• 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!

My waf dipole speaker design. Waffle.

JeyB

Member
Joined
Jul 8, 2022
Messages
21
Likes
19
Interesting! Glad it works out :)
No doubt this sounds great 3d wise, Sl experimented in the very same way you did.
Only thing I'd be worried about is hearing the drivers separately due to how close they are. Also, for adding another lx521 baffles into the system, I'm 100% sure SL himself said it makes no difference to using something like minis or what you already have, so there's that, probably he would know :)
Well, the proof is in the cake. Just place them in this way and listen, it's free. The 4 drivers sound at the upper baffles is perfectly integrated at that distance of 80 cm to each ear. In my experience this the minimum distance before they collapse as individual drivers.

Despite this ultra near field disposition, the sound is psychoacoustically projected at about 6 m away and even more, transcending the room limits in some recordings. The top baffles are not perceived as sound sources. When listening, if I look at them or move my head side to side it is impossible to perceive any sound coming from their physical location, to the point that they seem to be powered off. Some friend told me he thought that the sound (phantom scene) emanates from the entire room itself. In my setup the top baffles and LXsub4s are like sculptures, because it's not possible to think of them as transducers, since the sound is absolutely decorrelated from them.

On the other hand, the LXsub4s in this setup cancel every standing wave up to 100 Hz and is the most astounding thing I have ever heard in my life. True directional low frequencies that are only excited at the listening chair location with total room modes cancellation. There is no audible low frequency content in the rest of the room, only in the transversal line that crosses the longitudinal axis of the opposed LXsub4 bins. So much less low frequencies to the walls, floor and ceiling means happier neighbours! But don't think the instruments with low frequencies are located at 50 cm, no. They are also projected far away in the room, wherever the microphones placed them in the mix. Summing up, despite being completely surrounded myself by nearby drivers, the phantom scene is happening in front of me at a considerable distance, adding positively to the musical experience.

Look at this graph, the magenta line is a measurement taken to the LX521 with LXsub4s bins under the bridges at 2,2 m equilateral triangle setup. The yellow line was the measurement taken with the subs at 50 cm from my ears, the same setup I showed in the photographs enclosed in my last reply. The difference is so great that I call this setup LX521 881 (0,8 m distance from each top baffle to my ears and 1 m between top baffles). I think there is so much to still discover from the LX521...


Screenshot_20220929-032241_Gallery.jpg
 
Last edited:
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
I somehow managed to miss your former post! Sorry about that. What I'm referring to is this: http://noaudiophile.com/RMAF_2015_Day_3/

What I believe is actually happening is that lx521 is so good at what it does that it does it a bit too good. Yes, the voices sound actually really good, but something is a bit off, it's hard to express, probably because we are acustomed to hearing our own voices inside our rooms, and not really orchestras, so it bothers us a bit more when we hear voice in the phantom image very disconnected to the room itself. If that makes sense. :) just a theory. This would probably be remedied by your approach fully I believe.

Yes, my bins do the low end extension a bit better than lx521, and are smaller, but they do suffer from resonance appearing which is a relatively low q one and thankfully really easy to eq out. Lx subs are great, but they still do vibrate at about 50% of what they would without cancelling (which is still very nice). And yep, no need for a bridge :)

Also, yes, the ability to disconnect the top baffle and subs is actually true for all the speakers, not just dipoles (which probably make most use of it). Anything below Schroder would benefit from that, but it's not very practical.
 
Last edited:

Holmz

Major Contributor
Joined
Oct 3, 2021
Messages
2,108
Likes
1,283
Location
Australia
Out of curiosity ... I'd be interested in seeing a measurement with the Linkwitz top end swapped on to your new bass bin. Reason I'm curious is that Linkwitz spend some time optimizing his baffle shape. Duplicate his crossover points and dial them into your other unit.

^this^

One could probably try various wings attached to slowly flare the baffle, or felt wings, or star shaped wings to avoid a baffle step.
 
