One Day Ticket for Stax Galaxy - The STAX Compendium

After a brief wait my SR-300LE arrived yesterday from Japan for $630 shipped.


I let them energize while I listened to my Airbow SC-1's before a quick comparison. It was easy to quickly switch back and forth due to the SRM-T1S's two Pro sockets.

Immediately the 300LE sounded a little louder, more upfront, and dare I say bassy, wow. I took a few measurements as well...

300LE



AIRBOW SC-1


300LE vs AIRBOW SC-1

For the life of me I couldn't get them to seal to my coupler, and they had no Stax fart either. As @sorrodje mentioned in his L700 measurements, it appears they behave like Lambdas do when they have a broken seal, and the leak is designed into them to bring up the midbass at the expense of low end extension. Not to mention the increased distortion in the low end compared to the SC-1.

I played around with a McDonalds straw and breaking the earpad seal when I had the SR-207/252 set, and ultimately let them go due to the grating treble. Once in a while when listening to the SC-1, I yearn for some MOAR BASS and break the seal for a few tracks, but don't intend to permanently mod them. The 300LE gives me that fun bass when I feel I want it, and can happily live beside the SC-1, and Lambda Pro in my Stax stable.

I haven't heard any of the newer Lambda's (L300, L500, L700) so I can't compare how these 300LE fair against them. From memory I like them more than the SR-207 (that unrefined treble, OUCH), SR-202 (lacked technicalities, maybe it's age had something to do with it, but the SC-1, and Lambda Pro I have kicked it's ass in my book). My head is fairly round and wide (bowling ball?) so the lack of mobility in conforming to some people's head doesn't affect me, but I could see how it could be a problem for folks with a narrow head.

 

Looks like a considerably rougher L700. I own the L700 and they are one of the smoother and more relaxed sounding lambdas in the treble that I've heard outside of the original normal bias. Also seems to drop off in the bass much faster with much more midbass hump. Perhaps the pads or clamp are to blame?

 

Impossible to compare my 300LE measurements to @sorrodje L700's, as they are two different measurement rigs. Could be the earpads, but I played with the clamp and positioning on the rig over and over, and almost all the results were identical. I'll admit I was shocked by the bass roll off myself. In the first post of this thread the difference between a good seal and a bad one is shown in a measurement. The thing is, the Stax Lambda I have encountered have all shown increased bass with a broken seal, these 300LE do not increase whatsoever when the seal is broken, supporting the earlier idea that the broken seal is built in. Maybe it has more broken seal than the L700, I don't know, and there is only one way to find out.

I would like to measure your L700 if possible, and re-measure the 300LE at the same time so conditions would be the same. Then we could have a legit A/B to debate. I also wouldn't call the 300LE FR rough by listening to them so far.

 

My curiosity got the best of me and I didn't feel like waiting around to try a L700 side by side with my 300LE, and more importantly to try pad swapping. So I ordered a set of L700 pads. Quite a bit more pricey than the old 507 lambskin pads at $120 shipped vs $70 shipped. I couldn't find a Japanese seller that could ship to the States for a reasonable price, so I was stuck with ordering from Stax USA's website which is run by Woo Audio. They still had to order them from Japan and the wait was approximately 10 days from time of order to delivery, which isn't too bad all things considered.

L700 pad on the Right and L300LE (L500) pad on the Left

The L700 pads are noticeably thicker and more plush. They were also a little harder to mount, and were a lot more snug than the stock pads. Better seal? My hopes were immediately raised, as I had tried taping around the perimeter of the stock pads to get a better seal, and it worked quite well, bringing low end extension measurements very close to the Lambda's of yore, but looked way to fugly. Some dude on HF also used some thin weather stripping to seal them, but I wanted to avoid going that route.

My initial measurement looked like the seal problem was fixed, the bass bump was gone and extended much lower then the stock pad. Then I mounted the old stock pad for a direct A/B... what? the bass bump was gone on these too, but didn't extend as well as the L700 pads, interesting. Did the stock pads get broken in more and become supple and seal better? They were stiff right out of the box. Or did the driver loosen up a little? I really don'y know, Regardless the L700 pads seal better and extend further down low.

