Quote:
Originally Posted by
Aaron Smith 
True. Zheka mentioned he wanted to put an LT on the SI's though...I don't have much experience modeling LT's, but the excursion seems to go through the roof when doing one on these drivers. Although that is probably largely mitigated as it somehow didn't register to me that he plans to run the 2 drivers off of one channel of the NU3000DSP, so they would only see 500 watts or so apiece instead of the 1000 I was figuring. The Behringer stuff may not do so well with low impedance loads at low frequencies, but that may be another story...
Also missed that you were keeping the CHT and also didn't realize you had the Epiks...you have the modal thing covered quite well. Has anybody seen my coffee? Perhaps I should keep my 'help' to myself in the future.
I bet in the real world the extra excursion from the LT would probably be a non-issue even at 1000 watts per driver with the headroom provided by all of those drivers.
But hey, I think I got the D2 config right!
Quote:
Originally Posted by
zheka
i do not know what the proper approach is. I am still learning. Should I model on based on the amp peak ratings or RMS?
Quote:
Originally Posted by
zheka
Yeah, once I get second Inuke I will hit the 1kW per driver sweet spot. Perfect.
Quote:
Originally Posted by
zheka
to summarize.
The real world power limitations for the SI drivers in HT context are dictated by excursion. The power to get the driver to bottom out depends on size and type of the enclosure.
In most cases, 1000 W is a safe match. Smaller sealed enclosures can use even more power.
However the thermal limits can be reached with as low as 600 W long term output signal.
fair enough?
Quote:
Originally Posted by
Ricci
Right. 600 w is the recommended power from SI. This driver will crash hard if you give it too much power in the deep bass in a large enclosure. That is a suggested limit based on mechanical excursion related issues when used in a large enclosure. SI has to cover their ass for the inevitable guy that clanks one too hard running it off a 2400w amp in a big box. The coil has some sort of thermal limit too but that stuff is based so much on the material, duration and enclosure alignment that it is hard to quantify with a rating. I could probably purposely cook most drivers with a 500w amp if I wanted too but unless you are going to run the sub flat out with compressed dub or techno for 4 hrs straight thermal handling in a home setting really shouldn't be a problem.
This is exactly why I've advocated controlling excursion with the air spring. You maintain a 2nd order roll off, protect the driver, have less in enclosure costs and construction, more placement options and can safely apply enough power to use both ends of the driver, all for the price of some additional 2HD down low.
With that out of the way (because it's a no brainer), I have to say I think there's a lot of confusion in the CEA 2010 results, IMHO, because Josh inputs 11,000W at 80 Hz and 800W at 20 Hz to get those results. Obviously, if you're using 1KW per driver, you aren't going to see anywhere near the results in your system that you see in the max burst output results from DB.
The CEA 2010 methodology is for a system only, not to see what the max output of a driver is at each of the frequency centers of the 1/3 octave tone bursts with unlimited amplification. IOW, why doesn't Josh make the PB-13 Ultra sub passive and connect his K-10 to the driver for the CEA 2010 tests? Because the results would be irrelevant to anyone buying the sub.
It's my opinion that the same holds especially true for a DIY system.
If 1KW is the reasonable limit for amplification, 1KW should be used to get the max burst results. Posting +10dB higher results is irrelevant data and misleading, especially in the case of a sealed L/T system, where the opposite of that methodology is the reality. Since there is no such thing as a frequency specific, variable output voltage amplifier and, with an L/T you use less power up top and maximum power at the low end, not the other way around, there's no way you'll see 125dB at 80 Hz and 106dB at 20 Hz in the finished system.
1KW with an L/T of 6dB boost means you can only run the driver at 250W, keeping the rest of the reserve power for the <30 Hz transients. If you're going to design your sealed, L/T system based on WinISD, use +6dB of L/T boost and input 250W. Adjust enclosure size to that metric (because it is a fixed piece of the puzzle) regarding excursion and stop your model graph at 10 Hz because everything below that in the model is incorrect information.
The idea offered by many folks that somehow the 1KW is available for the top end as 'headroom' is errant, like saying black is white. The only way you can use the extra power above the L/T is to change the frequency response of the system. There are 2 ways to do that; 1) aggressive limiting, which changes the FR infinitely on-the-fly or 2) HPFs, which change the FR predictably for the desired output level up top.
Once you've gotten past that, take your model result numbers and do the math for desired output by multiplying the system times 'X' multiples.
A 4 X SI 18 sealed, L/T system with 4KW is a perfectly reasonable system that should give excellent results. 4 drivers ups the sensitivity to 95dB making the 1KW above 30 Hz formidable (within 3dB of Josh's single driver results) and the 4KW for <30 Hz gives a safe delivery of the 20+ available liters of displacement.
If you want to utilize the +6dB above 30 Hz, apply a HPF at around 20 Hz and there ya go, giving up some flat in-room extension for higher playback level.
Missed that silly drop-down box...it should be displayed better, but I can see why they don't want it easily visible. 
