• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is the LXmini by Siegfried Linkwitz the answer for home audio?

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,505
Likes
3,351
Location
Detroit, MI
I think they sound very good with studio music as long as you understand the inherent volume limitations.

I also use them as speakers for my TV and they do very well in that regard (good speech intelligibility, natural sounding).

I highly recommend adding two lower cost sealed subwoofers and eliminating the low end boost on the woofer as it will give you a bit more headroom and a lot more low end extension. Subwoofers are very easy to integrate as the woofer without boost has a 60 Hz LR2 roll off.

Michael
 

F-Bomb

New Member
Joined
Feb 11, 2021
Messages
3
Likes
0
Thanks for the comment, Michael.

I don't expect the LXminis to sound as big as floorstander speakers. But since I have a medium-large room (~50 m2), I was indeed looking at complementing them with subwoofers, under the form of a LXmini+2 system.

You suggest using lower cost sealed subwoofers. Is mainly due to the price point, or do you feel they sound better than the open-baffle subwoofers of the LXmini+2?
 

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,505
Likes
3,351
Location
Detroit, MI
50 m^2 is a pretty big room, not sure they would be the best fit for that to be honest.

I had considered the LXmini+2 but as you mention the Seas L26RO4Y subs are quite expensive ($330/each) and I personally didn't see much of a reason to pursue dipole subwoofers when the LXmini woofer is a monopole. I have not heard the LXmini+2 so I cannot comment beyond that. I also already had sealed subs on hand from previous projects (2 x Dayton RSS315HFA-8). The big advantage of the Seas L26RO4Y is that it allows an upgrade path to the LX521.

There is a lot of really good discussion on OPLUG about integrating subs with the LXminis, when I mentioned low cost sealed subs I had @dreite 's Peerless 830668 design in mind.

Michael
 
OP
A

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,439
Likes
7,948
Location
Brussels, Belgium
50 m^2 is a pretty big room, not sure they would be the best fit for that to be honest.

I had considered the LXmini+2 but as you mention the Seas L26RO4Y subs are quite expensive ($330/each) and I personally didn't see much of a reason to pursue dipole subwoofers when the LXmini woofer is a monopole. I have not heard the LXmini+2 so I cannot comment beyond that. I also already had sealed subs on hand from previous projects (2 x Dayton RSS315HFA-8). The big advantage of the Seas L26RO4Y is that it allows an upgrade path to the LX521.

There is a lot of really good discussion on OPLUG about integrating subs with the LXminis, when I mentioned low cost sealed subs I had @dreite 's Peerless 830668 design in mind.

Michael

Hey Michael, I head somewhere that the LX521 have an F3 of 30Hz, did anyone do measurements of the bass units in half space? I'm thinking of using the bass units (starting with the single driver one, sometime upgrade into the double driver ones) as satelite subwoofer for multi-subs.

I'm not sure how dipole and monopole subwoofers would blend in a room. Any idea how that would work out?
 

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,505
Likes
3,351
Location
Detroit, MI
I have not seen measurements for the LXsub2, there are ground plane measurements of the LXstudio (LXmini + LXsub4) on SL's website.

Probably a good question for OPLUG. To be honest I am not sure that there is any difference in DSP between the LXsub2 and LXsub4 which to me suggests the response should be somewhat similar.

No idea on how blending dipole / monopole subwoofers would work other than they will couple to the room differently.

Michael
 

wizardofoz

Member
Joined
Feb 16, 2019
Messages
41
Likes
34
Location
Singapore
We did do a listening test with the LXmini +2, LXstudio and LX521 and IIRC determined that if you were starting with the LXmini then +2 was good for a smaller room, but if you had a larger room and starting scratch the LX521 was where you should be going - the Studio really didn't come close to the LX521 in terms of capabilities in a bigger room. 50M^2 is a much bigger room than I have my LX521's in. We did the test in a room ~50M^s too.
 

