I think the thread is derailed from the original idea to use subs to mimic the ESL radiation/directivity at lower frequencies.
That's what it says in the thread title.
Happens all the time, some guy jumps in without having a clue of what the discussion is about and starts beating his chest.
Ohh well. It was nice while it lasted.
Jan
That's what it says in the thread title.
Happens all the time, some guy jumps in without having a clue of what the discussion is about and starts beating his chest.
Ohh well. It was nice while it lasted.
Jan
Well, to quote Newton, F=MA. I have repeatedly seen pushback on terms when discussing low frequency reproduction and this dislike for describing it as speed or fast or whatever. Whatever you want to call it, the more it weighs, the more force it takes to accelerate and decelerate the mass. The less rigid it is, the more it will flex and distort.. I agree with you that a low Q sealed box will provide a low distortion and linear bass, but my issue is from 25-to-80HZ. I have listened to more than a few bottom octave drivers. I've heard all sorts of vent noise, cone breakup, materials resonances, Floppy drivers and mushy bass. The Gradient does well- but as I have to imagine the pieces exist to update that work with modern components.
Again, we are talking about a SUB. Not the monitor.
A correctly designed sealed sub does NOT have the boom. It blends very well in a room.
When you go adding boost to a sub, that means you have not modeled the excursion or measured the distortion. I have.
To each their own. You can lead a horse to water, but you can't make him drink.
A correctly designed sealed sub does NOT have the boom. It blends very well in a room.
When you go adding boost to a sub, that means you have not modeled the excursion or measured the distortion. I have.
To each their own. You can lead a horse to water, but you can't make him drink.
Yes- Contrapunt. it's the link in my first post. That is virtually identical to the Gradient solution developed for the Quad. My guess is that it sounds similar. But they are still using the active EQ, which I'd like to eliminate if possible. Either that or dramatically improve.
Ideally it would be a driver/alignment that had the ability to perform without EQ. I was just writing a question about the Ripole- will continue below.
What about the Ripole? In looking at that configuration it appears to use the opposing driver's compression loading to lower the resonant frequency by as much as 10db.
Physics is physics- it's the ultimate "no free lunch" environment. so my first guess is that there's a trade-off.
Physics is physics- it's the ultimate "no free lunch" environment. so my first guess is that there's a trade-off.
Off-Topic, but considered important:
Yes, the backside output can be annoying. You have to deal with it in order to avoid an intense, early reflection. On the foto in my former post a test setup ist shown where a mirror is deflecting the backward emitted beam away from the listening position. It works. In the graphs below red shows an optimized result by deflecting/absorbing some earlyer, unwanted reflections. Note that the backward reflection shows aup as a negative spike at 0.147 in the black graph. In the red graph it has been "mirrored" away.
On-Topic again:
Unless Qts is >0.7 the resonance of a driver will not cancel any of the -6dB SPL slope for a LF dipole system. E.g. a standard Q of 0.4 for a Bass dipole will negatively add up it's own diving SPL slope to the SPL slope of the dipole. Ending up at a slope of << -6dB towards the lower end. This is the cruel tradeoff for LF dipole speakers. And this is why bass drivers with Qts > 0.7 are most welcome within LF dipoles. We would really best need a big, pro-quality, high SPL driver with a Q around 0.7 ... 1.0 for a best fit. But to my knowledge, there is no such driver.
What are the spec's of these Gradients and Daitons mentionned first place?
Yes, the backside output can be annoying. You have to deal with it in order to avoid an intense, early reflection. On the foto in my former post a test setup ist shown where a mirror is deflecting the backward emitted beam away from the listening position. It works. In the graphs below red shows an optimized result by deflecting/absorbing some earlyer, unwanted reflections. Note that the backward reflection shows aup as a negative spike at 0.147 in the black graph. In the red graph it has been "mirrored" away.
On-Topic again:
Unless Qts is >0.7 the resonance of a driver will not cancel any of the -6dB SPL slope for a LF dipole system. E.g. a standard Q of 0.4 for a Bass dipole will negatively add up it's own diving SPL slope to the SPL slope of the dipole. Ending up at a slope of << -6dB towards the lower end. This is the cruel tradeoff for LF dipole speakers. And this is why bass drivers with Qts > 0.7 are most welcome within LF dipoles. We would really best need a big, pro-quality, high SPL driver with a Q around 0.7 ... 1.0 for a best fit. But to my knowledge, there is no such driver.
What are the spec's of these Gradients and Daitons mentionned first place?
https://aespeakers.com/product-category/dipole/ have some, Qts almost 1, but at a price. That's when I looked for restoring the originals.
Jan
On your first point, I definitely agree.
