Hello Erin,
Thank you very much for the shown measurements and for bringing Siegfried Linkwitz´ design to people’s attention.
The whole concept is indeed very different to conventional “box” speakers and the design goals -with respect to polar radiation pattern etc. are quite different. The usual textbook formulae (Floyd Toole etc.) is something that Linkwitz did not always fully agree with. His findings are well documented on the Linkwitzlab website, many AES papers and videos can be found there.
The design result of LXmini is an unrivaled presentation of sound stage and phantom imaging, as you already experienced.
Nevertheless, let me please add some remarks:
Your measurement shows a pronounced, ca 4dB increase in frequency response on axis between 2.5kHz and 4kHz. This is certainly audible and must sound unnatural. I wonder what is going on. Siegfried Linkwitz original outdoor gated measurement (graph is shown on the website under Linkwitzlab /LXmini/Design) did not show this “bump” in frequency response. More than 100 auditions we did were free of any comments about an unnatural sound or a boost in the 2,5-4kHz region.
Siegfried Linkwitz “The resulting acoustic frequency response on the full-range driver axis is designed to be flat +/-1.5 dB from 100 Hz to 20 kHz under free-field conditions. It is 3 dB down at 45 Hz and rolls off at 15 dB/oct rate below that.”
Maybe you are not listening to the original LXmini as per design, but to a deviation of it? Where does this difference originate from? Does it come from the electronic side or from the acoustic/speaker side? You may easily check if the electronic side (a corrupt filter setup in the MiniDSP unit). The filter curves of the MiniDSP may be visualized with a (free) app like REW or ARTA and then compared to the original Linkwitz design curves, that you find on pg.23 of the LXmini construction plans and on Linkwitzlab/LXmini/ASP.
I think it is worth sorting out that issue, as it can be the start of a very joyful and satisfying experience with LXmini (or even LXmini+2/LXstudio for increased bass punch).
Upgrading to analog signal processing (Nelson Pass ASP from DIYaudio store or Linkwitz Store) would be a further step up as it avoids these ADC/DAC conversions of the MiniDSP.
The other obvious measurement is the distortion “hump” at 1,4kHz. As mentioned by Davey, it was discussed earlier in the owner’s forum, called OPLUG, in 2017. Is it audible at domestic listening levels? Some posts from year 2017 feature this:
“It is evident that the 1.4 kHz distortion appears only at elevated SPLs and goes up very quickly with the output levels. It shouldn't matter much for listening at moderate SPLs, and loud playing is not an LXmini's forte anyway. That being said, getting rid of this distortion peak wouldn't hurt...”
How to get rid of that this distortion if you find it perceivable? One owner addressed this 2nd order distortion by applying three small self-adhesive felt pads. He measured the 2nd order harmonics in that 1,4kHz range to be lowered from ca 2.5% (-32dB) down to 0.6% (-44dB), but he didn't comment on an audibility of that measure.
Compression measurements: 102dB@1m are unpleasant and damage our hearing. Linkwitz LXmini is not designed for this. At 50Hz, the woofers could have even been driven beyond their mechanical limits. For loud party applications or for PA, there are a lot of other speakers around (horns etc.)
OTOH, we demoed LXmini/+2 to a 20+ people audience at Burning Amp Festival at Ft Mason/San Francisco. The room was larger than many living rooms (ca 60 sqm / 650sqft). LXmini+2 was able to reproduce at fully satisfying volume levels and dynamics.
Erin, I hope you find the time to sort this out and that Linkwitz LXmini is not just a “measure-and-put-away” speaker for you. Indeed, it can be the start of a very satisfying journey into Siegfried Linkwitz´ designs, that allow a live-like audio experience, that is so different from many “box-type” speakers.
Best Regards, Frank Brenner / Linkwitzlab