radulescu_paul_mircea

GP in which enclosure? The single sealed? Or did you managed to try it into Skhorn?

It seems Eminence went for a different chassis. The old one probably was too similar to those from B&C

It does have over 3 dB higher sensitivity at 1.41 V across the band from 28 hz up, 1 vs 1 but it doesn't get a lot more maximum output, at least not between 28-45 hz. The Ipal drivers can take +4dB more power so they compensate.

Good work guys! Attacking the subject from so many angles. Regarding a piston in a sealed enclosure. It will never be a good Idea. The efficiency of a piston in an infinite baffle is dependent on the force^2* SD^2/ MMS^2*Re and some constants in the fraction like speed of sound and density. This means that any increase in force squared/mass squared will give an increase in efficiency. But also having bigger radiation surface. Also, there is a relationship between mass and reactance which will change the behavior quite a lot and will make the system non linear. So a hugely powerful piston would be very good only if attached to a huge light weight surface and an amp made for constant current instead of constant voltage to be able to properly work with huge reactance. But a horn system is a way of making this behavior change. Higher compression ratio with long enough path and big enough mouth will improve efficiency and could make use of a higher force. But at the cost of quality, having higher distortion and nonlinear behavior, like power compression, mechanical noises. I feel that for pro audio and my own goals , a dual coil dual 21Ipal enclosure, with 4 times the thermal capability and twice the excursion, with an amp capable of delivering enough sustained power for enough time in a 150*100*70 cm enclosure, tuned to do 30-150 hz full and flat, having 3 dB more than Skhorn XL with K20 bridge maximum burst output at 50-100 hz and 6-8 dB more from 50 hz down. This year I was able to play with Danley BC218 . They were stunning, fully flat and hugely powerful from 26 hz up! More output than my pair of Skhorn XLs, cleaner, deeper, less compressed, huge directivity. They are made with good but normal drivers, 18SW115 so we can see here how the enclosure contributes to the end result as much or more than the driver's capability. But way too big for anything else than a huge stadium or a funny outdoor festival. And if a better driver was being used , the enclosure would be better? If the answer is yes, how much could you increase the drivers performance until you reach the limit of the enclosure?

Graphene drivers already are out there. MAGICO is using them.

Are you talking about TH50, aren't you? Those were my thoughts also

I have noticed you said about using a "8002" amplifier.which one is that ?

I will state my idea of this "perfect" pro audio driver for the most powerful subwoofer one could make in a given volume. The enclosure volume is maximum 1000 liters, weight is 175 kg max, frequency response should be -6dB at 25hz and +3 dB at 120Hz from the 40-50 hz output. This driver might already exists, but it is not readily available, it is the 24" driver from the new Funktion one F124. It is dual 8 ohms 6" coils, BL is 50 Tm but that is the only thing I really know about it, I didn't even get the chance to hear it. But now, going back to my idea of it, I know that any driver will in fact be just the engine part of a sub and it has to be matched to the enclosure for really good performance. Also , some of the qualities I will state about that driver might be negated or might be improved by the specifics of the enclosure design, but I will still play this game. The best driver IMO is always big. This means heavy most of the time. A 21-24 inch driver, with dual 8" diameter differential drive 1 ohm coils each. Length of each should be 70 mm , copper clad alluminum or just Al, flat wound ribbon wire, with inside and outside turns. The coil former made from a better conductive material than fiberglass and also a better thermal inertia, I am still searching for one with similar hardness/density with those properties. magnetic gap should be 30 mm, but with the proper extension on the pole pieces to decrease magnetic losses and to keep the magnetic lines perpendicular over the coils for the whole length. overhang should be 30 mm a side plus 10 mm the other, giving it a perfectly flat BL curve for the first 10-12 mm for maximum impulse response accuracy on the higher end of the spectrum, but this will also give a lot of excursion capability, good low end response and also cooling. This means it will need a smart magnetic circuit design, and the way the guys at Digital Design are doing it looks like a winning strategy IMO but it could be improved. Using radial magnets on the inside and outside of the coils, fixed in place by iron bars in anodised alluminum casings on the faces opposed to the coil (inside and outside of the motor toroid with the coils in the center), but with iron pole pieces shaped for conducting all the field perpendicular to the coils, The bar design with the visible coils are making the cooling easier if smart port positioning is used, and from the inside, there are a number of ways one could use the diaphragm displacement to cool the coils. From my calculations, about 6-8 kg of neo and 12-16 kg of iron and 2.5 kg of anodised Al would be needed to make it close to the M-Force motor strength but with way lower Le/Re. It will have much lower thermal inertia and it will still be harder to cool down, but the lower moving mass would make it more efficient. Diaphragm made from a laminated carbon fiber/aramid honeycomb/ carbon fiber sandwich, low mass , huge strength, W shape with the dust dome and cone in a single piece. The centering of the driver in the magnetic gap could be made with a spider separation ring that could be screwed into the chassis directly with all the parts already glued, like A&D Audio from China is already doing. Spider materials and shape should be similar to that of the new NTLW5000 drivers but probably at 12-15" diameter for enough excursion, dual layer separated by a ring with enough distance to ensure the very long coil stability. The total weight will be somewhere between 30-35 kg if the whole chassis is made of aluminum, encasing the whole motor like Faital pro are doing with their drivers. I would make a single driver enclosure with this one and it would finally satisfy my stupid wishes. With smart cooling it could take 2.5 kw of real power indefinitely with 6dB pink noise in band inside the enclosure with all setting needed. It might still not be exactly where M-FORCE+M-DRIVE are, by using a 30" diaphragm, but not too far either. What's your opinion?