Chromatischism

Major Contributor
Forum Donor
Joined
Jun 5, 2020
Messages
4,799
Likes
3,744
We did do a listening test with the LXmini +2, LXstudio and LX521 and IIRC determined that if you were starting with the LXmini then +2 was good for a smaller room, but if you had a larger room and starting scratch the LX521 was where you should be going - the Studio really didn't come close to the LX521 in terms of capabilities in a bigger room. 50M^2 is a much bigger room than I have my LX521's in. We did the test in a room ~50M^s too.
Or LXmini + your own more powerful subs.
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,632
Likes
6,226
Location
.de, DE, DEU
For a few analyses, I took the horizontal measurements from the LXmini and scanned them. I cannot exclude small errors during scanning, but on the whole the frequency responses should be correct. The 15° frequency response below 500Hz seems to have been affected by ambient noise, have treated this frequency range as 0°.
Unfortunately, there are no vertical measurements, which would certainly have been particularly interesting due to the driver positioning.

For all comparisons, especially with those based on NFS measurements (from Amir or Erin), please note that the LXmini measurements are heavily smoothed and therefore appear much smoother.

So, first the frequency responses from 0° to 90°, as they are also shown on Linkwitz's website.
Then the frequency responses are normalized to the on-axis frequency response.
1629901917639.png 1629901931246.png
Up to 1 kHz there are no conspicuous characteristics. It behaves as would be expected from a sealed speaker. The low influence of the cabinet has a positive effect; the radiation is very even up to 1 kHz.
When the Seas full range driver (crossover 900Hz) takes over, the radiation narrows down considerably. This is the transition from monopole to dipole.

Now let's look at the directivity of the speaker +-90° and estimate the horizontal constant coverage.
To do this, we look at the sonogram normalized to the on-axis FR and the -6dB limit:
1629904759928.png 1629904779215.png
Above 1 kHz, the radiation narrows considerably in several steps. If you consider the -6dB limit above 2kHz, you get a constant coverage of about +-40° (80°). It is therefore a narrow, rather than wide, radiating speaker. This is similar to the radiation behavior of a small horn.

Where's the magic in that?
It becomes "special" when you look at the directivity +-180°. For this purpose, we look at the horizontal normalized sonogram (+-180°)
1629908478886.png

Now you can see a narrow side lobe that forms behind the loudspeaker (look at the respective edges of the diagram at +180° and -180°) and that possesses appreciable sound pressure level in the range 1.2-5kHz.

If we look at power response and directivity index in the horizontal plane, the graph is choppy, but smooth on average. In part, erratic changes are included, which is actually not desired.
1629910202109.png


Update: But with the help of DSP, you should be able to compensate for some of this.
For comparison, with a "normal" speaker this would look something like this (only horizontal FR's too):
1629910802982.png


Seas FU10RB full range speaker used
I only know a review in Hobby-Hifi 2012-3 about the Seas FU10RB full range speaker.
The most striking thing is a resonance around 1.5 kHz, which can be seen in the impedance measurement and in the CSD in the review. Probably a cone or/and surround resonance.
This resonance causes a peak in harmonic distortion with 5% HD2 around 1.5kHz@90dB. This is unpleasant, but only affects a narrow frequency range.
The frequency response is linear for a full-range driver and has only minor breakup resonances in the high frequencies. Rather early sound pressure level drop in the high frequencies >10kHz.
 
Last edited:

MKW

Member
Joined
Jul 1, 2020
Messages
51
Likes
46
@ctrl thank you for those measurements. I'm in the camp of measurements are valuable, but ears/brain are the final word. I find that if something sounds good, it will measure good too, and vise-versa.

A little story for my perspective... lifelong (50'ish) audio enthusiast, started building speakers during high-school years, when I would relish visits to hifi shops and the inspiration those visits would generate. Fast-forward through car audio in my military years (and still), and as an employed adult with young children (minimal audio activity), to recent times where our children are now adults and a return to focus on my passions - one of which is audio. With a new house, came opportunity for a separate/new 2-ch system. HT is Denon AVR6400H running GoldenEar Tritons with Martin Logan Motion-series surround and in-ceiling Atmos.

Back to the 2-ch system... Originally had Martin Logan e-stats, but they took too much space, went on a tear of bookshelves, armed with REW and calibrated mic. During that investigation, I happened on the LX521 and it's "little brother", the LXmini - these look interesting, inexpensive, unique, small-footprint. Ordered a 3D printer (because every project deserves a new tool - young husbands take note), and the LXmini kit with flat-pack from Madisound. Built the speakers in about 4 days, including hardware store trips and painted finish. My first impressions - wow, these are amazing!

