Offset drivers.

[pre65]

I was looking at Google for "offset drivers" and came across this web site. Not really what I was looking for but interesting none the less.

They are suggesting the voice coils should be in line.

Any comments ? (open to ALL members )


http://stonessoundstudio.com.au/stone/a ... offset.htm

[Scottmoose]

Nothing startling here; I haven't had time to read all the page linked, but what it's referring to is an intrinsic part of speaker (and crossover) design. That's assuming you have a target FR and phase alignment in mind.

Sticking with that side of things, essentially we're talking about time-alignment of the drivers. To take the simplest / most common case as an example, i.e. a 2-way with a dome tweeter and a cone midbass, if they are mounted on the same flat baffle the acoustic centre of the tweeter in the front - rear plane (sometimes called the z axis) is further forward than that of the midbass due to its smaller size and the differences in physical construction. How much varies depending on the size of the drivers. Assuming the listener is aligned on the axis of the tweeter, then you have a time delay in the woofer's output relative to that of the tweeter due to its acoustic centre being further from the listener. The vertical and horizontal distances between the drivers also have influence here. In any event, it's an issue that needs addressing, since if the phasing is off you can get large variations in the response. There are a number of things you can do to align the acoustic centres, or compensate for an offset:

1/ Move the listening axis so the tweeter is further from the listener (i.e. raise it or lower it, the latter being done most easily by inverting the speaker)
2/ Use a sloping baffle to physically move the tweeter back & align the acoustic centres of the drivers
3/ Use a stepped baffle for the same purpose
4/ Use a delay network on the tweeter
5/ Use asymmetric crossover slopes
6/ Some combination of the above

Sloped / stepped baffles have a following, but they only tend to work optimally when the listener is directly on-axis with the speaker. You also have to contend with off-axis tweeter performance with sloped baffles (since it's firing at an upward angle) and potentially problematic diffraction effects with stepped types, which is why some also feature large quantities of foam or other damping materials around the multiple sharp edges. Delay networks can also work well, but assuming a passive filter, the price is complexity. The most common solution as far as passive crossovers are concerned is to use asymmetrical acoustical and / or electrical slopes to provide closer phase-tracking between the drivers (different orders have different amounts of phase shift so you can work with / use that). Whether that is superior or inferior to alternatives such as physically aligning the acoustic-centres in the front-rear plane is one of those debatable points without a definitive answer, since there are often a number of other variables involved that you can't separate out, and it depends on the design goals for the speaker. All very interesting if you like this sort of thing (I do, but I'm a lost cause).

[IslandPink]

I was going to comment but Scott has covered it much better than me.