Prev: Question about paralleling inputs
Next: Sony To Close Recording Media Plant
From: Leon on 18 Feb 2010 21:46
On Thu, 18 Feb 2010 17:33:04 -0800 (PST), Krooburg Science <krooburg(a)gmail.com>
wrote:

>On Feb 18, 12:24�pm, L...(a)nospam.com wrote:
>> If I put 2 equal powered speakers (subs) together, I guess I get 3db more SPL
>> plus 3db from coupling effects? Or is it 6?
>
>With a set of subs, doubling power into that set nets a 3dB increase
>in output not taking power compression into account. Doubling boxes
>but with the same power *dissipated* (half the power per box from the
>original number) nets a 3dB increase. Doubling an boxes *and* amp
>power nets a 6dB increase. If you have an amp that's putting out 100
>watts into 8 ohms and you add another identical 8 ohm box, the amp
>power out nearly double from supplying 2x as much current. So
>inherently when you double subs connected to an amp or double up subs
>& add amp, you're essentially doubling power and boxes for a 6dB
>increase in output.
>
>With top (mid/high) boxes, the net increase in SPL for doubled boxes
>in an *array* and doubled power is only ~3dB due to a combination of
>destructive & constructive interference (comb filtering). Putting
>boxes together with a combination of coupling and non-coupling drivers/
>frequencies, that is the result. This is because drivers and boxes
>need to have their distances within a 1/4 of the wavelength of the
>frequencies they're reproducing. After that distance, they start to
>comb. Low frequencies have relatively long wavelengths, but upper mids
>and highs don't. With subs in clusters, the drivers and frequency
>couple very well. Not so with top boxes past the lower frequency
>bandpass. Here's a link to an audio frequency wavelength calculator if
>you feel like plugging in some numbers:
>
>http://www.sengpielaudio.com/calculator-wavelength.htm
>
>> If I add to the bunch, how does it increase?
>
>As stated before, doubling subs and amp power nets a 6dB increase. So
>it's an exponential relationship of box count vs. output:
>
>1 sub = 0dB baseline
>2 subs = +6dB SPL gain over 1 sub
>4 subs = +6dB SPL gain over 2 subs (12dB more than 1)
>8 subs = +6dB SPL gain over 4 subs (18dB more than 1)
>16 subs = +6dB SPL gain over 8 subs (24dB more than 1)
>32 subs = +6dB SPL gain over 16 subs (30dB more than 1)
>and so on....
>
>> Does it matter if I put 4 cabs in a line, or 2 on top of 2? What about 2 in each
>> corner compared to 4 below mid stage? ( I know not to tower them up 1x4!!)
>
>With only 4 small subs, it's not going to make much if any difference.
>You need a pretty long line of boxes to achieve any real pattern
>control. Now the larger the sub array and the arrangement, the more
>difference it's going to make on the pattern control & coverage. The
>longer the array the more pattern control will be exhibited end to end
>due to interference from one end of the array to the other. The
>frequency the pattern control is effect to depends on the array
>length. This is true in both horizontal and vertical arrays. Depending
>on the venue and the array implementation, this can be an asset or a
>detriment to audience coverage.
>
>> I have some cheap subs ($200) that are rated 118db @ 400 watts (their maximum),
>> so if I put 4 together what could the rating go up to?
>
>Take the SPL ratings with grain of salt, especially with cheap boxes.
>There is likely no qualifier of what frequency they're talking about
>and calculated specs, the most common, often don't come close to real
>world performance. As for the answer based on the "calculated" spec, 4
>subs vs. 1 would give you a 12dB increase in level if they're packed
>together in an outdoor free field. If you place them in separate
>areas, they will comb filter and the average level will be less than
>the 12dB calculation. However, in a room things can be much different
>due to boundary reflections. It can increase SPL or decrease it
>depending on room modes.
>
>>
>> I want to be able to match up with 4 - 450 top boxes.
>>
>> Thanks!
>
>K.

Thanks for the long post! I'm going to save it and read it a few times!