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
I spent quite a bit of time with the driver selection for the minimal possible baffle shape to match my goals, but, yes, I bet the lx top baffle would work great with the new bins. That being said, I actually perfer my baffle (take that with a bag of salt obviously) to the lx521, but the difference is sound is at least to me very small. Vocald sound a bit different, that's about it.

No baffle step corrections would be needed at all, just do a lr4 on the bins/low mid and that's it. There is no baffle step to speak of, as the bins are in fact dipoles, and at 100 Hz the wavelengts is 3.43m so plenty of room to integrate there (although, at such f I wouldn't even bother worrying about integration).

@JeyB
80cm for full integration with lx is great. It's just impractical for me unfortunatelly. There is little doubt that your setup is optimal.

Also, +1 for the "sculptures". Everyone notices that first. Speakers seem to be sculptures, sound comes from somewhere else. Amazing stuff.
 
Last edited:

JeyB

Member
Joined
Jul 8, 2022
Messages
21
Likes
19
I somehow managed to miss your former post! Sorry about that. What I'm referring to is this: http://noaudiophile.com/RMAF_2015_Day_3/

What I believe is actually happening is that lx521 is so good at what it does that it does it a bit too good. Yes, the voices sound actually really good, but something is a bit off, it's hard to express, probably because we are acustomed to hearing our own voices inside our rooms, and not really orchestras, so it bothers us a bit more when we hear voice in the phantom image very disconnected to the room itself. If that makes sense. :) just a theory. This would probably be remedied by your approach fully I believe.

Yes, my bins do the low end extension a bit better than lx521, and are smaller, but they do suffer from resonance appearing which is a relatively low q one and thankfully really easy to eq out. Lx subs are great, but they still do vibrate at about 50% of what they would without cancelling (which is still very nice). And yep, no need for a bridge :)

Also, yes, the ability to disconnect the top baffle and subs is actually true for all the speakers, not just dipoles (which probably make most use of it). Anything below Schroder would benefit from that, but it's not very practical.
I really loved your bass bins, so small and with full force cancelation. I bet they can displace a great spl without even a slight vibration. Not that the LXsub4s are bad in any way, they are also excellently designed and executed with state of the art drivers. How much is the cost of your subs relative to the LXsub4? Have you provisioned any subsonic filtering in your crossover? Subsonic content can do real damage to the Seas L26RO4Y drivers inadvertently for the user, so I usually play new music files at low volumes as SL suggested.

I fully agree with you regarding the great advantage of a woofer structure independent from mid/highs. The best location in the room for low frequencies rarely coincides with the best location for mid and high frequencies. I can't imagine myself owning a classical loudspeaker with woofers integrated in the main structure. Again a great solution from SL, often underrated.
 
Last edited:
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
I believe if you go my route it's about 200 usd/eur cheaper. Not that much. L26RO4Y is around 330 usd in europe, you need 4, and you need 8x SB23MFCL45-4 at around 140 usd.
 
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
Also, yep, I did put a high pass lr4 filter to help protect the subs.

I played them during testing to the point that I was very uncomfortable with the amount of bass in the air. My whole body was very uncomfortable. And they were perfectly fine. So I wouldn't worry about the drivers much, also, when you see them in person, they are as robust as it gets. SB really did a good job with these.
 
Last edited:

JeyB

Member
Joined
Jul 8, 2022
Messages
21
Likes
19
Also, yep, I did put a high pass lr4 filter to help protect the subs.

I played them during testing to the point that I was very uncomfortable with the amount of bass in the air. My whole body was very uncomfortable. And they were perfectly fine. So I wouldn't worry about the drivers much, also, when you see them in person, they are as robust as it gets. SB really did a good job with these.
Just for curiosity, when applying the HP LR4 did you notice any group delay effect? Sometimes visibly "awkward" measurements are completely unnoticed to the ear.

Regarding the voices issue you were explaining with the LX521, I have been playing quite a lot of voice recordings these past days just to check if I could perceive the same effect you talked about. Thereafter I can conclude that, in my own setup, there is no issue at all. But I have realized some interesting findings.

