CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS
Lobby => Headphone Measurements => Topic started by: ultrabike on November 04, 2013, 04:37:27 PM
-
Measurements form the SD Coronado Meet:
Frequency Response (no smoothing)
(http://www.changstar.com/index.php?action=dlattach;topic=1205.0;attach=4387;image)
CSD right
(http://www.changstar.com/index.php?action=dlattach;topic=1205.0;attach=4389;image)
CSD left
(http://www.changstar.com/index.php?action=dlattach;topic=1205.0;attach=4391;image)
Seal with these was difficult. Someone helped pushing them slightly, but perhaps needed more pressure against the pads.
-
Standard Lambda pro I presume?
Thanks for posting!
-
I think it was a Lambda Pro indeed. Seems this is Darinf's Lambda SR-407. Seems similar to the SR-404 (http://www.changstar.com/index.php/topic,352.0.html (http://www.changstar.com/index.php/topic,352.0.html))
BTW I think I know why the bass rolls off early with this one and the SR-009. I measured them as I do most of the open cans... With a lossy baffle. Closed cans I measure with a rigid baffle and some damping inside. This makes a big difference in the bass region (not so much in the rest of the spectrum).
Given the Stax use sealed pads I probably should have measured them as closed cans. Unfortunately, I don't have them Stax with me to measure again.
-
That's a tasty 100-1000.
-
What's the speed of that impulse response? The 'back wall' section such as at 3 kHz to 6 kHz? All I can see is it's less than 1.34ms.
-
That's a tasty 100-1000.
Yup. Ignore the < 100 response given strong dependency on seal.
What's the speed of that impulse response? The 'back wall' section such as at 3 kHz to 6 kHz? All I can see is it's less than 1.34ms.
I think it's a pretty clean CSD. Measurement sampling rate was set to 44.1 kHz. If I get the chance to measure again, will try 96 kHz.
I heard these ones with some strings playing. Pretty immediate, but still a tad little bright IMO. Could be the recording. Great detail, but difficult to get the whole picture in noisy meet conditions.
-
Same amp used as in your SR-009 measurements?
-
Yup. SRM-252. There was an SRM323A but didn't use that one.
-
Ultrabike, could you explain why the treble is so low? Is it because of Stax using the FF curve and you using a different one? Do you still have the raw measurements?
Thanks
-
Unfortunately I don't trust these measurements that much as I do the ones for the SR-307 since I didn't use a sealed baffle. Measurements are actually raw. No compensation was used, and this is the case for all the measurements I have so far.
Can't talk much about the treble on this particular cans since I didn't give them as much time as I gave the SR-307. However, notice the depression (similar to the SR-307 (http://www.changstar.com/index.php/topic,1820.msg49360.html#msg49360)) around 2 to 5.5 kHz followed by peaks in the 5.5 to 11 kHz area. My best guess is that such a depression in the high mids combined with peaks in the treble area are responsible for the somewhat bright perception of these cans (and perhaps it's etherealness).
Going by memory, I didn't find these (and the 307) painfully bright as I did the DT-990. Nor did I find them analytical. But there was a sparkle on them.
-
Is there a thread explaining or showing your measurement apparatus? I've never seen a measurement technique that produces a 'flat' raw ideal for headphones.
From what I understand of human hearing and the recording process, surely a dip of >10dB around 2-5.5khz would cause a serious problem with diction in vocals (particularly "K", "t", "Ch" and "ss"). I've never had that problem with Stax (except an underpowered 007 MK2) - if anything, I've had the opposite. Granted, this could be a new change from the older lambdas - which are the only lambdas I have used.
Thanks.
-
Is there a thread explaining or showing your measurement apparatus? I've never seen a measurement technique that produces a 'flat' raw ideal for headphones.
Here is a pic (the one with the Sephora box):
(http://cdn.head-fi.org/f/f6/900x900px-LL-f6f1a34f_HDF0024_9744CC_blog.jpeg)
It produces these:
http://www.changstar.com/index.php/topic,987.msg35940.html#msg35940
IMO the key is to use materials that minimize internal reflections, and to be careful about seal when headphones require it. Note the rig does not employ an ear simulator (I only use that for IEMs), therefore compensation is less of a requirement, and measurements tend to be more repeatable in the higher frequencies.
From what I understand of human hearing and the recording process, surely a dip of >10dB around 2-5.5khz would cause a serious problem with diction in vocals (particularly "K", "t", "Ch" and "ss"). I've never had that problem with Stax (except an underpowered 007 MK2) - if anything, I've had the opposite. Granted, this could be a new change from the older lambdas - which are the only lambdas I have used.
Thanks.
I actually felt some vocals seemed behind some instruments with the 307s (again, spent more time with those). I don't necessarily think one would experience sibilance and similar issues. Note 2k and bellow are actually above or close to the levels in the treble region...
You are more than welcome to give more credit to these measurements of the perhaps somewhat similar 507 if you like:
http://www.innerfidelity.com/images/StaxSR507SE11049.pdf
They also show a 10 dB drop from 2 to about 8 kHz and which are probably closer to my 307 measurements (minus the null around 4.5k which I think is a measurement artifact - perhaps a reflection that the foam material inside the measurement was not able to properly absorb).
Futhermore, though perhaps there are certain similarities, I don't feel the 007 MK2 sounds or measures the same as the Lambdas (old or new), under or over powered, so I wouldn't group individual headphone issues or assets with all things Stax. If you are comparing to the old Lambdas, I guess I wouldn't know for certain (I think I heard a pair at THE Show some time ago). However, here are some impressions and measurements of the older SR404s which seem kind of similar what I got with these and the 307s:
http://www.changstar.com/index.php/topic,352.msg5581.html#msg5581
More measurements of the SR404s here:
http://www.innerfidelity.com/images/StaxSR404S2742.pdf
-
Thanks for the image. Regarding Tyll's measurements - I don't particularly like them, but I do like his raw FR graphs. In that case there, it matches more closely what I would expect for a lambda (insofar as something that is treble shy of an ID curve, since it follows the FF curve more closely).
From what I can gather about your rig, some part of the mechanism is compensating the value on your behalf, most likely by design. Otherwise every headphone would be reading earbleeding treble (as is my understanding).
Thanks for the response.
-
No worries. I think measurements should at this point be interpreted in relative terms. If it helps, I measured these cans as I measure every can that requires seal...