It's not performing "worse", it's just that the mass loading version is showing slightly less efficiency. It's pretty much what I would have expected as that's what MJK's simulators predict. If you make the opening smaller you will lose yet a little more efficiency.
Using other drivers instead of the AN-15's would not change this basic issue: mass-loading or slot-loading like this will generally lower the low freq cutoff, scrub off a couple of dB of efficiency and more than likely create a peak in the response in the 200 to 400Hz range.
This also applies to Nelson Pass' slot-loaded speakers, and also the ripoles, which are closely related. Basically, it's a good way to achieve low frequency extension if you have the efficiency to give away and can reduce or eliminate the peak @300z
In your measurements, the low end cut-off point does not seem to be changing but I'm guessing that is limited by measuring in the room...
e.g. a situation where the distance from the back wall is creating a fairly wide and deep cancellation in response at 30Hz, even if version A is flat to 37Hz and the mass-loaded version B is flat to 30Hz their apparent in-room LF roll-off could still look almost identical with the -3 dB point at 40Hz.
In practice, because of room placement issues, I've found that the "practical in room limit" for OB is about 40Hz, below that it may just be better to augment with a cardioid or a mono-pole type woofer. It's not a law, just a rule of thumb for what seems to work best within fairly practical limits.
But for the range between 40Hz and 300Hz, so far I've found the simple OB to be the best sounding arrangement in most rooms.
Keep on cranking!
Using other drivers instead of the AN-15's would not change this basic issue: mass-loading or slot-loading like this will generally lower the low freq cutoff, scrub off a couple of dB of efficiency and more than likely create a peak in the response in the 200 to 400Hz range.
This also applies to Nelson Pass' slot-loaded speakers, and also the ripoles, which are closely related. Basically, it's a good way to achieve low frequency extension if you have the efficiency to give away and can reduce or eliminate the peak @300z
In your measurements, the low end cut-off point does not seem to be changing but I'm guessing that is limited by measuring in the room...
e.g. a situation where the distance from the back wall is creating a fairly wide and deep cancellation in response at 30Hz, even if version A is flat to 37Hz and the mass-loaded version B is flat to 30Hz their apparent in-room LF roll-off could still look almost identical with the -3 dB point at 40Hz.
In practice, because of room placement issues, I've found that the "practical in room limit" for OB is about 40Hz, below that it may just be better to augment with a cardioid or a mono-pole type woofer. It's not a law, just a rule of thumb for what seems to work best within fairly practical limits.
But for the range between 40Hz and 300Hz, so far I've found the simple OB to be the best sounding arrangement in most rooms.
Keep on cranking!
Sorry to hear you're feeling ill. Get better quick. Your posts are too much fun to go without.
Emerald's probably using the mass loading to get much more extended bass output while using less deep bass boost in the EQ. It's a trade off which is likely chosen as a result of the Alpha 15's "gigantic" 5mm excursion. Note the opening they're using also has a port tube on it. This starts looking like a band-pass filter, and that can indeed boost the efficiency, while reducing the mid-bass peak.
Also note they're using 48dB/Oct filters at 100Hz! This takes care of any residual mid-bass peak. And since the EQ is active, any efficiency disparity with the main element is inconsequential.
So I would say it's not marketing BS, just a clever way of optimizing the system parameters.
That being said, I am NOT a fan of higher order crossover slopes, and Emerals is definitely using them. For me, it's gotta be fully time-coherent or just not listen to music at all. Criminy, why else go to all the fuss using these expensive full-range drivers to begin with?
I think he's getting an increase in velocity near the slot, which would look like a SPL boost in the near field. But as distance increases from the array, the overall efficiency actually drops... I vaguely recall he said something to that effect....
Actually, it's basic physics: adding mass to a speaker while keeping the Bl the same will reduce the overall efficiency. However, that changes if you add a tuned resonant circuit (such as a band-pass load like Emerald Physics is doing) in that case you can indeed get a significant increase in output in the pass-band of the filter. Also, see the Bose subs for how a tuned band-pass can yield higher sensitivity in the pass-band, I once took one apart and found the basic driver efficiency was about 86dB for a sub that clocks in at 92dB.
I don't see where Nelson's system could look like much of a band-pass, so I can only assume the velocity increase is the reason for near-field SPL increase.
Member
Joined 2009
Paid Member
Yes I know, but like you, I am also interested in flea watt amps or at least class A amps of all designs.
My design goals are to produce a speaker that can be run direct from a cd player through a source follower/power buffer without the need of voltage gain. If I use the XLR output of my dac/cd player then I can achieve that.
My design goals are to produce a speaker that can be run direct from a cd player through a source follower/power buffer without the need of voltage gain. If I use the XLR output of my dac/cd player then I can achieve that.
Last edited:
Member
Joined 2009
Paid Member
that sounds like a noble goal ! - so it means using lots of woofers to get that sensitivity into the low registers for OB, as you are planning to do already. I'm not sure about the slot loaded baffle. I read quite a bit of the relevant thread and it left me unconvinced unless you plan to use it for near-field listening. I've never heard one though.
Hey Bigun, I don't think there is an intrinsic advantage to Nelson's slot-loaded approach, either in efficiency or in output or in distortion.
But there is one very cool way to use it, and this one does have a very noticeable advantage: place the woofers back to back with close mechanical coupling of the baskets. This reduces the architecturally transmitted noise by a very large amount. Also reduces the "modulation" issues caused by cabinet movement to almost nothing.
