p13 Just replicating/confirming measurements, and learning stuff a long the way... :)p7
-
I never realized these could be had so cheap! I'd take that for $100! Well done Ultra.
As for reviving the zombie thread, I like the idea of bringing up the old measurement thread for something like this, it keeps everything in one place.
-
Thanks Ship! Yes, Dakmart/Dahmart and Adgtron (which claim to be Senn authorized dealers) sell these refurbished/used through Amazon at very affordable prices. TBH I was a little afraid of being scammed, but these seem genuine.
Bellow are some overlays of before mod (stock) in white vs after mod (foam removal per Anax suggestion) in green/red. Seems the mod pushes the 2kHz to 4kHz (high mids) down a little relative to the rest of the spectrum. Liking the mod quite a bit. Thanks guys!
EDIT: Added IMO better visualization by aligning at 1kHz...
EDIT: Plots zoomed below for visualization
Right Channel
(http://www.changstar.com/index.php?action=dlattach;topic=65.0;attach=3281;image)
Left Channel
(http://www.changstar.com/index.php?action=dlattach;topic=65.0;attach=3283;image)
-
Yeah, Sennheiser calls the mod, the HD598 (almost 598).
-
So what does the 598 do better than these? I've owned the 555,595, and the 558 but am still curious to know if I am missing something. The 5khz peak on these does bother me a little and was wondering if the same can be said about the 598.
-
So what does the 598 do better than these? I've owned the 555,595, and the 558 but am still curious to know if I am missing something. The 5khz peak on these does bother me a little and was wondering if the same can be said about the 598.
598 is a little brighter, more airy and spacious. The physical difference is mainly a more open web type chassis grill behind the driver rather than a honeycomb. That change alone is audible.
-
So what does the 598 do better than these? I've owned the 555,595, and the 558 but am still curious to know if I am missing something. The 5khz peak on these does bother me a little and was wondering if the same can be said about the 598.
598 is a little brighter, more airy and spacious. The physical difference is mainly a more open web type chassis grill behind the driver rather than a honeycomb. That change alone is audible.
I removed the plastic chassis web grill...thing...and the fabric as well, leaving it completely open much like the HD600/HD650. The metal grill was still in place. I'm curious how those changes would have measured. At the very least, I thought it looked nicer that way. I don't think it changed the sound much, but a little.
-
When I tried the 598's the look made me feel like I had just stolen them from some dude's luxury yacht. They sounded a bit more rich/lush compared to my AKG's but I felt they had a clearer and more detailed sound at the same time and weren't sibilant in nature either. They were also super comfortable.
I had a very positive experience and am wondering if the 558's would be just as good to my ears.
-
EDIT: Post moved here since these are more HD558 relevant.
stv014:
Attached below is the "imptest" loopback file for my HD558 (80 mVrms with no load and using the 2i2).
Also the sweep results (https://www.dropbox.com/s/blgqig4u8xpfb18/sweep_HD558_80mVrms.wav) (same 80 mVrms, using 2i2 and HD558 as load).
-
Also the sweep results (https://www.dropbox.com/s/blgqig4u8xpfb18/sweep_HD558_80mVrms.wav) (same 80 mVrms, using 2i2 and HD558 as load).
This seems to show some D/A or A/D related problems, since there is a high amount of aliasing/imaging, and apparently a 22 kHz lowpass filter, even though both the playback and recording sample rate was 96 kHz. Is it possible that the signal was resampled by Windows ?
-
Yes. Here (https://www.dropbox.com/s/1gvm7j4fdya4ud3/sweep_HD558_150mVrms.flac)is hopefully a better recording at 150mVrms @ 1kHz and 96kHz fs.
Thanks! :)p7
EDIT: NVM. I think it still has quite a bit of aliasing. I'll see what I can do to get the 96 kHz recording working.
-
Yes, it still seems to be resampled at some point to 44.1 kHz.
-
Alright... how about this one (https://www.dropbox.com/s/bsn9wqasu3u3at2/sweep_HD558_150mVrms_2.flac)...
-
Alright... how about this one (https://www.dropbox.com/s/bsn9wqasu3u3at2/sweep_HD558_150mVrms_2.flac)...
Is it a digital loopback ? It does not have any distortion and the frequency response is perfectly flat below about 16 kHz, but there is still a brick wall lowpass filter at ~20 kHz, and noise shaping. I think I will use the first 150 mVrms file for now, but I have also created a 44.1/16 version (https://www.dropbox.com/s/xg0gx5gux5abvwi/sweep_44100Hz.flac) of the test file, maybe if it is already 44.1 kHz, then it will not be resampled by the operating system or drivers.
-
Yup. It was in impedance finding rig mode...
Here (https://www.dropbox.com/s/xp23dnmus7kpxko/sweep_44100Hz_HD558_90dBSPL_mic.flac)is the sweep recording of the 44.1/16 version through the headphones SPL (headphone characterization rig)
-
Here (https://www.dropbox.com/s/xp23dnmus7kpxko/sweep_44100Hz_HD558_90dBSPL_mic.flac)is the sweep recording of the 44.1/16 version through the headphones SPL (headphone characterization rig)
Unfortunately, it does not seem to have gone through any analog hardware (the distortion and frequency response is again too good for that), but it is not a bit perfect copy of my sample either. I think with the new settings you use since "sweep_HD558_150mVrms_2.flac", the recording is just a digital loopback through the audio driver.
