preference for using driver near its distortion limits

[mazurek]

Short version:

Crossed over at 160Hz and 2000Hz at 24dB LR acoustic, which is preferable? I want the midwoofer to have a box no bigger than 0.25 cubic feet, as most space will be taken by the (sub)woofers.

Peerless HDS 6.5" (Ruled out 5.5", not enough SPL at 160Hz)
http://www.htguide.com/forum/showpost.php4?p=235104&postcount=42

High end distortion will be kept just barely flat by the filter.

Scan-speak 8530K (Ruled out 8535, box too big)
http://www.zaphaudio.com/5.5test/

Low end distortion will be kept flat by the filter.

I want my next system to be a three way with high spl, high fidelity. For that reason, I am enforcing 2000 Hz as the lower limit on the tweeter crossover to achieve reasonable dynamic range (Linkwitz' max spl spreadsheet says 112 dB max at that freq). Also, I am keeping the (sub)woofer below 160 Hz, so that it may relieve the midrange, but can safely play loud bass without driving up distortion.

Thanks, sorry about the which driver is better question, but I thought I'd drive up the value a little by adding interesting details.

 

[ScottG]

Short version:

Crossed over at 160Hz and 2000Hz at 24dB LR acoustic, which is preferable? I want the midwoofer to have a box no bigger than 0.25 cubic feet, as most space will be taken by the (sub)woofers.

Peerless HDS 6.5" (Ruled out 5.5", not enough SPL at 160Hz)
http://www.htguide.com/forum/showpost.php4?p=235104&postcount=42

High end distortion will be kept just barely flat by the filter.

Scan-speak 8530K (Ruled out 8535, box too big)
http://www.zaphaudio.com/5.5test/

Low end distortion will be kept flat by the filter.

I want my next system to be a three way with high spl, high fidelity. For that reason, I am enforcing 2000 Hz as the lower limit on the tweeter crossover to achieve reasonable dynamic range (Linkwitz' max spl spreadsheet says 112 dB max at that freq). Also, I am keeping the (sub)woofer below 160 Hz, so that it may relieve the midrange, but can safely play loud bass without driving up distortion.

Thanks, sorry about the which driver is better question, but I thought I'd drive up the value a little by adding interesting details.

Assuming you mean distortion to be non-linear distortion.. You need to ask yourself how much "NL" distortion at what spl.. ANY time you "throw" high spl's into the "mix" your apparent distortion figures at 1 watt/1 meter can be more than a little deceptive to distortion level at say 100 watts or more.

Furthermore even high-order crossovers will typically not help significantly here (..unless we are talking about utilizing a driver within 1 1/2 octaves of its resonance). NL distortion above the resonance at high power levels is largely due to VC heating (and inductance).

This suggests a driver that:
1. Has a large VC - particularly a large diameter VC (in conjunction with#2)
2. Has good venting
3. Has Faraday rings (which help with distortion at low levels and at high levels as a heat-sink)

Hopefully this will help you to make a choice between the two, OR perhaps suggest an alternative.

 

[mazurek]

Thanks for the detailed reply, I believe the scan-speak has a larger voice coil, both should have good venting and faraday rings according to reviews.

I suppose I will focus more on motor construction for spl, and just use the distortion testing to decide crossover limits.

In some ways, I should just jump into pro-audio, then the sensitivity is that much greater, and you don't need tons of power (and associated heating) to get dynamic range. Its tough though because I don't know where all the reviews are for pro drivers.

Here are some pro drivers I looked at briefly, they look like they need more response shaping, and don't like being played below 400 Hz. This would change my design even more.

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=294-652

http://www.partsexpress.com/pe/showdetl.cfm?&DID=7&Partnumber=294-650

 

[ScottG]

I should just jump into pro-audio, then the sensitivity is that much greater, and you don't need tons of power (and associated heating) to get dynamic range. Its tough though because I don't know where all the reviews are for pro drivers.

I was thinking the very same thing.. (though not with regard to sensitivity per se)

Pro manufacturers actually design the drivers for high spl's.. their distortion levels may not be as good at lower levels but they are often better at high power levels.

The real problem here is off-axis response to 2 kHz. Trying to get something without substantial losses off-axis can be difficult, particularly because most pro drivers are at least 8 inches in diameter.

With that in mind BMSpro has a new 5 inch driver that may be worth while:

http://www.bmspro.info/index.php?show=item&usbid=10284&id=5055170

its fs is high, (its obviously a midrange driver), but with an aperiodic enclosure you can keep distortion relativly low near resonance.