Green L700 pad
Yellow L300LE pad


Are the 300LE the poorman's L700? I still don't know, and they are selling out in Japan, and US retailer's delivery dates have been pushed back significantly. Now with the plush L700 pads these really do feel like a Limited Edition, thanks Stax, I like 'em.

EDIT: I forgot to mention the Stax fart is back with the L700 pads

 

Just want to thank JK47 and others for posting at least superficial and even better measurements of the L300 and other headphones. They're very valuable. These combined with a headfi 300Limited Seller describing the 300Limited as more base heavy, V shaped than the l300 have recalibrated, steered me back to the l300.

 

Reviving this old thread to post some collected info and random musings on Normal Bias round driver models, which have become my obsession (and favourites) in the last couple of years. Very interested in collecting opinions / comments on those, of course and apologies if this sounds a bit like a long-winded rambling: it's the first time I try to organize thoughts I've only been discussing at length with @Kabeer so far.

My interest was sparked by the first pair of SR-5 I bought years ago: I immediately loved their mids so started exploring their siblings. Despite having most known variants of the SR-5N (white gold, black gold with cloth cable, black gold with rubber cable) I could never warm up entirely to their sound, so I left them alone and moved on to the SR-X lineup.

One of the first SR-X I bought came from Japan and was a strange animal: SR-Xmk3 frame, and SR-5 white cable. These quickly became my favourite, with all the good qualities I loved in the SR-5 and none of its defects. After some time accumulating SR-Xs and SR-5s and having set up a measuring rig I started cross-measuring examples, and transplanting SR-5 drivers into the SR-Xmk2 and mk3 frames from the bad examples with busted drives I got.

Measurements (I will paste some below) show that SR-5 drivers transplanted in SR-X frames (mk2 or mk3), are very closely matched in FR to original SR-Xmk3 drivers. And they do indeed look identical visually. Problem is, they still sound slightly different to me, with thicker mids in the SR-5 and a touch more brightness in the SR-X, while using the same frames.

So I turned to the data in the Japanese fc2 site and cross-referenced the diaphragm thickness reported there for all models, which differs from the one usually mentioned in forums, and even here (@sorrodje states 2um for the SR-5, which seems to be correct for the SR-5N based on data below):



The data above might explain my preference for SR-5 drivers and the SR-5 weightier mids / SR-X airier treble as a result of diaphragm thickness. An interesting trait also emerging from the table is driver technology trickling down from the top-line SR-X models to the more consumer-oriented SR-5 and SR-5N.

Here are some raw measurements on my flat plate rig of SR-X versions, and comparisions with SR-5 headphones and driver transplants in SR-X frames. At the end of this post I'll post a measurement of my HD600 so you can do the mental FR gymnastics to translate my rig's data.

First, SR-X revisions compared. mk3 is averaged between all the different mk3s I have (some with intact pleather pads, some down to cloth, some replacements as close as possible to the originals).
red=mk1 / green=mk2 / blue=mk3


Variations of my mk3s (different pads on each).


Now let's move on to the SR-5 and SR-5N and compare to blue mk3 above which is the middle of the three I have.
blue=mk3 / dark yellow=sr-5 / light yellow=sr-5n (gold/black, cloth cable)


Leaving aside the decrease in bass in the SR-5n, is the difference between srSR-5 and SR-Xmk3 due to the driver? housing? cable? Let's compare the SR-5 to its driver transplanted in an mk3 frame, and in an mk2 frame.
dark yellow=sr-5 / dark green=mk3 frame / light green=mk2 frame


The frame makes a huge difference in the upper mids and lower treble regions. Last measurements compare the two transplants above to an intact SR-Xmk3 with its own drivers.
green=as above / blue=sr-xmk3


Similar amount of variation we observed between SR-Xmk3 samples. Also, the smaller cup opening in the mk2 vs the mk3 does not seem to make a lot of difference in FR.

Going back to what I wrote at the beginning of this post, SR-5 and SR-Xmk3 drivers seem to behave identically in terms of FR once put in the same frame. Despite this, I still notice some differences which seem larger than what could be ascribed to sample variance, and kind of go hand in hand with the diaphragm thickness data derived from the fc2 site (and possibly other factors too, of course).

Phew, turned out a lot longer than anticipated. Here is the SR-Xmk3 shown in charts above compared to my HD-600, as a frame of reference for my rig's measurements.