Usually, multichannel studio recordings play a great role regarding credibility of a real person singing in the listening room accompanied by other instruments. At the studio, the voice is generally recorded into a cabin booth, completely wall to ceiling padded. As a result the sound engineer can obtain a voice stream free of any reverberation and extremely detailed thanks to a very close microphone . This is great for control/editing purposes but very bad for plausability rendering. The main character of the voice is extremely straightforward and usually has some cues that detract from credibility when played on a good sound system. This is, when I listen to a real person talking in my living room he/she is about two meters away from me, his voice is surrounded, and at the same time being modulated, by the acoustical environment of my room, so as a result it's perceived with naturality by my ear-brain. If I bring that person to the toilette, his voice changes radically due to a very reflective new environment, let's say RT60 about 700 ms. But if I try to get abstracted of the environment and mentally place the toilette modulated voice into the living room space, then there will be great incongruencies that my brain would inmediately detect.

The cabin booth voice at the recording studio produces a very similar situation. Given the fact that it's extremely detailed to be comfortable/plausible in the complete mix in my opinion, it is then downmixed with instruments recorded at different spaces in the studio, like the drums for example. When the mix is edited, usually digital reverberation is added to try to compensate the dry voice recording and as a result the plausibility goal jumps through the window. If the singer has some technical shortcomings further agresive equalization is engaged (this is the worst it can be done imo and the sibilants responsible), the engineer adds some little autotuner here on that notes, a pick of compressor there. Then the producer (Mr. Money) imposes his own criteria to make a product that can compete in the market. Further compression, additional voice equalization, changing levels for all the band, after all it's his own money which is going to be in the air, but that is another story.

This is what I perceive in lots of recordings I have been playing along this week. I think it is not a problem with the LX521, on the contrary. Voices sound in the room but decorrelated from the instruments most of the time. Maybe is this what you and your friends perceive?

When listening to soprano/tenor arias accompanied by piano and recorded live at symphony halls, for example, the thing become much much more easy to the brain. Digital reverberation is not needed, autotuner is unnecesary, equalization is taken to a minimum, nice hall reverberation surrounds both singer and pianist and as result the experience becomes greatly benefited by true profesional singer&pianist, with great voice training, able to project the interpretation with sublime voice gesture, expression and dynamism.

Given a usual assistance to these musical events, such enough to give your brain a sound database written in your memory, is mostly beneficial to produce the magical trick that some sound equipment is able to do: to be teletransported to the real hall with the real singer and the real piano player. This is my experience here. I would add that after assisting to a live representation and within half an hour to getting to my home, then played some similar recording and being unable to distinguish between what I have listened live at the hall a few moments ago and what the LX521 is projecting in my living room. It's a so addictive experience that I can't encourage enough reccomendations to go to listen live music at symphony halls.

My advice is do not take seriously the opinion of those who are not used to listening to live music in concert halls.
 
Last edited:
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
Yes, you are absolutelly correct in that one should be cautious of such bias, especially audiophiles that listen only to "audiophile grade" speakers and not much live at concert halls. I must admit that I too am probably in this group. Several people told me this, but they only have typical box speakers and are probably acustomed to the sound, with all the colorations one expects. There is also the fact that music is made for very poor dispersion characteristics speakers, and when you play music on something more revealing it suddenly becomes much more apparent. I remember this effect when I got my stax headphones, they don't forgive.

Don't get me wrong, I adore lx521. What we're speaking about are the smallest of details, nitpicking about stuff that's probably quite pointless. The main goal of my speaker was to make it as slim as possible while keeping the characteristics of lx521 and possibly adding some benefits like force cancellation, ability to put them on wheels, etc. I took a conscious design choice to radiate more of vocal mids to the sides and see what happens. Well nothing much actually, and when I applied a different baffle (with wider sections) and a different EQ, I got very close to lx521 directivity wise. Sounded pretty much the same so I opted to keep it slim.
Makes sense that all of that with vocals might just be bias. They do sound a bit different and I do perfer them from my baffle, but that's just such a nuance I wouldn't say it's a necessary benefit over lx521 or something like that. From the lx521, at least in my case, vocals sound crisp, easily located, but a tiny bit tonally weird. As I said, quite possible that it's just bias, but it's definitelly different then on regular speakers, even good ones, for sure.
Basically I needed a more portable version of lx521 mostly, the rest is just toying with ideas. :)

