Using the ideas of Linkwitz Lxmini with better loudspeaker drivers ?

[Tangband]

Making a baffle less speaker with better drivers seems to be very interesting - what about a Lxmini speaker with better drivers - like a Kef uniQ for the 300 - 20000 Hz and a purifi 8 inch woofer , or a Dayton 8 inch woofer from the reference series . It would then be a 3- way dsp speaker and much bigger than the original Lxmini . The tube for the bassdriver has to be about 20 cm in diametre.

I have been experimenting with putting a baffle around one of my condencer microphones - this made the recorded sound much more diffuse and unclear - why should there be any better result with a broad baffle in a loudspeaker ? - there is clearly a coloration depending on early reflections within 1-5 ms . I think Linkwitz had something very important going on in his latest constructions and he seemed to try to diminish the baffle size and instead correct the frequency response with the help of a dsp .

All you DIY:ers - is this a bad or good idea in a future loudspeaker building project ?

 

[mcdn]

The bass driver does run out of excursion fairly early. Finding a better driver for the mid/treble could be tricky. Maybe the 7nit from the Kali UNF? https://www.kaliaudio.com/in-unf

 

[ROOSKIE]

Linkwitz site is still live and all of his reasons for his designs are well documented.
He was very competent.
Building your own version will be tons of fun and Linkwitz's documentation should allow you to build something very nice.
Consider if you want dipole highs as KEFs tweeter is not open back.
I actually have drivers here and plans to do something reasonably similar to what you describe however everything is still on hold while I deal with Winter and other projects.

 

[Rick Sykora]

Forgive me if this seems candid, but it is my nature from decades of work experience...

Rather than how, suggest you list what your requirements are for your proposed speaker. This will help you determine whether an LX mini is a good pattern to use or not. Linkwitz had some specific goals in mind, not clear more expensive drivers would be much better. I own a comparable speaker in the Mirage OMD-5. On another extreme, I own Epique CBT24s. I often listen to my Amazon Echoes. They all project sound differently than conventional speakers but are hardly the same.

On the other hand, if you have time and money to burn on a learning experience, have at it. In any case, some more specific goals would help others understand how they might help.

 

[Tangband]

Linkwitz site is still live and all of his reasons for his designs are well documented.
He was very competent.
Building your own version will be tons of fun and Linkwitz's documentation should allow you to build something very nice.
Consider if you want dipole highs as KEFs tweeter is not open back.
I actually have drivers here and plans to do something reasonably similar to what you describe however everything is still on hold while I deal with Winter and other projects.

Looking forward to your building

 

[Tangband]

Forgive me if this seems candid, but it is my nature from decades of work experience...

Rather than how, suggest you list what your requirements are for your proposed speaker. This will help you determine whether an LX mini is a good pattern to use or not. Linkwitz had some specific goals in mind, not clear more expensive drivers would be much better. I own a comparable speaker in the Mirage OMD-5. On another extreme, I own Epique CBT24s. I often listen to my Amazon Echoes. They all project sound differently than conventional speakers but are hardly the same.

On the other hand, if you have time and money to burn on a learning experience, have at it. In any case, some more specific goals would help others understand how they might help.

Thanks for input.
This learning process is fun, but it might also be expensive. I would like to , at first, build the real lxmini to have a reference speaker and then try to make something different, with less distortion and higher spl with the same cardioid dispersion. 2 way or 3 way with active dsp crossovers.

Edit : maybe 6 dB higher spl ( for each speaker ) and 10 Hz lower bass than original lxmini, with the same cardioid dispersion.

 

[Tangband]

The bass driver does run out of excursion fairly early. Finding a better driver for the mid/treble could be tricky. Maybe the 7nit from the Kali UNF? https://www.kaliaudio.com/in-unf

Looks interesting !

The ultra cheap Peerless tc9fd18 seems to be superior to the seas driver . Much lower distortion at 1 kHz.

Peerless TC9FD18-08 | HiFiCompass

The THD measurements at voltage level 2.83 Volt were carried out with High Pass Filter "on" (2nd order Butterworth type, 80 Hz cutoff, HPF2-80) The THD measurements at voltage level 4 - 11.2 Volt were carried out with High Pass Filter "on" (2nd order Butterworth type, 150 Hz cutoff, HPF2-150)...

hificompass.com

Kef uniQ is also probably better, so are probably the Omnes cx3.1. Both of these must be complemented with a tweeter radiating backwards. The tube will have to be much bigger.
Tectonic have some interesting drivers also. Visaton B80 seems interesting, slightly better than tc9fd18.

 

[Tangband]

Another question is about using a 20 cm diametre tube for a 8 inch woofer in such a projects. The dispersion at 700 Hz will probably be slightly different than the original lxmini - but will the sound be worse or better ? One solution is maybe the use of a 350 Hz dsp equalised crossover point If crossed to a cardioid Kef uniQ ?

Beaming point for a 8 inch woofer :

Speed of sound) 34400/(Diameter of 8 inch cone)20.32 = 1673 Hz.
Beeming point in the seas woofer for original Lxmini would probably be around 3000 Hz ?

Original Lxmini uses a 12 dB/oct dsp crossover at 700 Hz, so with a 8 inch woofer , the crossoverpoint at 700 Hz might be to high because of the beaming starting around 1600 Hz.

 

[test1223]

Here is the review with spinorama of the LX Mini from Erin.