Edit: I have a few graphs from sweep_HD558_150mVrms.flac:
Frequency response (http://img22.imageshack.us/img22/1144/8vsx.png) (I could have set a higher vertical resolution; the high frequency ripple and roll-off shows that there was probably some low quality sample rate conversion applied somewhere)
Crosstalk (http://img199.imageshack.us/img199/5007/9qj7.png)
Distortion vs. frequency (http://img10.imageshack.us/img10/2540/jax6.png) (only D2 and D3 are included, anything higher mostly just adds noise)
Distortion vs. level (http://img189.imageshack.us/img189/9094/tthq.png) (0 dBr corresponds to 0 dBFS in the original sample file; the 15 kHz result is not shown because of the playback/recording problems)
From the same sweeps it would also have been possible to generate these plots:
- THD vs. frequency with only one channel driven (it did not really differ from the both channels case)
- THD vs. frequency into the other channel when only one is driven (this did not happen in any significant amount)
- linearity (basically, gain vs. input level - it would be somewhat interesting with high distortion)
- distortion plots for specific harmonics
-
Yes. I tried the latest Audacity using wasapi and no go. This one here (https://www.dropbox.com/s/92lx760axolxskg/sweep_44100Hz_HD558_90dBSPL_mic_rightonly_2.flac) is only right channel through Windoze DirectSound, which is the best I can do for now, since I only have one mic currently. For REW I use ASIO.
The 150mVrms file that stv014 used for his results was probably done using the impedance test recording setup. Hopefully I didn't mess up this time.
-
Graphs created from sweep_44100Hz_HD558_90dBSPL_mic_rightonly_2.flac:
Frequency response (http://i39.tinypic.com/wjyvpv.png) (it shows that the bass response is slightly compressed at higher SPL)
THD vs. frequency (http://i44.tinypic.com/2qm1bwg.png) (includes D2 and D3, the -15 dB plot is dominated by noise)
THD vs. level (http://i40.tinypic.com/29c136x.png) at 60 Hz and 1 kHz (at low levels it is mostly just noise). 15 kHz is not shown because of the limited bandwidth.
Linearity at 60 Hz and 1 kHz (http://i43.tinypic.com/2zf2xxy.png) (this again shows how loud bass is compressed). The plots are normalized, so the absolute levels do not matter, and at 60 Hz and low levels noise is significant.
By the way, is 90 dB SPL the level corresponding to 0 dBFS in the test sample, or to -3 dBFS (the level of the first sweep) ?
-
I believe it was -3 dBFS, but not sure.
-
Adding calibrated HD558 Impedance curve. Ignore octave smoothing comment.
(http://www.changstar.com/index.php?action=dlattach;topic=65.0;attach=4259;image)
-
Sorry for resurrecting a very old thread...
I currently wrote 2 different articles around HD558:
1) The first one is dealing with HD558 in general + comparing old vs new pads (measurements) + examining pad burn-in effect (measurements): http://headaudio.weebly.com/clanky/ochozene-vs-nove-nausniky-recenze-sennheiser-hd558
2) The second one is focused on modding - I experimented and found out again that ring-shaped felt rules them all!:
http://headaudio.weebly.com/clanky/modifikujeme-sennheiser-hd558
Again, the site is not in english but all the graphs are fully english... And I will be happy to comment here in english if needed (or use Google translator). For now, I will post the most relevant graph I guess:
(http://headaudio.weebly.com/uploads/1/2/6/3/12635125/990525_orig.png)
Green curve is HD558 w/ HD598 mod and a few days old pads. White curve is w/ two felt rings inserted in between driver and plastic dust cover (see images inside the article for better understanding) while gray curve represents three felt rings inserted. Ignore 8.5 kHz dip on the graph and partially the one around 15 kHz as well.
HD558 w/ two felt rings really rocks in terms of neutrality... The headphone just felt neutral from the first second. It also destroys my other headphones here in terms of value (HD201, K712 Pro, HE500). HD558 w/ three felt rings sounds the most clean and grain-free even though it's already on the darker side of neutral which can be felt sometimes. I personally cannot decide which version is better... Both are significantly better than stock and both are the best in their way.
An interesting question to finish off: Driver burn-in vs earpad burn-in - how these two relate to each other, how big their impact is on your new headphones especially during the first days of listening? I would like to test a brand new headphone with a set of older but fully functional pads to get closer to answering this question (as I already tested a brand new set of pads with an old headphone)...
EDIT: One ring, again, does almost nothing... Four rings are impossible to fit inside and would theoretically cause the headphone to sound too dark. And I, again, tested other materials as well... But felt just rocks.
EDIT 2: Forgot to add - measured with SA-31SE (around 1 ohm output impedance).