[JoshK]

Can anyone tell me what the relationship with multiple drivers and efficiency is?  Or point me to a reference where I can look it up?  For example, if I have 8 x 86R drivers, what would my resulting efficieny of the array be?  Is it dependent on whether the drivers are wire in parrallel or series (or combo therein)?

[Rick Craig]

Can anyone tell me what the relationship with multiple drivers and efficiency is?  Or point me to a reference where I can look it up?  For example, if I have 8 x 86R drivers, what would my resulting efficieny of the array be?  Is it dependent on whether the drivers are wire in parrallel or series (or combo therein)?

It depends on how they are wired and how the crossover is designed. If you let me know which drivers and what your target is for sensitivity I can help you.

[JoshK]

I was more interested in learning the math behind it then having any particular idea in mind.  My background is in math, I just have never seen such a reference.

[Jason1]

Each time you double the number of drivers the sensitivty increases by 3db

8 drivers would be a 9db increase over a single driver.

[WerTicus]

yeah that makes sense but i wonder if its accurate?

[JoshK]

Thanks Jason,

Even if its not 100% accurate at least it gives me a feel for the relationship. I understood the impedance issues, as a application of Ohm's law, I just didn't know a law or a rule for sensitivity.  I have just been reading up for fun and playing with ideas in my head but wanted to make them realistic.

[JohnR]

Strictly speaking, the *efficiency* goes up 3dB. In general, it appears that "sensitivity" refers to the output for a given voltage input rather than power input.

[JoshK]

Thanks John, that actually makes sense too.

[WerTicus]

so is there a way of calculating that john? or do you just have to tune line arrays in the box? (well you WOULD anyways but you know...)

[Rick Craig]

The following link has a program called "BoxyCad" that you can download to calculate the output for multiple drivers.

http://www.audiodiycentral.com/

[JoshK]

Thanks Rick.

[JoshK]

Spse I was creating a line array using 10 Seas midrange/midbass drivers and 8 Silver flute ribbon tweets each side.  Specifically these drivers:

Seas L12RCY/P  (H1207)  8 Ohm, 86db, 6L VAS, 48 Fs (rec'd range 55hz - 3.5Khz)
Silver Flute YAG20  8 Ohm, 90db, 3k - 20khz

Wiring the ribbons alternating Serials and Parallel to acheive 8 Ohm would yield 99db efficiency?  Likewise, would wiring two groups of 2, and two groups of 3 in alternative series/par yielding 10 ohms produce an efficiency of approximately 96db?  

I came up with this guess using the following equation:

Individual Drivers Sensitivity + 3 * Log(base 2)[ # of drivers in array]

which would be the equation if it held true that each doubling of # of drivers produced 3db more efficiency.  

Second question, if I did this correctly, is 4 db sensitivity differential too much to pad in a passive xo?  

What else do I need to consider for this matching these two drivers?  Assuming of course that I could find a suitable xo slope in the 3k - 3.5k region that would work with the both.

[_scotty_]

JoshK,you will need at least a 24db/oct slope xover to minimize comb filter effects in the horizontal plane. Here is a link to designing near field line arrays. Check out the pdf doc by Dr. Jim Griffin on the left side of the page.   http://www.audiodiycentral.com/
  Also check out what Linkwitz has to say about loudspeaker design and line arrays  http://www.linkwitzlab.com/frontiers.htm#E
     Good luck with your project,Scotty

[Rick Craig]

Spse I was creating a line array using 10 Seas midrange/midbass drivers and 8 Silver flute ribbon tweets each side.  Specifically these drivers:

Seas L12RCY/P  (H1207)  8 Ohm, 86db, 6L VAS, 48 Fs (rec'd range 55hz - 3.5Khz)
Silver Flute YAG20  8 Ohm, 90db, 3k - 20khz

Wiring the ribbons alternating Serials and Parallel to acheive 8 Ohm would yield 99db efficiency?  Likewise, would wiring two groups of 2, and two groups of 3 in alternative series/par yielding 10 ohms produce an efficiency of approximat ...

The ribbons don't couple the same as the point source woofers do. This is due to their narrow vertical directivity and they won't give you any SPL gain unless you configure them for a lower impedance.

[JoshK]

The ribbons don't couple the same as the point source woofers do. This is due to their narrow vertical directivity and they won't give you any SPL gain unless you configure them for a lower impedance.

Thanks, that is the sort of stuff I am looking for.  If I actually decide to build a line array I will likely go with a pre tested kit like yours, I am mostly just learning for fun.

[nathanm]

Sorry, I don't mean to sidetrack the thread but here's a something I always wondered about:

Are series-connected speakers in phase with each other or only if they're run in parallel hookup?

[_scotty_]

Yes, they are in phase.

[JoshK]

JoshK,you will need at least a 24db/oct slope xover to minimize comb filter effects in the horizontal plane. Here is a link to designing near field line arrays. Check out the pdf doc by Dr. Jim Griffin on the left side of the page.   http://www.audiodiycentral.com/
  Also check out what Linkwitz has to say about loudspeaker design and line arrays  http://www.linkwitzlab.com/frontiers.htm#E
     Good luck with your project,Scotty

Scotty,

Thanks for the fantastic reference!  Dr. Griffin's white paper is exactly what I was looking for!  Gives me as much understanding as I will need for my purposes.