lx521 top baffle on the waffle bins might in fact be a winner combo, but seriously in my eyes these things are quite interchangeable. Any properly designed, with directivity in mind, serious ob solution will sound great and similar. It's kinda the point.

Btw, thank you so much for taking the time to discuss and write! :)
 
Last edited:

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,731
Likes
2,922
Location
Finland
Very nice speaker, Ardacer!

Just a question about voicing, in previous post you say that vocals sound crisp. Previously you showed measurements, and this graph tells me why they sound crisp. I suppose this is one speaker near listening spot, long gating (500ms?) Upper mids are elevated 6dB...

otriGWt.png
 
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
Hi :)
thanks!
Nope, I said vocals sound crisp on my lx521. This image is a measurement of my speaker, 1m distance, position between upper mid and tweeter assembly, raw. The same positioning, gated measurement is the ruler flat one above it. I don't really sit and listen, I move around and listen, that's why there is a rta maximum of my room there too. Vocals sound great on both lx521 and waffle. Waffle gives subjectively a bit tonaly darker and pronounced vocals than more analytical ones of lx521. Both are good and crisp clear.
Subjectivelly I like mine better, but as I said it's just a preference, I really like lx521, that's why I own it too and won't be selling it.
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,319
Location
Ottawa
Thanks for posting your design Ardacer! I own PAP trio 15’S and am contemplating something like you have done.

Question…how important is the rear firing tweeter in your opinion? Does omitting it make a substantial difference?
 
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
I remember reading about how JohnK and SL debated about this very topic. John advocated for the use of the rear tweeter because of the general tonal energy balance in the room, SL was skeptical. You can still find it documented online. Aftermath was that SL decided to add a rear tweeter.
So I guess that settles that. I'd say it's not all-important, but it makes sense and isn't going to cost that much.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,731
Likes
2,922
Location
Finland
About the rear tweeter (or planar tweeter open back) - I remember reading about that debate too, and if we look at SL's home setting it is obvious that he wouldn't hear the difference. But in more normal, more reflective rooms it really helps to balance the sound, by levelling treble room response and decay response. It is not so important to get dipole pattern very high, but we just project sound energy backwards to be reflected back in the room as psychoacoustically late reflections.

Been there, done that - monopole > twin mono > planar

Ps. Thank you Ardacer for promoting the work of John Kreskowsky! His work for multiway dipoles is IMO as important as Linkwitz's, and fortunately his website is restored http://musicanddesign.speakerdesign.net/products.html An European college that I respect and thank very much is Rudolf Finke, and his site is still running but without active updates http://www.dipolplus.de/ It's in German, but you can find a very good paper in English at Downloads subpage. We must remember also a guy down under, Gainphile was first one using Minisdsp for EQ and xo, as well SLOB for bass! http://gainphile.blogspot.com/
 
Last edited:

JeyB

Member
Joined
Jul 8, 2022
Messages
21
Likes
19
Hi again Ben! I really appreciate your efforts to construct a new transducer employing the LX521 basic principles (OB) but adapted to your own necessities and tastes.

IMHO what you have accomplished it's not an easy task. You have also selected a completely different driver complement and optimized their polar response in the new Waffle top baffles and bass bins. This is something I couldn't do because I lack your technical expertise. So, how not to comment positively about your dedication and determination? You've also showed a lot of respect for SL and JohnK, two of the most advocates and wise-spreading OB designers of our time. So I agree with you again. Like @Juhazi also mentioned, Andi Gainphile is another great OB designer with full credit in my view (the Princeton Univ. fully validated his exemplary R16 true constant directivity loudspeaker).

