CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS
Lobby => Headphone Measurements => Topic started by: Marvey on December 01, 2011, 12:51:32 AM
-
STAX SR-009 Frequency Response
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=1621;image)
-
It is quite amazing how well the Paradox T50RPs measure compared to something as expensive as the SR-009.
-
It is quite amazing how well the Paradox T50RPs measure compared to something as expensive as the SR-009.
Have you had a chance to hear them - the Paradox?
-
It is quite amazing how well the Paradox T50RPs measure compared to something as expensive as the SR-009.
Have you had a chance to hear them - the Paradox?
They are due to be delivered today!!!
I received my pair of Paradox's from Luis yesterday and posted my initial comments at head-fi (post #5321): http://www.head-fi.org/t/452404/just-listened-to-some-fostex-t50rps-today-wow/5220#post_7936071. (http://www.head-fi.org/t/452404/just-listened-to-some-fostex-t50rps-today-wow/5220#post_7936071)
Purrin / LFF: If you'd like, I can "re-enter" these initial opinions as a new topic under headphones or a "review" section if you create one.
-Tim
-
It is quite amazing how well the Paradox T50RPs measure compared to something as expensive as the SR-009.
I always get paranoid. To me, the SR-009 sound much better. Could be due to the open design vs. the closed and sealed design of the Paradox....who knows. ???
Purrin has heard them btw....
-
Hey marv,
I am considering giving a second shot to acoustic simulation of the sr009. Your measurement on baffle plate would be of tremendous help to calibrate the model. Would you be keen to join the project?
That means I need to know your dirty little secrets though (equalization curved, test rig acoustic conditions). Obviously, since you want to keep the secret sauce to yourself, I would present simulation results with equalization applied and not divulge the baffling conditions.
Care to help?
Arnaud
-
Adding measurements taken at the SD Coronado Meet.
Frequency Response:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4379;image)
CSD right:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4367;image)
CSD left:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4369;image)
Relatively noisy meet conditions not optimal for distortion measurements.
These are no bass monsters. Relatively clean CSDs.
Amplifier was SRM-252. I believe Marv used a DIY-T2.
-
I'm certainly not going to argue about the 009 not being a bass monster, but this graph seems a bit far from what I'm hearing...
-
Could be seal Ship. I've heard some people actually use rubber bands to press the Stax against the pads. Dunno.
-
Are you going to post the Lambda measurements?
-
Ya I'm certain your measurements are wrong. This makes it look like the 009's have less treble than the LCD2.1
-
Are you going to post the Lambda measurements?
I'll post the Lambda's in a minute.
Ya I'm certain your measurements are wrong. This makes it look like the 009's have less treble than the LCD2.1
Possible. From mids to treble they don't look that different to me from the IF results though:
http://www.innerfidelity.com/images/StaxSR009.pdf (http://www.innerfidelity.com/images/StaxSR009.pdf)
http://www.innerfidelity.com/images/StaxSR009SZ91278.pdf (http://www.innerfidelity.com/images/StaxSR009SZ91278.pdf)
http://www.innerfidelity.com/images/StaxSR009SZ91278afterburnin.pdf (http://www.innerfidelity.com/images/StaxSR009SZ91278afterburnin.pdf)
A rigid plate measurement would extend the bass, but if I do that to all open headphones they would appear as bass monsters. Probably should have measured with rigid plate given seal is critical to these as with closed headphones.
I think the seal problem is possible. For the Lambdas someone pushing slightly the cans to the pads (EDIT: those probably should have been measured with rigid plate as well).
Here are the results for the FR with no smoothing:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4393;image)
Unfortunately, I don't have current access to electrostatics cans/amps right now to attempt another set of measurements.
-
The fact that these use sealed pads and seem to require good seal to get the bass out suggests that I should have used the rigid baffle approach I use for closed cans.
-
So the IR at 5kHz is around 0.5ms, that's pretty fast right?
-
Well, to me CSD cleanness means there are not phase and ringing weirdness going on.
I usually think of speed in the context of BW. A very short impulse response would correspond to fairly large BW. However, all of the SD meet measurements I did were sampled at 44.1 kHz.
-
Could be seal Ship. I've heard some people actually use rubber bands to press the Stax against the pads. Dunno.
The gimbals holding the cups of my 009 are beginning to disintegrate and the cable's been cutting out. I've been thinking about sourcing a whole new frame for the cups which would allow for increased clamping pressure. After all it'll be a hell of a lot cheaper than sending it in. I don't want to play around much with damping, the only problem I have with the tonality is that lower bass dip. Maybe a different bias voltage might also improve things.
-
Seems good seal (and perhaps clamping pressure - less air gap) with electrostats is important to get the bass (and sub-bass) out of the depths... In fact, I think all of them use sealed pads, which is why I think I should have measured them as closed cans facepalm.