The ports are the corner iteration from their own 2*15" version, the VTX V25. I've seen that one in real life and I've heard another version of the VTX with single 18" driver, but with their nice dual flared tubes and they really really work way better than a similar surface square vent. I bet they went for a higher motor force , higher moving mass with lower Q, increasing sensitivity on the upper end.

Looking at the cutout of the 2269, it seems that the design for NTLW5000 driver is pretty much the same. The pole pieces are the same shape, the magnet inside is single 2 cm thick ring of about 4" external diameter with a center whole inside of 2 cm, meaning about 1.1 kg of neo. The external metal ring is also closing the magnetic circuit, but not having a similar heat sink, it might not dissipate the heat as well. All my tests show that the driver behaves absolutely stunning up to 50 mm peak to peak, after which it starts to become noisy. Klippel results show flat BL until +-10 mm, similar to that of a split coil design, but then it dropps with a very low slope, remaining at 75% at +-17.5 mm.

http://www.jblpro.com/www/products/tour-sound/vtx-a-series/vtx-b18#.XNwR5LdRU0M Yes , I got my drivers a few weeks ago, they are absolutely stunning. I love the way they perform and the new materials used for them

Very nice!! I love that driver. This year JBL launched a new sub in the VTX line with a new driver and they say it is a big Improvement, +4 dB at 80 compared to 2269H and better power handling, 1400 W instead of 1200W. It seems that they are not using the AES standard of power rating, instead they use Klippel measurements. A 2kw driver AES like 21SW152 can take around 1.1 kW nominal based on RMS voltage in the nominal Impedance on the Klippel LSI reports, with 4-5 dB power compression. This is probably the case with these drivers also. Now it's time for the new NTLW5000 models . I have a pair of 21s with 4 ohm coils , first of them ever. The motor force is over 200 , coils are 2 of 38 mm each minimum from my calculations and verifications, copper, wire gauge is 0.6 mm, magnetic gaps are 12 mm . The 2 coils are 8 ohms each and linked in parallel. The spider suspension material seems pretty much the same as that from this JBL driver.

In my experience, the driver is excursion limited in a lower tuned enclosure and even though the motor force is very high, moving parts very good, I still like the 18" Ipal version from B&C more. It got to 1-3 dB max burst SPL from 21Ipal, so a very good performance but the thermal stability and overall capabilities are also 3 dB lower, but the price is very similar. In fact , I like the RCF 21N551 better, it is very similar concept and weight, but way higher excursion, mostly after xmax. But that one doesn't come with 4 ohm Impedance so one will need a pretty huge amp or bridged amp to power it to the max output.

Nice! Very good efficiency. I had mixed feelings with this driver and still have.

Huge voice coil and very heavy. They compensated for the Neo quantity by using copper , increasing the length of wire in the field to increase BL instead of using Alluminum wire with way more neo, like 21Ipal

The output power of a K20 vs an X4 K20: 145 V RMS for less than a second at 2.05 ohms both channels driven Then it limits to 58 V then it goes down to 52. This means 10.5 kw per channel for long bursts 1680 watts for 4-10 seconds 1350 watts per channel for almost a minute then thermal protection. Similar output is offered at 2.5 ohms , 70 V RMS for 4-10 seconds. This time constant depends on the internal temperature. The X4 can do 2950 W for 2 seconds at 4 ohms, then 1600 watts per channel at 4 ohms for 20 seconds single channel. 2*2650 w for 0.8 seconds then 950 watts per channel at 4 ohms for 12 seconds 2 channels driven After 20 respectively 12 seconds, the amp shuts down the output, going into protection. There is also a review of the X4 on Production Partner showing interesting result with all channels driven at the same time with the same signal. Ipal module is pretty small. Maximum output power is 4500 watt for 0.3 seconds then it will reduce to 2500 watts for 3 seconds then it limits to 700 watts continuous for 35 seconds then down to 400 watts for long term. K3 at 2.3 ohm load will deliver 3150W per channel for less than 0.4 seconds both channels driven, will then go down to 1450 for a second then slowly down to 500 watts a channel in 1.5 seconds. Why I receive this thread? Because now Powersoft launched a new model, X4L , a forced bridged X8 amp, that has 4 channels with 300 Vpeak and 140Apeak. If 2 X4 channels can deliver that sustained power and program power, then I believe the new X4L will surpass the K20 if asymmetrical loading is used. The idea is that a series loaded 2*21Ipal loaded Skhorn presents a pretty high minimum impedance for a K20 so mostly the output is voltage limited in normal use. Bridging the amp would be too much for the long safe run and very expensive. With this new model, voltage will not be a problem anymore, the maximum average output power is high enough for using 8 21 Ipal on 4 channels and. It can be almost 3 dB louder on peak output than with the K20 but with awesome processing and three phase ready.