Linkwitz LXmini Speaker Review

Linkwitz LXmini Review

www.erinsaudiocorner.com

Original Lxmini uses a 12 dB/oct dsp crossover at 700 Hz, so with a 8 inch woofer , the crossoverpoint at 700 Hz might be to high because of the beaming starting around 1600 Hz.

Yes I would use even a smaller midwoofer due to the lower diffraction of the driver and tube and add another woofer from about 300Hz where the room transition begins. The Peerless GBS-85N25PR03-04 https://products.peerless-audio.com/transducer/73 or one of the newer Blisma M142A-6 might be a good mid-woofer since the reflections from the driver might be better due to the dome like cone.

Omnes cx3.1 is a very good driver it might be a good experiment to use two of these in a dipole configuration with an extra crossover of the backwards driver to adjust the room reflections.

 

[nowonas]

In all respect, in order to improve the LX Mini I think you need to study the concept in more detail and really understand what can be improved before you start your project. In order to fix a problem you must understand it.

The reason for the low crossoverpoint (700hz) is due to making the speaker cardioid and thus reducing room reflections and anomalies in the crossover region. This is probably one of the most important characteristics of this speaker, and by moving the crossover point to 1600 you have destroyed this capability in addition to have created big problems with vertical disperity and that the speaker will not act as a single source anymore. In my opinion, you are creating another speaker with bigger problems than you started with. The cardioid effect is created by mixing the positive phase from the woofer with the negative phase from the dipol. If you start calculating the dipol effects at 700hz you will also se that the crossover point and distance and arrangement between woofer and fullrange is selected very carefully by mr. Linkwitz.

Further, before changing the seas drivers I would also check the QMS and QTS of the Peerless drivers to see if they can be used without baffels. SEAS drivers are very high quality in addition to be very consistent to specs. Further, they are showing very honest measurement using industry standard measurements (50db) _without_ any smoothing. Peerless is using a 100(!) db scale for their frequency plots to make them look nicer than they really are, and the build quality and consistency is simply not the same as SEAS.

Hope this helps and good luck with you project!

 

[Tangband]

In all respect, in order to improve the LX Mini I think you need to study the concept in more detail and really understand what can be improved before you start your project. In order to fix a problem you must understand it.

The reason for the low crossoverpoint (700hz) is due to making the speaker cardioid and thus reducing room reflections and anomalies in the crossover region. This is probably one of the most important characteristics of this speaker, and by moving the crossover point to 1600 you have destroyed this capability in addition to have created big problems with vertical disperity and that the speaker will not act as a single source anymore. In my opinion, you are creating another speaker with bigger problems than you started with. The cardioid effect is created by mixing the positive phase from the woofer with the negative phase from the dipol. If you start calculating the dipol effects at 700hz you will also se that the crossover point and distance and arrangement between woofer and fullrange is selected very carefully by mr. Linkwitz.

Further, before changing the seas drivers I would also check the QMS and QTS of the Peerless drivers to see if they can be used without baffels. SEAS drivers are very high quality in addition to be very consistent to specs. Further, they are showing very honest measurement using industry standard measurements (50db) _without_ any smoothing. Peerless is using a 100(!) db scale for their frequency plots to make them look nicer than they really are, and the build quality and consistency is simply not the same as SEAS.

Hope this helps and good luck with you project!

You didnt read properly.
I never proposed a crossover at 1600 Hz . I was writing about beaming. I proposed a lower crossover, at 350 Hz instead, when using a 8 inch woofer.
It gonna be hard to improve a lxmini, ofcourse.
Linkwitz has already constructed a better speaker and how it would look - the lx 521.

 

[nowonas]

I see now that you proposed a lower crossover, so you are right, I had a wrong assumption about the crossover point. By moving the crossover to 350hz you would have great horizontal and vertical dispertion, But, I do not think that you would be able to create a cardioid pattern for the speaker as LXMini does. To create a cardioid pattern, it is necessary to have the right distance between woofer and fullrange driver at the crossover point in order to enable phase cancellation with the dipole. With a crossover point at 350 the distance would be much longer due to longer wavelength at lower frequencies. .

I am not so sure if Linkwitz believed that the LX 521 is a superior speaker, it is just different, with much higher requirements to the setup and room. In fact very othen he recommends the LX Mini over the 251 and Pluto due to this. ( see i.e https://www.linkwitzlab.com/LXmini/Introduction.htm )

By moving the crossover to 350hz, I would look at some of the FAST/WAV designs at DiyAudio site instead of modifying a LX Mini. I have heard a couple of them (using SEAS drivers), and they really are amongst the best systems I have ever heard.

 

[Short38]

I purchased the LX Mini plans and constructed a variant because of a congenital inability to follow instructions. Be that as it may, if you have not done so, be sure to purchase the plan set from the Linkwitz web site before starting a project based on Linkwitz concepts.

I ended up with an odd ball active 3 way. I love it but have not measured it. Picture attached. Morel mid and tweeter, Dynavox woofer in a vented box. dbx 3 way
Xover at 800 and 5000. Aged Emotiva upa 700 fed by OPPO 103.

 

[Timpman]

I know this is quite different from the original design of the LX minis but has anybody considered the same exact design but instead of a full range on top from 700 Hz, use a line array of flat panel drivers?

 

[Timpman]

I know this is quite different from the original design of the LX minis but has anybody considered the same exact design but instead of a full range on top from 700 Hz, use a line array of flat panel drivers?

Yes, I have designed a similar speaker to the LX Mini with a line array instead of the full range cone driver that Linkwitz uses. I am also nearing completion of a set of LX Minis that I am building just for fun so it will be interesting to compare the two.