Still good distortion for a dynamic even at 100db. I bet it'd be fine even with an extra 10db. Someone got a rattly driver, it happens. If one's that worried about it then send it in for a replacement or refund, or start listening to music at sane volumes instead of bottom octave test tones at deafening levels.

Ultra, thanks for posting those measurements. Clearly visible just what I was missing with these and explains things much better then my analogy with our better half's :-).

Wanted to share my impressions of the K7XX. Full measurements to come, hopefully within the next couple days. I did have some experimental measurements floating around in another thread, if you want an early look.

The first thing I noticed is that the K7XX sounds both a bit warm and also a bit bright and metallic at the same time. I suppose in that sense, it has a bit of a U-shaped sound signature. It's also a bit grainy sounding, but I suspect this is mostly due to the upper frequency response characteristics more than anything else. Mostly clean and clear sounding, but just a touch grainy and dirty. Not the smoothest headphone I've heard by any means. However, it does do a fairly good job resolving small details. Not the best, but pretty good, though some elements are smeared together due to the treble issues I hear. Sometimes the extra warmth gets in the way a bit here too...just a tiny bit.

Now, going entirely off of faulty memory, the K702 65th AE sounded more "shouty" where as the K7XX just sounds brighter and grainier. They might also be more similar than I remember, so take that with a grain of salt. Either way, the K7XX does become somewhat irritating and fatiguing for me rather quickly with harsh music, especially something with lots of metallics/cymbals.

Thankfully, the broad bass emphasis and warmth does help cover up and balance the upper end out to an extent. I wouldn't be able to deal with it otherwise. It's not the cleanest, clearest sounding bass ever, nor is it the punchiest, but I do like the lower end balance overall.

In terms of soundstage, these sound rather wide, but not very deep in a front-to-back manner. Center imaging can be a bit lacking at times, too left and right sounding.

Comfort seems to be rather good. Very large pads, very little clamp, rather light, and the auto-adjusting headband works well. The pads are a bit stiff and cheap feeling, but nothing major. Build quality isn't half-bad, but some of the plastic used feels a bit cheap. It does not feel like it will fall apart or break easily, however.

All said, though, while these aren't entirely my thing, I think they're a really good deal at $200.

Thankfully, if your tastes are at all like mine, I've found using some of that 1/4" open-cell foam I've linked before in other threads works surprisingly well when placed in front of the driver. Cut it in a circle and just place it there. No need to take off the pads, as if it's cut to size, it'll hold in place due to friction. And the pads are large enough that the foam doesn't irritate the ears. This smooths out and attenuates the upper-mids and treble rather well. I actually quite enjoy them like this.

How loud? (I have no idea what 82% or 100% is in audacity or 3 O'clock on O2 without knowing output voltage of source). How do you know it's 20mW? Can you measure with a DMM with a steady state 30Hz on the headphones? Most importantly, are you testing or actually listening to music when the flub-flub happens? The K7XX driver isn't that big. 30Hz test tones tend to do a lot of screwy things to headphones with small drivers, e.g. Grado.

Your AT has a huge advantage in that the driver is sealed. Open-back headphones need to work harder (more excursion) to maintain same bass level.

Ppl says K series needs amp because they sound kind of dead from an iPhone and also because they are less efficient compared to Grado, AT, Denon, Fostex (dynamic) 'phones.

If the K7XX performs fine while listening to music, I wouldn't worry about it.

my K7xx runs fine at volumes significantly higher than what I would ordinarily listen to and with tracks having significant sub bass. That is good enough for me. No way am I running test tones at high volumes for the sake of science and seeing if I can eff up my perfectly good cans.  Sorry if some are having trouble in the context of listening to music at their own normal volumes if that, in fact, is the case for some.

1/4" in this particular case. You could probably double up on 1/8" too. Let me know if you need an exact link.