Efficency Confussion

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I apologise in advance as this may be a common question, but I have done a search both here and on google and some of the answer I have found contradict others and I'm quite confussed!

I have 2 mid/bass drivers rated at 88dB efficency and I'm running them parallel in MTM configuration with a 88dB efficency tweeter. What is the efficency of the mid/bass drivers?

Following on from this what, if any, padding will be required to the mid/bass drivers?

As I've said, I have searched, but there is so much contradicting evidence I'm not sure which to follow! :xeye: :bawling: :xeye:
If you run the 2 mid woofers in series you should end up with the same SPL and no padding would be required.

Running them parallel will boost as much as 6dB and would be roughly 94dB. I don't think it's a good idea to pad the mid woofers.... seen a post around here somewhere on that.

Well the standard metric is that so long as the frequency is low enough, and the drivers are mounted close enough, you get 3dB extra efficiency due to increased radiator area.

However when you consider:
1. baffle step
2. these are mid / bass
3. they are mounted MTM

The only way to get a decent answer to your question is to measure in box.
The speaker is being partially built into a wall, so will effectivly be infinite baffle.

The mid/bass drivers wil be used from 350Hz - 1950Hz in part of a 3way WMTMW system.

So if I follow rabbitz advice, and wire them in series, they would effectivly stay at the same efficiency?

Pinkmouse - Yes I do intend to measure them in box once their complete. But I would like to get them as close as possible, then have a small amount of tweeking once I get all the relevent equipment together! (The woman indoors probably won't be too impressed with her living area filled with all my electrical and audio gubbins!)
Mutley666 said:
So if I follow rabbitz advice, and wire them in series, they would effectivly stay at the same efficiency?

Well depends on if you're talking efficiency in terms of 1watt/1m or 2.83V/1m .....

1watt / 1m will likely hover somewhere around the original 1watt / 1m figure.

2.83V / 1m will likely hover approx. 3dB below the original 2.83V / 1m figure.


Joined 2003

At the 1950Hz XO point the midbasses would have to be < 2.2" c-t-c apart to get any useful coupling, so matching the tweeter's voltage sensitivity to a single midbass is what's needed if they are wired in parallel, or 91dB if both the midbass and tweeter are nominally 8ohms.

Series wiring isn't a good plan since there will be no averaging of specs, causing one driver to do most of the work if not very closely matched, and Le will sum instead of halving, so impedance compensation will probably be required.

Also, since these are in an IB, you'll have to high pass the midbasses much higher than where the woofers are XO'd or you'll have a hump in the response in the BW where they have good acoustic coupling.

I was under the impression that the calculation for distance between the drivers vs frequency was; 13560/c-t-c distance in inches = frequency.

13560/7" = 1937Hz

Or have I got that wrong?

The drivers I have as follows;

2x 6.5" Woofers, 86dB sensitivity, 8Ohm (above and below the MTM)
2x 4" Mid/Bass 88dB sensitivity, 8Ohm (MTM with the tweeter)
1x Tweeter 88dB sensitivity, 4Ohm

I had planned to wire the 2 woofers in series and the 2 mid/bass drivers in series and L-pad the mids and tweeter to match the woofers.
Would this work? Or do I need to rethink this?

Visualize it this way, as a poster on another forum likes to say, 'sound is round', so draw two circles on a piece of paper so that they just touch. This is the point where there's no longer any acoustic coupling until you're far enough away that they physically appear as a summed point source to your ears. As the 'balls' increase in size there's more overlap (percentage of radiation impedance gain) until at some point they overlap enough to increase to the theoretical limit of ~ +3.01dB.

Their circumference = 1WL, so assuming the SoS is 13560"/sec and the circumference is 12", then 13560/12 = 1130Hz. The distance between their centers will be = to their diameter, or C/pi, ergo the 'HF' frequency where gain begins is 13560/pi/c-t-c spacing, or ~360Hz.

Theoretically you want the c-t-c spacing to be at least whatever the XO's -12dB point is to ensure that the two sources appear as a single one, but obviously this is phyically impractical beyond a few hundred Hz so the search was on to find the average 'audiophile's' audible comb filtering distortion tolerance in the human speech BW, which based on the popularity of MTMs and current line array design theory apparently is 1WL of the XO point.

If you want to use this as your driver spacing criteria, that's fine, whatever 'floats yer boat' ;), but for calcing matching sensitivities, etc., you want to use reasonably technically correct parameters.


There is a big difference between sensitivity and efficiency. By way of example, both the Seas M15CH001 and Scan-speak 15M4531K have a sensitivity of 90dB/2.83V. But the SS will draw twice the current and therefore require twice the power for the same acoustical output. Therefore is has less efficiency than the Seas.

Most people here refer to the XXdB rating of drivers as their voltage sensitivity. It's easy to make a 92dB/2.83V system. Much harder to make a 92dB/w system.


This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.