At the time, I was hunting Tekton Double Impact, and found a pair w/factory upgrade, grills & spikes - quick trip to Knoxville and they were mine (awesome speakers!). But next to LXmini... they sound like speakers - similar to everything else.
Here's a list of speakers that were directly A-B'd (gain-matched) against LXmini and sold because they don't play in the same league as LXmini; Tekton DI w/factory upgrade, Martin Logan e-stats, GoldenEar Triton 3+ (kept for HT), Buchardt S400, SVS UltraBS, Elac D2B6.2 (kept for garage), Martin Logan XT35 & LX16, various Polk. LXmini's bring the 'magic'. Closest sq competitor of the contenders were the ML panels, but even those didn't quite measure-up.

Tinkerer's gonna tinker... All above was done with the miniDSP 2x4 (came with the kit) and Emotiva 4-ch amp, but I've progressed through 2x4HD, 4x10HD, DDRC-88A/BM (for Dirac), and even miniDSP nanoDIGI (2X8 all digital) with a pair of stereo Topping D70's (to eliminate any possible deficiencies in miniDSP DAC's). All the DSP upgrades were done using an ATI (made) 6012 - the Emotiva was replaced due to high noise floor - couldn't stand that.

Current configuration... The manufactured sub (used to fill-in the low range) was a weak point, so I built a pair of Phoenix (alt) OB-subs (also a Linkwitz design) using the 12" Peerless drivers recommended by SL, which I found NOS (discontinued years ago). Side note: If you intend to build OB-subs, I can suggest other options that will work well. For the subs I built, the cabinet design was modified to accommodate a Hypex FA123 (preamp/ADC/Ncore 125+125+100 amps/DAC) 'plate amp' in each, and they are run in stereo (I do not subscribe to mono bass theory). With Xover and PEQ settings mapped to Hypex HFD, the pair of plates DSP and amplify all 6 channels (LXmini = 2 per side + Sub = 1 per side) for a very tidy system.

Someone might ask what feeds the plates - currently using NAD C658 (so I can play with Dirac) XLR to the plates. Also have NAD M12 (which I love, but no Dirac) coax SPDIF to plates - eliminates AD and DA conversion (which is always beneficial). Also still have DDRC-88A/BM (or 4x10HD) with ATI 12-ch if class AB strikes my fancy.
If someone is thinking of trying LXmini, I highly recommend it.
 
Last edited:

F-Bomb

New Member
Joined
Feb 11, 2021
Messages
3
Likes
0
For the subs I built, the cabinet design was modified to accommodate a Hypex FA123 (preamp/ADC/Ncore 125+125+100 amps/DAC) 'plate amp' in each
Nice ! Do you have pictures or plans you can share? I am curious about how you integrated the plate amps.
 

Hasse

Member
Joined
Feb 5, 2020
Messages
10
Likes
30
Probably better than Mr Linkwitz original design? Maybe….. but I would be surprised!
 

MKW

Member
Joined
Jul 1, 2020
Messages
51
Likes
46
Nice ! Do you have pictures or plans you can share? I am curious about how you integrated the plate amps.

LXmini built

20191110_172337.jpg


3D-printed driver struts
20191110_172442.jpg


Comparing speakers with Emotiva
20191110_182215.jpg


Later, comparing floor-standers. Martin Logan e-stats and LXmini have a very similar sound. If we are throwing-out subjectives, I'd call the ML "transparent" and the LXmini "holographic" - haha, figure that out from their marketing references. The Tektons are wonderful, but sound like speakers. Close your eyes and the Tektons put you in the front row, while LXmini put you on the stage - more intimate and emotional. ML are great too, but the sweet-spot is way too narrow and they want more space from the front wall. LXmini are typically placed 24" from front and side walls, side-wall boundaries are not balanced in the listening space - for LXmini, this is no problem due to the cardioid/omni dispersion.
20200611_072615.jpg


Using the Node2i connected directly to Hypex plate amps via coax SPDIF - this is a great and inexpensive 'lifestyle' solution. I'm currently using a NAD M12 (smoother tube-like sound) and/or NAD C658 (because it has Dirac - which I haven't played with yet).
20210410_213039.jpg


Linkwitz Labs Phoenix subs using the "Alt"(ernate) Peerless 830500 XLS drivers (2x12" in each cabinet) and modified to include Hypex FA123 with terminals so they can be serial or parallel and connected with either FA123 or standalone amp, and 4-way Speak-On connector for LXmini.
20210425_091133.jpg


