Gradient's been doing this longer than John K or Siegfried Linkwitz. In fact, Jorma Salmi started writing about this thirty two(!) years ago : https://secure.aes.org/forum/pubs/conventions/?elib=11885
I respect and admire the work of all of them, but Gradient has been quietly writing the book on cardioids and dipoles for decades.
Believe me, I would have dismissed them too, until I heard the Helsinkis and my mind was thoroughly blown.
An externally hosted image should be here but it was not working when we last tested it.
But don't take my word for it. Earl Geddes conducted a blind listening test of his Summas (which are my reference), his JBLs, and the Gradient Revolution. The Gradient Revolution was right up there with the Summa, and the JBL fared the worst. I wish I could find the results of the listening test, but this thread was the closest I could find : Audio Asylum Thread Printer
Before you accuse Jorma Salmi of being a 'novice', consider the fact that the boomiest parts of your room are in the corners. We all agree on this, correct? Walking into the corner of any room, and the bass gets louder. Well guess what happens when you orient a dipole 45 degrees? Yep, you just aimed the null into the corners. Clever, no?
The Gradient Revolution takes it a step further, and allows you to rotate the midrange and tweeter independently of the bass dipole. So you can choose one of nine different configurations, depending on what works best for you.
Patrick,
I have built both JK(NaO MkII) and SL(LXMini) "clones" both sound very similar in my room environment. The main different being the JK's system is a much larger system hence more suitable for the a medium to large room. I have a small room the LXMini is more than adequate, it is because of this when I saw the LXMini I thought it would be idea for taste of speakers taking into account my room size.
Both are very good sounding systems with a lot more similarities rather differences because both are based on similar concept.
I have built both JK(NaO MkII) and SL(LXMini) "clones" both sound very similar in my room environment. The main different being the JK's system is a much larger system hence more suitable for the a medium to large room. I have a small room the LXMini is more than adequate, it is because of this when I saw the LXMini I thought it would be idea for taste of speakers taking into account my room size.
Both are very good sounding systems with a lot more similarities rather differences because both are based on similar concept.
Patrick,
His objective was not a cardioid response. His discussion seems fairly clear on this if you read the pages on his website. The cardioid response was yielded in the crossover range due to the monopole/dipole transition, but not because he was (specifically) targeting that pattern.
Maybe I'm reading his description and getting something different from it than you?
Regardless, the speakers sound terrific in my listening room. 🙂 I'm sure that doesn't help you or answer your questions.
I've never auditioned Gradients but I'm sure they're terrific speakers. I'm wondering about that photo with all the speaker setting atop milk crates though. Is that the way they were evaluated? 🙂
It seems the LXmini is probably not a speaker system you'd be interested in. That's too bad.
Cheers,
Dave.
His objective was not a cardioid response. His discussion seems fairly clear on this if you read the pages on his website. The cardioid response was yielded in the crossover range due to the monopole/dipole transition, but not because he was (specifically) targeting that pattern.
Maybe I'm reading his description and getting something different from it than you?
Regardless, the speakers sound terrific in my listening room. 🙂 I'm sure that doesn't help you or answer your questions.
I've never auditioned Gradients but I'm sure they're terrific speakers. I'm wondering about that photo with all the speaker setting atop milk crates though. Is that the way they were evaluated? 🙂
It seems the LXmini is probably not a speaker system you'd be interested in. That's too bad.
Cheers,
Dave.
Last edited:
Actually I'm pretty close to trying to build a knockoff. Just trying to figure out why you'd *want* it to go monopole -> cardioid -> dipole
An externally hosted image should be here but it was not working when we last tested it.
Beolab doesn't talk about it much, but their speakers are cardioid above 2khz or so. I know a lot of people don't like their sound, but I'm a fan. I like the way that the soundstage "floats" above the loudspeaker
In this thread, I made a Beolab lens with a Fostex 3" : http://www.diyaudio.com/forums/multi-way/239808-28-days-later-3.html#post3614006
These three pics show the following:
1) The on-axis sound of my Beolab lens clones and the sound 180 degrees behind the speaker.
2) The forward radiation of the LX Mini
3) The rearward radiation of the LX Mini
See how the shape is similar? I'd argue the Beolab lens does the job a little bit better (not a lot) and I think that's mostly because of a lack of diffraction. (The response of my lens is completely unequalized, while the LX Mini *is* equalized.)
