LXmini Clone

[01svtL]

I'm wondering how I can use these with my current setup. I am running an NAD t758(v1) as my pre, since I need room correction in my room. I have it going out, unbalanced, to an Acurus A150 stereo amp. I understand the miniDSP is needed as it acts as the crossover, so how would I integrate it into my setup? Would I need a second stereo amp, or could I, say, run from the NAD to the full range drivers and then run the woofers from the Acurus? Any idea if the NAD's front two channels are shut down with the power amp in play? How would the miniDSP connect to the NAD so I can still utilize room correction?

 

[NTK]

I'm wondering how I can use these with my current setup. I am running an NAD t758(v1) as my pre, since I need room correction in my room. I have it going out, unbalanced, to an Acurus A150 stereo amp. I understand the miniDSP is needed as it acts as the crossover, so how would I integrate it into my setup? Would I need a second stereo amp, or could I, say, run from the NAD to the full range drivers and then run the woofers from the Acurus? Any idea if the NAD's front two channels are shut down with the power amp in play? How would the miniDSP connect to the NAD so I can still utilize room correction?

Welcome to ASR.

You will need another stereo power amp to drive the full range drivers of the LXmini. Here is a connection diagram from Linkwitzlab [link]. Substitute your NAD's main L & R pre-outs for the Oppo Blu-ray disc player's analog outputs. The miniDSP will function only as the cross-over, and is transparent to the NAD. Room correction for the NAD will work normally (i.e. as if the NAD is driving L & R speakers through an external power-amp).

 

[01svtL]

Welcome to ASR.

You will need another stereo power amp to drive the full range drivers of the LXmini. Here is a connection diagram from Linkwitzlab [link]. Substitute your NAD's main L & R pre-outs for the Oppo Blu-ray disc player's analog outputs. The miniDSP will function only as the cross-over, and is transparent to the NAD. Room correction for the NAD will work normally (i.e. as if the NAD is driving L & R speakers through an external power-amp).

Excellent, thanks for the reply!

I'm currently also using an HSU ULS15-MK2. Think this would integrate just as well/the same as it does with my current speakers, assuming room correction gets it right? I had to run it 6 times, with slight adjustments in placement each time, to finally get it to blend with my mains to sound full range. I'm assuming it would just be a matter of the same with the LXmini's, correct?

 

[NTK]

Excellent, thanks for the reply!

I'm currently also using an HSU ULS15-MK2. Think this would integrate just as well/the same as it does with my current speakers, assuming room correction gets it right? I had to run it 6 times, with slight adjustments in placement each time, to finally get it to blend with my mains to sound full range. I'm assuming it would just be a matter of the same with the LXmini's, correct?

Yes.

 

[Jer]

You need DSP and measuring equipment. Without it, you don't know what you're up to and you can get the wrong impression.

Yup- have my umik1 now to help EQ.
It’s definitely interesting.
Loving the dispersion of sound and how low they go with the bigger driver.

 

[diyhobby]

Yup- have my umik1 now to help EQ.
It’s definitely interesting.
Loving the dispersion of sound and how low they go with the bigger driver.

Visaton Fontanella also has a similar Kit 2 way Omni available

 

[Jokerbre]

Updated XO and measurements.

Quasi anechoic measurements. Gate 250hz. Smoothing 1/24.

Reverse polarity.

 

[laserluxxer]

Cool, i build the "same" speakers like 3 years ago when TC9FD's could be had for under 10€.
These really are one of the best price to performance speakers.
With EqualizerAPO and a 20€ AV-Receiver one can get a great sounding small system for under 150€.

My meassurements also roughly coincide with yours.
If these speakers are used in front of a wall i would recommend basotect on the rear wall beacause the rear response does not look as smooth.
Free standing they are just awesome.

 

[Jokerbre]

Mid-bass is Mivoc?

 

[laserluxxer]

yes
Right now im printing these parts:
https://www.reddit.com/r/diyaudio/comments/13299zb to give it a cleaner look. I modified the part where the woofer is screwed to in 3d cad and it even fits the mivoc.

 

[olieb]

Updated XO and measurements.

View attachment 278992

Very nice!
FR and in room response look really good. Great work!

What XO do you use? LR4? Linkwitz uses 2nd order to have more overlap. The idea behind it is to combine monopole and dipole behavior to get a cardioidish result. But this means higher excursion for the full range. Contour plot in horizontal plane look very similar.
The original LXmini has -25dB THD around 1.5kHz and 86 dB as measured by Erin!

Linkwitz LXmini Speaker Review

Linkwitz LXmini Review

www.erinsaudiocorner.com

I do like the rubber coupler construction of the original though. Very little vibrations from mid bass or full range drivers going into the big pipe.