20210425_091028.jpg


Notes about the subs:
- Plans, details and (most valuable - as always) Siegfried's comments are available free on Linkwitz Lab site Phoenix (alt) OB-Sub
- I up-sized dimensions (a tiny bit of height) to fit the FA123 plates. The most important aspect of OB-subs is the distance between front and back of cones (to prevent pressure-wave cancellation) - this aspect was not altered.
- I recommend building an enclosure for the plates on the back - I did not do this (slight mistake). I was thinking it would be better for cooling the plate amps if they were open and receiving air-blasts from cone movement. While that is correct (no problems cooling), what I underestimated was how many airways there are in the plates. I had to seal them all to prevent short-circuiting. I did that with; electrical tape over the vent slots (planned this), hot glue around every other opening (planned this), silicone caps on all connectors (I did not anticipate this). Subsequently ordered plugs (pictured above) from PartsExpress after removing hot-glue from the XLR connector locks Neutrik DummyPlugs
- A pair of these subs can be can be economically made from one sheet of MDF, see attached cut-sheet.
- They DO require DSP 'trickery' to make them work properly. You will have PEQ's to eliminate cabinet resonance at high freq. (near the crossover point) and boost (to extend the low freq. output).
- I used the specified Tymphany/Peerless XLS 830500 12" drivers (made in Denmark), which I was fortunate to find NOS reasonably priced from a private seller. They were discontinued many years ago (likely hard to find); however, I see the design as applicable to many drivers, and there are numerous available today that could accomplish the same (if not superior) performance. Find substitutes here or call Eminence (if you're in the US) - they are in KY and very good to work with (used in Spatial OB-speakers). Eminence website
- Another good (more costly) solution would be GR/Rythmik direct-servo OB-subs.
- Most difficult part of these builds (LXmini and subs) was the painted finish. There are many ways to prep MDF, I highly recommend using Shellac on all cut edges (thank me later). Yes, it is made from beetle poop (you weren't going to eat it anyway - right?). And if I had to do the finish over again, I'd probably use ExoHyde at PartsExpress

As you can see, and have experienced if you're a 'maker', everything doesn't always go as planned. That's part of the experience that makes it worthwhile. Let's face it... anyone can shop for great gear - there is no challenge, no victory, no sense of accomplishment in that. It's like guys who rev their bikes at the stoplight and ride fast in a straight - literally anyone can do that. Challenges and opportunities for greatness are found in the corners, where preparation, attitude, awareness, knowledge, skills and comfort-zones are tested - same as building speakers. And honestly, these are some of the easiest speaker builds I've ever done... and I still hate painting/sanding/painting/sanding...
 

Attachments

  • CutList Optimizer - 2021-02-27.pdf
    214.1 KB · Views: 102
  • 20191110_182347.jpg
    20191110_182347.jpg
    666.4 KB · Views: 150

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,632
Likes
6,226
Location
.de, DE, DEU
Probably better than Mr Linkwitz original design? Maybe….. but I would be surprised!

Of course, everyone has their own taste. Measurement-wise, I have strong doubts that the speaker delivers good values - maybe that's why measurements are missing.

In the presentation of the speaker it says:
Both drivers operate full range electrically and avoid the monopole to dipole null that would occur around a fixed crossover point. This helps maintain a symmetrical figure eight pattern associated with classic dipolar dispersion.

The Jordan full range driver has a, on average, straight frequency response but is already wavy from 1kHz upwards with max +-5dB.
At 90dB sound pressure level, HD2 and HD3 reach the 1% mark in the frequency range 2-4kHz (German Hobby-Hifi magazine, 2014-5).
The Xmax of +-5mm is okay for a 5'' driver, but quickly reaches its limits with full range dipole use.

The dipole use (high excursion of the full range driver) and the (comparatively) high harmonic distortion should lead to high IMD even at moderate volume - which is reduced in the LXmini by a lower excursion, because of the crossover.

Because the 8'' woofer is used as a full range monopole driver, the sound power frequency response of the Hestia speaker should be "special".

The reflections and the resulting interference due to the large surface area of the woofer cabinet are not likely to be insignificant.