Believe me, I'm not trashing the LX Mini in the least, just wondering if it's possible to modify the enclosure shape to smooth out those polars
There is a profound difference between NaOs and LX mini - below 1kHz LXmini is not dipole, it's omni. LX521 is more like NaOs. Me thinks that low mid dipolism is of more importance than upper mids.
I see it as a trade-off for small size.
You want to make a small speaker that is dipolish. The lower you get in frequency, the required size for dipole/cardioid increases dramatically - something you can control by choosing to go monopole at some point. Then you cross from dipole to monopole and - guess what - you get a cardioid response in the transition.
There is also this other thing I have noticed. while tinkering with speakers For roughly the same frequency response, dipole bass/mid-bass excels with certain types of music, particularly large orchestral, double bass and generally acoustic strings instruments, while stuff like classic/hard rock and anything containing drums seem to lack something (slam ?) and sound more realistic with monopoles (to my ears). Call it weird, but it is as if "velocity" instruments would sound better with velocity speakers (dipoles) while "pressure" instruments like drums sound better with pressure speakers (an active cardioid midbass could be best of both worlds).
Its just a consequence of the physics involved.
You may not want it, it just happens.
I own the Pluto's and think they are very good. But I'm sceptical about the LXmini. It sure gets great reviews, hopefully I can get a chance to listen to a pair.
Actually, as per my measurements of several speakers mono or dipole, cardioid response around 500-3kHz would be optimal. In decay and RT measurements we usually see these Fs decayed/reflected the most.
It is practically/physically impossible to do a speaker that is dipole all through it's range. It is nice that we are able to achieve different radiation models and set transition ranges at different F. Sadly this is a never ending story, because rooms and personal preferences vary so much. Understanding these limitations and challenges is worthwhile!
The closest things to the LXmini that I had designed were a speaker prototype called the Bird Cage and the now discontinued NaO Mini. The Bird Cage had a down firing woofer, dual (front to back) 4" mids and a Neo3 dipole tweeter. The speaker was monopole at low frequency with a cardioid transition to dipole mids and tweeter. That transition was around 300 Hz. The NaO Mini was a simple flat baffle MTM dipole panel (but no rear tweeter) coupled to a self powered monopole woofer. Transition between mids and the woofer was dependent on how the woofer/panel was positioned with the recommended positioning yielding a cardioid transition centered around 120 Hz..
Going back to the Gradient speaker, and dipole woofers in general, angling dipole woofers so they are not aligned with a room axis makes it act more like a monopole with regard to room modes, but it is still unable to pressurize a room as monopoles and cardioids do. Once the dipole woofer is angled, modes in the side to side as well and front to back, and all associated modes, become excited. It then becomes only a matter of source strength as to the level of modal excitement. Since all low frequency sound in an enclosed space is modal in nature, what you hear is totally dependent of which modes are excited and at what level. For this reason, and the fact that a dipole radiates less power into the room for the same axial response level, a dipole woofer must be boosted even further to achieve the same level of bass as a monopole or cardioid. It is for this reason that with, for example, the Note II, I recommend adding a monopole sub woofer to the system with SLS woofer rather than substituting XLS or XXLS woofers. Adding a self powered sub provides the bass impact some find missing with dipole woofers and is a more cost effective solution to low frequency response than staying dipole and using an array expensive, long excursion woofers. And if you don't want the bass impact on certain types of music, you can turn the monopole sub off.
I've been building horns of various types since the 1990s.
This year is the first time that I've built a front loaded horn that does two things:
1) it goes into the midrange
2) it uses a driver with a lot of displacement
I was startled by how 'punchy' it is, really fell in love with it.
Here's my theory:
In a vented box, back loaded horn, or tapped horn, the room is not being pressurized. (Because the back chamber is not closed.)
A sealed box can pressurize a room, but there's an impedance mismatch between the room and the cone.
An externally hosted image should be here but it was not working when we last tested it.
But a front loaded horn reduces that impedance mismatch, so it can 'push' the air in a way that you can really feel. It's quite remarkable! And fun. Similar to those water jets at the Bellagio.
Hi,
Total off topic... sorry :
On the photograph : due to the magnet size and the tube size in relation to the diameter of the cone : Is the tube here just to tie the magnet like it seems the rear cone seems radiate outside ot this tube (not Inside : too less space between magnet and Inside diameter !) ?
In other words : is the tube a load or just a simplier & cheaper way to tie the driver for a NB (non baffle) design ?
1/2 cents... sorry. (Just can(t understand this lake of high frequencies : are the réflexions with - (minus) decay xxx m second on the walls are enough ? a way to corelate with the slower bass ? This speaker is really strange in relation with what was made before !)