The quads perform remarkably well down to the 50hz region without a sub and the 6-panel versions have every bit of bottom end you'd need for light jazz, etc. I have both the 63 and the 989 and the 989 is actually an exceptional speaker when rolled off at somewhere in the 85-95 hz region. It brings out a bit of warmth that's not there in the 63
My early investigation led me to some recommendations for replacing the discontinued Peerless drivers in the Gradient- one suggestion was the automotive version of this driver https://aespeakers.com/shop/ibau-woofers/ib12au/
FS at 24 and a Qts at .51. A hundred bucks less and five times the power capacity (100 vs 500) What am I missing?
In the reading I've done on Linkwitz' work on dipole subs (IIRC) the advantage of his V-shaped woofer orientation was a lessened interaction with room walls. I have found very little elsewhere that anyone has done anything with his arrangement aside from building his LX521. I have spoken to peole who have heard it and the feedback is quite positive. I have no idea why it seems to be relatively ignored.
I know from personal conversation with Siegfried at the time that he was also concerned with the sub 'swaying' (loss of a better word) as a result of high output levels.
At the time I understood that the V-shape countered that to some extend. I don't remember any details..
Jan
At the time I understood that the V-shape countered that to some extend. I don't remember any details..
Jan
The V-Shape reduces some of the whole system dynamic back-force in terms of vibrational control. Actio (Moving parts of the Driver + Air Load) = Reactio (Baffle). And it's also a way to implement a push-pull-configuration of both drivers in order to linearize for membranes excursion asymmetry. It has nothing to do with room control, then. The LX521 behaves as a dipole, it's two openings and it's frame geometry/depth is determinating for the dipole characteristic. And not the V-Baffle inside.
The V-Baffle might have the advantage over the W-Boxing that there is no distinct cabinet depth. So no lambda/4 resonance like in a W-shaped boxed dipole (such as the one on my fotos). E.g. take a W-Box-Shaped Push-Pull with an inside size of 34cm in order to accomodate two 12-inchers. This box will have a first lambda/4-Resonance at 250Hz, which will call for some equalization and/or sooner/steeper cutoff. The V-flavor has no such resonance and therefore might be used further up with no need for equalization. That seems to be the merit of the V-Form.
The merit of the W-shaped Dipole instead is total cancellation of vibrational forces. As the V-Form does not perform this natively, S.Linkwitz hung a mobile internal drivers-carrier sub-frame into a static outer supporting frame. So the vibrations from the inner frame does only very weakly transmit onto the outer frame.
Go CamillaDSP!
You will have to perform a closed-range/nearfiled miking to assess the inbuilt drivers behavior within theirs supporting construction. And then convolve this measurement with a theoretical dipole behavior/slope (-6dB/Okt). And then inverse this. You will get a response that will rise by some 10dB/Okt or so towards lower frequencies. Then, you will have to convolve this response with your target X-Over behavior. The result of all that will be your filter, ex the room response. You see that you will choose a low cut frequency wisely.
Taking the spreadsheets on Linkwitz and the further infos on his aesspers.com site. Fill in the AESSpeakers Dipole15 specs: SD 855cm^2 and Xmax=15mm. Take either the naked driver (D=20cm) or a square baffle of 60cm x 60cm (D=30cm). Then you get the following calculeted max. SPL's:
30Hz: 77dB naked / 82dB baffled / 88dB floor standing baffle (+6dB for the floor mirrored system)
40Hz: 85dB naked / 90dB baffled / 96dB floor standing baffle
50Hz: 91dB naked / 96dB baffled / 102dB floor standing baffle
60Hz: 97dB naked / 102dB baffled / 108dB floor standing baffle
These are all approximative, but realistic values. Not so much bass SPL for a lot of money, then ...
To understand your bass needs its necessary to know why you roll off the 989 in the 85-95 hz region.
Its meant to go down to 40Hz:
Because they are being used with the Gradient subwoofer right now. The original Gradient electronic XO crosses at about 95, but I don't use its output for the Quad as it's not a very high quality item to have in your signal path, and my room is exceptionally dead. Running the Quad just a little lower sound better- and measures more flat. Not that I care about measurement all that much.
Of course the Quad on it's own performs wonderfully down to the 40hz region. Would sound like a clock radio rolled off at 95 without the sub.
Why use a sub then?
What crossover do you use?
Can you post your measurements and explain how you take them?
The more information you provide the better the answers you will get.
You might want to specify your budget too.
Taking four panels, each 15cm x 56.5cm, and assuming an Xmax of 3mm, at D=30, then the Quad 63 should be able to provice a max. SPL of some 95dB @ 40Hz.