One more small note about this statement in relation to the LXMini:
The result is (Acording to Linkwitz) a very wide dispersion speaker that radiates the entire room (except the wall behind the speakers),
In post#88 I evaluated the frequency response measurements made by Linkwitz.

The speaker radiates rather narrow and not wide - above 2kHz, the -6dB limit is just +-45°, which corresponds to a narrow radiation.

Unlike a monopole speaker, the LXMini radiates significantly more sound energy to the rear in the frequency range 1-5kHz (and not less).

Below 0.6-1kHz, the speaker radiates like a normal monopole due to the high crossover frequency of 800Hz and the flat filter.

This says nothing about the psychoacoustic sensation when listening to the speaker, the (subjective) impression can be different.
However, according to the measurements available so far, the measured reality is as described.
 
Last edited:
OP
A

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,439
Likes
7,948
Location
Brussels, Belgium
In post#88 I evaluated the frequency response measurements made by Linkwitz.

The speaker radiates rather narrow and not wide - above 2kHz, the -6dB limit is just +-45°, which corresponds to a narrow radiation.

Unlike a monopole speaker, the LXMini radiates significantly more sound energy to the rear in the frequency range 1-5kHz (and not less).

Below 0.6-1kHz, the speaker radiates like a normal monopole due to the high crossover frequency of 800Hz and the flat filter.

This says nothing about the psychoacoustic sensation when listening to the speaker, the (subjective) impression can be different.
However, according to the measurements available so far, the measured reality is as described.
The comment you replied to is a word for word quotation from his talk, i only added the words in parenthesis. Unfortunately he is no longer around for you to ask him yourself about what he exactly means with that comment.

 
Last edited:

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,632
Likes
6,226
Location
.de, DE, DEU
The comment you replied to is a word for word quotation from his website i only added the words in paranthesis, unfortunately he is no longer around for you to ask him yourself about what he exactly means with that comment.
He will always be one of the GOAT.
Perhaps Linkwitz is referring more to the subjective listening impression left by the loudspeaker.

In any case, he also says about the LXmini measurements:
At 180 degree in the rear, the speaker has a small bright spot between 1.2 kHz and 6 kHz.
...The resulting polar response of the LXmini is omni-directional at low frequencies.
...In the tweeter frequency range the radiation is dipolar, but with the rear radiation dispersed irregularly and gradually attenuated with increasing frequency
 

suttondesign

Addicted to Fun and Learning
Forum Donor
Joined
Feb 4, 2019
Messages
733
Likes
1,310
Location
Bellingham, WA
I can only tell you that while the response is not particularly smooth and requires a bit of personal tweaking with the dsp for a given room, the wide-open quality of the LXmini’s sound is pleasing beyond belief. It cannot play loud, esp with classical, so you need the LX521.4 for that.
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,632
Likes
6,226
Location
.de, DE, DEU
the wide-open quality of the LXmini’s sound is pleasing beyond belief
The reason for this could be the backward radiated sound, especially in the range 1-5kHz. The additional reflections could provide for a particularly "wide perceived radiation".

Just like some manufacturers place an additional tweeter on the back of a speaker to achieve a similar effect.
 

MaxBuck

Major Contributor
Forum Donor
Joined
May 22, 2021
Messages
1,544
Likes
2,203
Location
SoCal, Baby!
These undoubtedly sound terrific. Aesthetically they look like sanitary sewer cleanouts.
 

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,795
Location
Sweden
The reason for this could be the backward radiated sound, especially in the range 1-5kHz. The additional reflections could provide for a particularly "wide perceived radiation".

Just like some manufacturers place an additional tweeter on the back of a speaker to achieve a similar effect.
Is there anything unusual and genius in this dipolar dispersion between 1-5 kHz ,? - the backward radiated sound is out of phase compared to the front sound from the 4 inch driver , this should (?) take away some of the unwanted room-reflections directly from the side of each speaker ?

Im really interested to test this approach with my Seas er18rnx and a Peerless tc9fd driver , must dig out my dbx pa2 dsp crossover again….

Another question. - Im really wondering of the purpose of the 3 inch pipe ? Is it doing anything to the sound ? The backward radiation from the fullrange driver should show no big difference with or without the pipe ? Or are there some psychoacoustics involved in this ?
DC27A25A-AA43-4344-A8FE-9FCD0F207248.gif
 
Last edited:
Top Bottom