PS : if any nice french maid it, please let me know by PM... if not too far from home, I will be pleased to hear this speaker !
Total off topic... sorry :
On the photograph : due to the magnet size and the tube size in relation to the diameter of the cone : Is the tube here just to tie the magnet like it seems the rear cone seems radiate outside ot this tube (not Inside : too less space between magnet and Inside diameter !) ?
In other words : is the tube a load or just a simplier & cheaper way to tie the driver for a NB (non baffle) design ?
1/2 cents... sorry. (Just can(t understand this lake of high frequencies : are the réflexions with - (minus) decay xxx m second on the walls are enough ? a way to corelate with the slower bass ? This speaker is really strange in relation with what was made before !)
PS : if any nice french maid it, please let me know by PM... if not too far from home, I will be pleased to hear this speaker !
Last edited:
The tube behind the full range driver acts to reduce the output directly to the rear to allow closer placement to the front wall.
Thanks,
I believe I understood the concept but not the technical explanation !
On the photograpgh the rear tube ( which is a smart way to tie a driver by its magnet if no rear center screw hole) diameter seems very near to the diameter of the magnet.
So I don't sucess to understand how the internal load of the tube can work ?! Again with the photograph: the rear cone seems bigger in diameter than the tube!
Is the tube not a diffusor on its external side ? Or if if it's an absorber in its load : is the litle open circular between the diameter and the magnet enough ? Ok a cone is beaming with the higher frequencies on its center; but here the rear cone radiation is closed by the relative large size of the magnet in relation to the cone one, and with the lesser size of the diameter tube in relation to the cone size: how the tube can absorb something ?
How can this "load" works tehnicaly as a closed load at highs fhz to a cardioid at low FS of the driver Fs range ? Is it just not due to the natural beaming of frequencies when they go up ? does an open driver at those frequencies beam naturaly from the low to the high like a cardioid at he botom to a sealed tweeter above (with a rear chamber... so sealed)
Why not e.g. a larger tube with equal diameter with the rear of the cone or even larger ? For example SL said he found on the LX521 the form and size of the bafle with some tries and errors. Well no critics here, just don't understang the mix between the easy tube because easily sourcable and the beginning of math due to the size (both diameter & length) in relation to the double function of tiying the magnet and open-cardioid load ?
I believe I understood the concept but not the technical explanation !
On the photograpgh the rear tube ( which is a smart way to tie a driver by its magnet if no rear center screw hole) diameter seems very near to the diameter of the magnet.
So I don't sucess to understand how the internal load of the tube can work ?! Again with the photograph: the rear cone seems bigger in diameter than the tube!
Is the tube not a diffusor on its external side ? Or if if it's an absorber in its load : is the litle open circular between the diameter and the magnet enough ? Ok a cone is beaming with the higher frequencies on its center; but here the rear cone radiation is closed by the relative large size of the magnet in relation to the cone one, and with the lesser size of the diameter tube in relation to the cone size: how the tube can absorb something ?
How can this "load" works tehnicaly as a closed load at highs fhz to a cardioid at low FS of the driver Fs range ? Is it just not due to the natural beaming of frequencies when they go up ? does an open driver at those frequencies beam naturaly from the low to the high like a cardioid at he botom to a sealed tweeter above (with a rear chamber... so sealed)
Why not e.g. a larger tube with equal diameter with the rear of the cone or even larger ? For example SL said he found on the LX521 the form and size of the bafle with some tries and errors. Well no critics here, just don't understang the mix between the easy tube because easily sourcable and the beginning of math due to the size (both diameter & length) in relation to the double function of tiying the magnet and open-cardioid load ?
Last edited:
I must have overlooked this fact, thks for the reminder.
BTW one question the omni directional wave below 1 kHz, do we have to allow for baffle step compensation below about 300 hz?
I hope Mr Linkwitz will contact Dr. Sean Olive and subject his designs to a double blind test. See: Behind Harman's Testing Lab
I'm very curious about the outcome.
The tube acts as both diffuser and absorber. A bigger radius will make it more of an absorber, a smaller radius will make it more like a diffuser.
Apparently this radius is the optimal compromise between diffusion and absorption.
I doubt that Dr. Olive does tests on request. They are very time consuming, thus expensive.
Last edited:
There will still be a floor bounce.
Point a 5" woofer forward, or point it up, it makes no difference to the radiation pattern below 2700hz. (Because the wavelengths are larger than the radiator.)
- Status
- Not open for further replies.