I just realized that using this sub with 2 vents covered, I get the same or more output from it than from a JTR 4000ULF from 13.5 hz up. I cannot bring it in my apartment though, it doesn't fit into the elevator and I live at the sixth floor, so there is no way this will work. But it means it's time for a single driver Skram build

What you hear in the video is the noise I get that keeps getting louder with time. It certainly is a suspension and glueing problem. Very similar on the coil and gap but not magnet and coil material. If the coil is copper it means it is thinner for the same Re if the length is the same. This might mean that the gap width could be smaller so even if the amount of magnet is not the same, it might have a similar magnetic field inside. But that's just speculation.

Noise not present Noise present

Yes, I've seen that. The overhang is 21 mm, pole piece is 19 mm so the coil length should be 61 mm, huge. Probably the best 21 pro driver for vented alignment. Also, the BL^2/re product is around 330 and and that double separated suspension should make a very good sound

First of all, wow, you already have that Eminence driver to test? I want a set of 4 but I have to wait for 4 moths at least. Second, that's odd, I have drivers from 4 different manufacturers, but only those from B&C make that noise. I'll post a video on YouTube to for you to hear it and relate with.

Can't wait to see your comparison results, to compare them to mine. I don't have the NLW9601 driver, this should be extremely powerful and efficient

I have a question for you guys about sine sweep measurement pressure level. what does it really show? the SPL value is a peak one? Looking at the measurements for the SKhorn on the SP6000 amp, the CEA2010 values and the maximum swept values are within half a dB. From my tests, the crest factor of the sweep is 6 dB and the level it shows is 3 dB lower than the peak and 3 db over LZF.

The Linea amp is a serious power amplifier. It will sustain 900 watts continuous indefinitely and it will indeed give you over 5000 watts of power into 2 ohms for any useful length of time per channel. It is using IGBT on the output stages and lots of capacitors on the channel and power supply. I managed to explode the final stage of one 44m10 playing my SKhorn XL version using Ipal driver squire fast , but the 20K is way more stable in 2 ohms. The problem was bus pumping from the very high back EMF. I would suggest you to used 4 ohm drivers and use each sub on each channel instead of bridging it . This way you will use all the DSP capability of the amp and you can make more configurations like cardio and end-fire. The difference in stability is not big, a bridged class D amp will be a tad more stable at high reactive loads , but the back EMF will discharge in the endstage instead of the capacitors so for the same output power , more heat will be generated. In my tests ,a pair of 21Ipal in series will have higher minimum impedance but lower average than 2 *4 ohm drivers in parallel in the same enclosure. The average current is also higher because of a higher reactance , higher moving mass . So the stress on the amp is higher but a Linea amp channel will be well suited for it. Regarding limiters, for pro use, I suggest to make use of all of those limiters. The X-max limiter is excellent to be used when you force the system and you hit the peak limiter or maximum voltage quite hard. It will help you with keeping the high pass where it should be no matter what the input stage is doing.

It really depends on the amp, musical content and limiting. Even a pair of 21Ipal in series on a K20 amp channel can get overheated in pro audio use. A lot of heat builds up if the music content is centered around 50hz octave, near an impedance minimum. In this area , not only the driver excursion is minimum, but efficiency is pretty low, current input is very high and no air movement happens in the back chamber. I did an experiment where I put dance,pop, rock, hip-hop, reggae, etc music in a player and did a spectral content average, a long term FFT, with about 10 hours average and the 50 hz band is +7 dB over the rest of the low end spectrum (35-200 hz) , with 30 hz being 13 dB lower and dropping from there quite fast. That type of music sounds absolutely awesome on these subs but they heat up very very fast so the limiting strategy has to be well understood and used with care. On different kind of music, like dubstep and DNB, the spectral content has a peak near 39 hz and 30 hz is -6 dB , a lot of under 30 content is present. This very well received by this sub, a lot of diaphragm excursion is present, the vents are working a lot and air is exchanged , so the drivers can keep on going, way cooler, way louder.