 

[Jokerbre]

Xo LR2. You can see in first post all DSP settings. Settings are change a little over time but i will publish final DSP data for all folks. THD is quite narrow, 2nd harmonic, i wound not bother with that.
Tube is made from inner tube and outer tube, between is filled with quartz sand.

 

[Jokerbre]

FR and in room response look really good.

Measurements are not in room they are quasi anechoic.

 

[olieb]

Measurements are not in room they are quasi anechoic.

Sorry I did not write what I meant. I was trying to reference the "predicted-in-room-response" (in a "good room") that is normally rather close to ER and similar (though "scaled") to power response.
My fault.

Xo LR2. You can see in first post all DSP settings.

I could not find XO specification in the first (or other) posts only EQ of drivers. Not all filters shown in the pictures.
Thank you.

Tube is made from inner tube and outer tube, between is filled with quartz sand.

Nice. That is another way to solve the vibration problem.

 

[MrHifiTunes]

As proof of the concept, I decided to make a clone of the LXmini speaker.
The results are excellent both audible and measurable.
The drivers I used are Mivoc swm 68 and TC9FD18-08 which in my opinion is a far better driver than Seas FU10RB (H1600-04).
I ordered the Seas L16RN-SL (H1480) waiting to arrive to make the final version.
If someone is an expert in designing parts for a 3D printer, they might be able to contribute to the design by designing the upper part that comes on the tube.
Off axis measurements are 0-22-45.
The microphone I use is Umik-1. It has a horrible feature to bring down the response over 18khz so I don't know the real situation above.
Crossover is terrible minidsp 2x4.

Why you put low shelf of 13db on the full range driver? Seems a lot. Both drives don't see baffle so I would expect the woofer puts out 80db/2.83v same for the full range except for the highest frequencies which see the driverfront as baffle.
I would expect attenuation of the highest frequencies.

 

[olieb]

Why you put low shelf of 13db on the full range driver?

This accounts for the dipole loss of the baffle-less full range when wavelength is increased and front and back wave cancel more and more.
More details here: https://www.linkwitzlab.com/models.htm
In particular headings A3 and D.

 

[MrHifiTunes]

This accounts for the dipole loss of the baffle-less full range when wavelength is increased and front and back wave cancel more and more.
More details here: https://www.linkwitzlab.com/models.htm
In particular headings A3 and D.

Still 13db sounds like a lot. In the link you send linkwitz use around 4-8db. I assume that for the full range in this design, you should look at the b=2a curve (7db). High boost will bottom out the driver I guess. Specially if you have an xmax of 3-4mm. Also not clear why you should extended it all the way down till 10hz. Just trying to understand and learn this concept

 

[laserluxxer]

Not OP but i built the same speaker.
I looked up my DSP settings and i also use a 14db low-shelf.
The idea behind it is to first equalize the response to be flat and then use a highpass filter.
The dipole rolloff can easily compensated with with a low shelf and that way you get perfect filter slopes.

 

[olieb]

Still 13db sounds like a lot. In the link you send linkwitz use around 4-8db. I assume that for the full range in this design, you should look at the b=2a curve (7db). High boost will bottom out the driver I guess. Specially if you have an xmax of 3-4mm. Also not clear why you should extended it all the way down till 10hz. Just trying to understand and learn this concept

The "baffle" of the FU10 is 98mm ø and the "piston" is 70 mm, so it is more like b=1.4a.
[EDIT: For the TC9FD the geometry is a bit different but the relation is similar.]
I did not find the 4-8db that Linkwitz is supposed to use, but the use case is somewhat different in his text anyway as he uses a 8" as mid.
The filter is supposed to linearize the FR of the full range so a high pass LR2 (≈700 Hz) can work as designed. There is no relevant signal reaching the driver in the 10 Hz range.
How high the shelf is chosen depends on how far into low frequency (beyond crossover frequency) you want to linearize - before applying the high pass.

 

[MrHifiTunes]

The "baffle" of the FU10 is 98mm ø and the "piston" is 70 mm, so it is more like b=1.4a.
[EDIT: For the TC9FD the geometry is a bit different but the relation is similar.]
I did not find the 4-8db that Linkwitz is supposed to use, but the use case is somewhat different in his text anyway as he uses a 8" as mid.
The filter is supposed to linearize the FR of the full range so a high pass LR2 (≈700 Hz) can work as designed. There is no relevant signal reaching the driver in the 10 Hz range.
How high the shelf is chosen depends on how far into low frequency (beyond crossover frequency) you want to linearize - before applying the high pass.

How loud can you play before you reach Xmax?
If the idea is to linearize the FR from the full range driver then assume the woofer isn't part of the equation right????
I can help it but 13-14 boost seems a lot for me.