I would also tag John Reekie from HifiZine.com, because after reading his "Prototyping a dipole bass system" article led me to empirically investigate about subwoofers positioning adjacently to the sweet spot in my actual LX521 room setup (with great success, by the way). John Kreskovsky also deeply wrote about this very same issue in his site. I've also enjoyed reading 2pi-online.de regarding comparissons between LX521 and his own omnidirectional designs, like Demokrit-T and Aristoteles and the great sonic similarities they all share. Also SL with his AES article "Room Reflections misunderstood?" touched the rear tweeter implementation in Orions. Summing up, lots of loudspeaker designers investigated in some or other way the optimum radiation pattern to accurately reproduce music in a home environment, so many that would be impossible to name everyone. But after reading much about OB designs implemented with electrodynamic drivers I would say that most of their findings converge to a common point more than diverge.

If my memory serves well, was SL who said that any OB design, even completely flawed ones, sounds nice and natural. This is the nature of the beast, I think. So I completelly falled in love with the LXmini, Studio and finally LX521. I talked to him in a few occasions via emails and aside being a very kind person he was also very conscious about what people may expect from a pair of loudspeakers. I remember very well that prior to recommend me to construct the LXmini, he advised me not to expect the typical commercial (box?) sound because I would be dissapointed in such case. He also inquired me about my preferred kind of music (acoustic or electric). And his advice proved to be right after time.

Most of us, since our childhood, have been passively indoctrinated in the box sound paradigm suffering its inabilities to evenly illuminate the listening room. So once listened to any OB design, every other conventional boxed design becomes futile in a home environment, no matter their complexity or cost, that's my humble opinion. Don't get me wrong, inside a studio acoustical environment boxed designs may well serve as a reference to take production/editing decissions. But only under controlled studio conditions with RT60 at about 200 ms and free of stationary modes below Schroeder frequency.

Also thanks @Juhazi for the link to Dipolplus downloadable pdfs that I've found so interesting to read and that I missed when I was at Rudolf Finke's website long ago.

Ben, I also use Stax SR-404 headphones to my own delight and a pair of LZ A4 hybrid balanced armature iems which I found very pleasing when I play my bass guitars over earphones.

Finally if anyone wish to listen to true holographic music tracks please check my personal playlist tribute to LX521 here:


The latest additions are the best ones since I'm putting the bar higher and higher. The speakers absolutely vanish when playing the content's list.

Best

Juan
 
Last edited:
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
Thanks for that playlist, I intend to abuse it a lot in the coming weeks. :)

Yep, Linkwitz, John and Gainphile are my major inspirations for this project and amazing people in general. :)

Thanks for all the support and comments you guys :)

I'm toying around with the idea of making a very small constant directivity coaxial bookshelf for fun. Maybe I'll get to it in a year or two after I'm done with headphones :D

The goals are stupidly ambitious, 1st make a working solid estat and planar, then try to make a true earspeaker open design - no enclosure headphone with as good bass extension as it gets, and add 3d realtime tracking and hrtf to simulate speaker placement and defeat the collapsing soundstage once and for all. I'm actually surprised nobody from the big guys (senns, audeze, stax, jbl, etc...) has tried to do it until now, they could for sure. eh.

Also someone asked about the group delay due to eq, I forgot to answer, no, I myself can't hear anything wrong in the bass region. But someone else might, I'm not a pro analysis tool like some guys
 
Last edited:

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,731
Likes
2,922
Location
Finland
Seems like we are a team!. Ardacer and JeyB! My 4-way dipole journey started in 2013, inspired by Jorma Salmi's Gradient 1.3 that I had listened at my "girl"friend's. In the beginning I needed and got a lot of help from reading sites mentioned and eg. JohnK and Rudolf themselves. John Reekie's way of using Minidsp actually made my dream possible ( passive is almost impossible even for experts)

John Reekie https://www.hifizine.com/2011/03/refining-a-4-way-open-baffle-speaker-minidsp-2x4/
Oliver Pirlich http://www.2pi-online.de/index.html