From: Leon on 18 Feb 2010 21:47
On Thu, 18 Feb 2010 21:52:52 +0100, Joe Kotroczo <kotroczo(a)mac.com> wrote:

>On 18/02/2010 21:24, in article vl7rn5tc815ldfklkj176lncpolu4fqcm5(a)4ax.com,
>"Leon(a)nospam.com" <Leon(a)nospam.com> wrote:
>
>>
>> If I put 2 equal powered speakers (subs) together, I guess I get 3db more SPL
>> plus 3db from coupling effects? Or is it 6?
>>
>> If I add to the bunch, how does it increase?
>>
>> Does it matter if I put 4 cabs in a line, or 2 on top of 2? What about 2 in
>> each
>> corner compared to 4 below mid stage? ( I know not to tower them up 1x4!!)
>
>The short answer: yes it matters.
>
>And by the way: vertical sub arrays are all the rage lately. ;-)

I didn't know that, never saw it done!

>A nice paper on sub arrays is here:
>http://www.voidaudio.com/pdf/bass%20guide.pdf

Thanks!

From: liquidator on 19 Feb 2010 07:53

"GregS" <zekfrivo(a)zekfrivolous.com> wrote in message
news:hlkba0$du9$2(a)usenet01.srv.cis.pitt.edu...
> In article >
> I was reading posts from some guy, I think in engineering, that the
> maximum
> addition of sound you can get is just over 10 dB. Maybe 12, from acoustic
> gain effect.

Unfortunately, him posting it doesn't make it accurate.

I would like to see where he is coming from, but sounds fishy to me.


From: GregS on 19 Feb 2010 09:24
In article <hlm1ka$c6d$1(a)news.eternal-september.org>, "liquidator" <mikeh(a)mad.scientist.com> wrote:
>
>"GregS" <zekfrivo(a)zekfrivolous.com> wrote in message
>news:hlkba0$du9$2(a)usenet01.srv.cis.pitt.edu...
>> In article >
>> I was reading posts from some guy, I think in engineering, that the
>> maximum
>> addition of sound you can get is just over 10 dB. Maybe 12, from acoustic
>> gain effect.
>
> Unfortunately, him posting it doesn't make it accurate.
>
>I would like to see where he is coming from, but sounds fishy to me.


I don't know if he lerks here, but I found that info I was looking for.
Here is a cut from some recent discussions.

In article <hj4qok$s8...(a)usenet01.srv.cis.pitt.edu>, GregS wrote:
>In <slrnhla62r.opt....(a)manx.misty.com>, d...(a)manx.misty.com (D. K.) wrote:
>>In article <7rkiqeF40...(a)mid.individual.net>, Phil Allison wrote:

>>>"Don Klipstein"
>>>> There is the matter of "half space reference efficiency" of cone
>>>> loudspeakers having some relationship to enclosure volume and low end of
>>>> frequency response.

>>>** Reference efficiency sets a kind of defined upper limit on efficiency.

>> It is not an absolute limit. Efficiency can be increased past that in
>>resonances, by restricting the acoustic environment to a smaller solid
>>angle than half space, and by horn loading.

>> And with an array of drivers, half space reference efficiency gets
>>multiplied by the number of drivers.

>> Half space reference efficiency is 0 dB on a plot of theoretical
>>frequency response using the Thiel Small equivalent circuit. Obviously,
>>if the number of drivers gets this past or anywhere near 100%, then the
>>on-axis sensitivity predicted (which often largely actually results)
>>depends on substantial directivity. At frequencies low enough for little
>>directivity, acoustic loading will depress the efficiency to 25% ballpark
>>or less, translating to 107 dB at1 watt, 1 meter or so in ideal half-space
>>with typical ratio of Znom/Re.

>Are you saying 107 is the maximum with arrays ?

Maximum with hemispheric coverage, unless ratio of Znom/Re is unusual.
At frequencies high enough for the coverage to be less, then the far field
sensitivity of an array can get higher than 107 dB @ 1 watt 1 meter (or 87
dB @ 1 watt, 10 meters - 1 meter may not be in the far field with a large
array).

<SNIP from here>

>greg

--
- Don Klipstein (d...(a)misty.com)
First  |  Prev  | 
Pages: 1 2
Prev: Question about paralleling inputs
Next: Sony To Close Recording Media Plant