Think of it this way: 8 woofers on a flat open baffle, 15" diameter each, 95dB sensitivity, each with a 125 gram effective moving mass (adding cone mass and air load)... that's looking like a whole 1Kg (2,2lbs!) of effective moving mass!
If the flat baffle is relatively low mass, say 20Kg, that flat baffle is gonna be moving around quite a lot at the unsupported end. Visibly so.
I've done this... it's plainly visible.... with 'stats also: I had Acoustat 2+2s, with big bass transients, we measured the movement at the top of the cabinet... can you believe a full 1/2 inch back and forth oscillation?? From "near-zero mass" electro-stats?
How could this be? Well, when you calculate the moving air mass of a 80" x 16" diaphragm, you'll see why!
OK, so what if we brace it and support it with a solid and heavy stand? Now the vibration that would otherwise move the flat baffle is being directly transmitted to the ground. Wow! Bass you can really feel with your feet! And.... later, we start noticing a huge amount of coloration coming from... the floor.
Instead, let's use Nelson's slot-load with 4x woofers on one side which get mounted back to back with the 4x woofers on the opposite side, with the baskets mechanically coupled. The force from the woofers cancels out, and voila, architecturally transmitted colorations are reduced to inconsequential levels.
The great advantage of using the "slot-loaded" approach is that it gives you easy access to the woofers (from both sides) for more complex mechanical mounting schemes such as what I'm suggesting here. And if you don't like the mid-bass peak from the slot loading, make the slot wider, put in some absorbent material and voila.
The 4x4 slot load as described above (with 4 in parallel = 2ohms, and then in series to get 4 ohms total) will yield about +12dB broad-band increase over a single woofer (at 2.83V/1m), thus allowing us to to "throw away" lots of that extra output with an entirely passive eq.... and thus we get good efficiency match with most full-range units.
Furthermore, if you don't like "wide-wing" OB's, fold'em back. Now you have a large "twinned" ripole which can kick some serious sss and look nice also.
Whatcha think, Melon? You up for it? I've got some protos' of just this in the works, I'll send photos when I can.... if only I can get my woodworkin guy to be more responsive. Somehow he seems to think other projects are more important even tho' I'm paying him equal $ per hour. Too bad I get wheezy if I'm around wood-dust for long, or I'd do it myself and dang the torpedoes!
Gary Pimm found that having the drivers in opposition, as is described by Jack, but in opposed mounting, with one front and one back exposed to the same volume, provided better transient response. I do have to say that any better transient response at those frequencies would certainly cause subdural hematoma over most of your body....as apposed to just sloshing about on your brain.
Bud
Bud
I'm chasing something, similar to getting struck by lightning, so outside of flying a kite with a key attached on a rainy day, what do you guys reckon? Is it the W or NP's slot loaded configuration
If I can't achieve the lightning strike, then a sledge hammer travelling at mach 10 will suffice.
I don't care about sub-40 Hertz
If I can't achieve the lightning strike, then a sledge hammer travelling at mach 10 will suffice.
I don't care about sub-40 Hertz
Last edited:
This will mostly be a function of frequency and SPL.
I have also done the same thing at SPL levels that I normally listen at, and didn't get quite near that level of oscillation. Regardless, I like all your ideas.
Last edited:
Hey Melon, thanks. About the "visible oscillation" I measured on the Acoustat 2+2's, I think it's safe to say that it probably isn't the best thing to have an un-supported panel frame which is all of 3" or 4" deep, nearly 7ft tall. That type of construction would really benefit from a support structure to keep it from moving.
A regular high-mass panel, supported with an adequate beam would not behave in the same way.
So it pays to have very solid support, right? Well, that depends on the mass of the assembly and the coupling to the floor, and the flooring itself, because then you are transmitting a lot of vibrations into the floor itself, and depending on the type of floor, that can cause some very serious and audible colorations.
But, lemme tell ya....
the version which pretty much eliminates all of these concerns is the force-canceling type.
You can even make it with very light but rigid materials, and still get pretty close to zero panel vibration or mechanically conducted floor vibration. (And yes, the acoustic wave coming from the woofers will still excite the floor, so it also pays to have a good floor!)
I implemented a force-canceling array with my omni system, the H3, (check the details on my website), and was absolutely stunned by the difference it makes when the mechanical conduction through the floor is mostly eliminated. (You could say I was really "floored" but that kinda goes against what I'm trying to convey!
)There's a reason KEF and other high-end makers uses force-canceling on their top-o-the-line units (e.g. KEF Blade). It's expensive to build, otherwise I'm sure we would see it on their cheaper products also.
One of the big "hidden" issues here is that the floor-borne vibrations usually arrive BEFORE the acoustic wave, due to the speed of sound through solids, into our feet and via bone-conduction, into our ears. (And there is not so much absorption through the solids and bone, meaning the in-ear levels can be relatively high in comparison to the acoustic wave.) And this can play all sorts of havoc with time arrival information AND create huge colorations in the bass range. Talk about room colorations... sheesh!
When you have a suspended hard-wood floor like I do, it's even more obvious because that's a big-sss bass drum!
So now, with my new-found conviction that OB provides the best room interface from 40Hz to 300Hz, I can't wait to get the new force-canceling OB woofer arrays built.
So by now you can probably guess which version has my vote. You could build'em back to back for tight mechanical coupling and easy adjustability of such, or try what Bud P. relayed about Gary Pimm's system. If you build them you'll be glad you did!
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.