My AINOgradient story https://www.diyaudio.com/community/threads/aino-gradient-a-collaborative-speaker-project.231353/
- I chose monopole downfire bass because of copying Gradient design and because of necessity to position the speakers rather close to the front wall!

oh2.JPG ainogneo83 vx out 0-180 polar.jpg siggi juha 2.jpgjormasalmi juhazi s.JPG
 
Last edited:

Els

Active Member
Joined
Sep 30, 2022
Messages
123
Likes
51
I think you will be amazed at how that will sound. I was a very early fan of SL but actually if you have a four way quad amped system using reasonably selected drivers and laid out in a reasonable fashion (which SLs series of speakers were always reasonable) it is actually hard to make a system that sounds bad. It will outperfom 99% of the overpriced systems out there.

I became less of a fan of SL when I realized he had fallen under the evil influence of Nelson Pass and was charging "licensing fees" for a simple crossover network which was advertised as more complex than it probably really is. Modern DSPs make anything you want childs play. Also high prices for plans. You do all the work and he gets a licensing fee for something you are improving on.

For WAF use veneers or designer formica. That latter forms a laminate that stiffens and deadens the surfaces.

Even in te 1970s I considered building my speakers as either dipoles or omni radiators but finally decided on a line source configuration.

What drivers did you decide to use? I am always interested in very high quality drivers for relatively bargain prices. These days drivers have gotten so expensive that you can buy Yamaha Be drivers used for competitive prices. Current fleabay rules make it pretty safe to do so.

I am about to build a center channel to go with my stereo (now FL and FR ) speakers. It will look something like your design above but will only be a
three way and only need to go down as far as 80 -100 hz the rest being taken care of by multiple subs and a low frequency channel.
Hello from Belgium, I very impressed with your system, mostly because of the inphessis on midrange. It seem that there is agreament that the Yamaha dome midrange is one of the best, I listened, recently to the ATC dome midrange, I was very, very impressed. How do think the Yamahas compare. What cut off frequencies are you using with the D 23? What JBl drivers are you using? Are there any JBL 12 inch drivers that would fit in the Yamaha NS 1000M? Thanks, Els
 
OP
Ardacer

Ardacer

Active Member
Joined
Jun 13, 2018
Messages
106
Likes
134
Seems like we are a team!. Ardacer and JeyB! My 4-way dipole journey started in 2013, inspired by Jorma Salmi's Gradient 1.3 that I had listened at my "girl"friend's. In the beginning I needed and got a lot of help from reading sites mentioned and eg. John and Rudolf themselves. John Reekie's way of using Minidsp actually made my dream possible ( passive is almost impossible even for experts)

John Reekie https://www.hifizine.com/2011/03/refining-a-4-way-open-baffle-speaker-minidsp-2x4/
Oliver Pirlich http://www.2pi-online.de/index.html

My AINOgradient story https://www.diyaudio.com/community/threads/aino-gradient-a-collaborative-speaker-project.231353/
- I chose monopole downfire bass because of copying Gradient design and because of necessity to position the speakers rather close to the front wall!

View attachment 234710 View attachment 234709 View attachment 234711View attachment 234712
That's some mighty fine speakers and mightier company you've got there :D
I went to check the gradient 1.3 out, there is one used available for 1500 usd/eur if someone is interested, seems like a good deal. I bet these are some good sounding speakers, they look great too. My wife would absolutely murder me if I put something like that in our room (she wouldn't, she's great, but she would hate them for sure).
I'm so sorry I never had the chance to meet Linkwitz, and Kreskowsky retired some time ago.

I'm really happy you said we three were a team. Thank you for that :)
Hopefully more people get inspired into making better dispersion speakers themselves. Remind me to send you two a pair of my headphones when and if they are finished and actually work good. I'll open a new thread about that project soon I hope, it will be a long time.
If you, Juan, or anyone else interested finds their way in Croatia, you're more than welcome to visit me and hear and see waffles and lx521! :)
 
Top Bottom