Not the 009s but this dude (http://www.innerfidelity.com/content/stax-srs-4040-ii-bump-problem) had a similar problem with the SR-404s, and rubber bands seemed to have worked for him.
-
Adding measurements taken at the SD Coronado Meet.
Frequency Response:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4379;image)
CSD right:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4367;image)
CSD left:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4369;image)
Relatively noisy meet conditions not optimal for distortion measurements.
These are no bass monsters. Relatively clean CSDs.
Amplifier was SRM-252. I believe Marv used a DIY-T2.
Holy cow! That's some amazingly fast decay in the treble there.
-
I'm certainly not going to argue about the 009 not being a bass monster, but this graph seems a bit far from what I'm hearing...
Chris/Ship:
20 dB down at 40Hz? I should hope so. Did not sound like that to me with Justin's pair on the BHSE at last July's LA meet. Or was that contributing to how floored I was by the inner detail further up? Psychoacoustic effect?
-
Nah, proly the fact I measured these as classic open cans... I would give credit to these measurements only 'bout 100 Hz up for now.
-
N3rdling came through and provided his SR-009, modded SRM-727 and SRM-T1!
Awesomeness Milos!!! :money: :money: :money:
This time around, cans were measured as I usually do closed cup designs that require seal. I believe Arnaud discussed the need for this with electrostats here (http://www.head-fi.org/t/498292/my-diy-electrostatic-headphones/585#post_8738868). Marv pointed out that the electrostats might have certain resonance issues. Seems the free air resonance of the diaphragm might be 100 Hz from Arnaud's discussion, which may explain the roll off I was getting at that frequency when measuring with sound absorbing, but leaky foam coupling around the pads (not exactly free air tho).
I listened to a few songs with these cans with the SRM-727 and did not hear a 100 or 80 Hz bass roll off. I liked quite a bit what I heard from them. Detail and cleanness is the greatest asset of these cans IMO.
So here are some measurements with the SRM-727:
Frequency Response
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4529;image)
Distortion right
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4531;image)
Distortion left
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4533;image)
CSD right
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4539;image)
CSD left
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4541;image)
Notes: I'm not certain the notch around 5 and 6 kHz is a headphone or coupling related. It does not show on all closed headphone measurements however. It seems to show up in IF measurements of possibly another 009 set too, though perhaps not as deep.
-
Measurements with SRM-T1:
Frequency Response:
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4543;image)
Distortion Right
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4545;image)
Distortion Left
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4547;image)
CSD Right
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4549;image)
CSD Left
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4551;image)
-
Comparison of channels using modded 727 and T1:
Right channel (727 red, T1 yellow)
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4553;image)
Left channel (727 green, T1 blue)
(http://www.changstar.com/index.php?action=dlattach;topic=23.0;attach=4555;image)
-
Ultrabike's latest measurements pretty much gels with how I hear the SR009 with a good seal: Middle-midrange emphasis with a slight downward slope toward the extremes (both treble and bass) with sub-bass rolloff.
Distortion is amazingly low - as we have all suspected.
-
Marv,
Were your original measurements in this thread done with the T2?
Chris
-
Ultrabike's latest measurements pretty much gels with how I hear the SR009 with a good seal: Middle-midrange emphasis with a slight downward slope toward the extremes (both treble and bass) with sub-bass rolloff.
Distortion is amazingly low - as we have all suspected.
Hm, but how does it retain a lot of airiness despite the upper treble roll-off? I recall you mentioned something about extended linear upper treble (10-20kHz) behaviour as almost a prerequisite for airiness.
-
I agree with Marv. While I felt these retained both airiness and bass response, I also felt they were somewhat deemphasized relative to the mids, which may be characteristic of many electrostats.
IMO, these don't have the airiness and soundstage of an HD800, or the bass response of some of the planars. There might be some mids coloration there. However, they are pretty high fi across the whole spectrum.
-
IMO, these don't have the airiness and soundstage of an HD800,
I agree, they are never as open or airy as the HD800, but at times I am very impressed with the imaging they can pull of. The fact that they can pull off the stat disappearing act is different than airiness.
I think of air starting a little further up than 10k, like somewhere well past 12k.
-
My original measurements were done with the SRM323, and obviously I did not get as good a seal. (Hence greater rolloff and bass mid-bass bump). My measurement rig and methodology were still a work in progress back then.
As far as the airiness factor, I suspect (haven't gone thru all the data), that UB's treble measurements tend to be a few db lower than mine. Even then, "air", i.e. > 12kHz measurements tend to be untrustworthy and very subject to placement factors. I tend to use my ears for > 12kHz than the measurements... after all, I don't hear the LCD3 as being an especially airy headphone despite a lot of measured energy in the last half octave.
-
Also, these measurements were done with 44.1 kHz sampling rate. If Milos and I meet again, or if I get a chance, I will measure again using 96 kHz.
As implied by Marv, coupling materials make a difference in the response probably depending on many factors like density, composition, thickness, and so forth.