reasons for Linkwitz' symmetrical driver placement

[thadman]

What are the reasons for Linkwitz' habitual use of symmetrical driver placement on dipoles? He uses this design in both his Orion and Phoenix designs.

[Davey]

What kind of disadvantage do you think is created with symmetrical placement?

Davey.

[454Casull]

Higher diffraction effect.

[dlr]

Originally posted by 454Casull
Higher diffraction effects.

Dipoles won't exhibit diffraction in the typical sense. The rear wave interacts with the front wave. The tweeter's likely will show some diffraction effects, however, since they are usually not a dipole when used as Linkwitz does. You'd have to have a tweeter with inverted polarity, equal amplitude and positioned directly behind the front tweeter to create a dipole. A large baffle as is typical of a dipole and typical crossover point also helps minimize the diffraction, putting it into the stop-band. Narrow dipoles with dome tweeters will have some diffraction. But Linkwitz has on-axis response down in his list of importance.

But more importantly, the knee of a dipole is based upon the distance to the edge. Positioning a dipole woofer or midwoofer offset, meaning closer to one edge, alters the dipole effect. I believe that it would not result in a standard 6db slope around the knee. It would be a complicating factor.

dlr

[thadman]

So symmetrical positioning for a dipole is where it is at? What is the most desirable shape for the edge, a teardrop (rear edge of plane wing) or a curved shape (complete roundover)? What are the benefits of wider vs narrow baffle other than the obvious (wider=higher efficiency, but less of sidewave absorbed, narrow=lower efficiency, but higher percentage of sidewave absorbed)?

[dlr]

Originally posted by thadman
So symmetrical positioning for a dipole is where it is at? What is the most desirable shape for the edge, a teardrop (rear edge of plane wing) or a curved shape (complete roundover)? What are the benefits of wider vs narrow baffle other than the obvious (wider=higher efficiency, but less of sidewave absorbed, narrow=lower efficiency, but higher percentage of sidewave absorbed)?

Go see the site link. It will answer more than I can by a long shot.

Music and Design

dlr

[Davey]

As is usual with most speaker design topics....SL has already addressed this on his website:

http://www.linkwitzlab.com/diffraction.htm

Also, I think you'd find Linkwitz recommending symmetrical placement of drivers in non-dipole conventional box systems as well. (If he had any interest in those.)


A typical box design measured on-axis will measure a somewhat "smoothed" response when offsetting drivers slightly. However, any off-axis measurements will also show a smoother or less ripply response even when the drivers are symmetrically placed. I know this practice of offsetting drivers is viewed as advantageous by many speaker builders, but it really isn't necessary.

In a dipole system the baffle width and shape is important to achieve a nicely shaped polar response (un-corrected) that can be equalized for dipole correction and still maintain an even power response. Trial and error experimentation with baffle shapes is necessary in all cases.

Cheers,

Davey.

[john k...]

Originally posted by Davey
As is usual with most speaker design topics....SL has already addressed this on his website:

http://www.linkwitzlab.com/diffraction.htm

Also, I think you'd find Linkwitz recommending symmetrical placement of drivers in non-dipole conventional box systems as well. (If he had any interest in those.)


A typical box design measured on-axis will measure a somewhat "smoothed" response when offsetting drivers slightly. However, any off-axis measurements will also show a smoother or less ripply response even when the drivers are symmetrically placed. I know this practice of offsetting drivers is viewed as advantageous by many speaker builders, but it really isn't necessary.

In a dipole system the baffle width and shape is important to achieve a nicely shaped polar response (un-corrected) that can be equalized for dipole correction and still maintain an even power response. Trial and error experimentation with baffle shapes is necessary in all cases.

Cheers,

Davey.

I have to pretty much agree. I also recommend symmetric driver placement in my dipole design guide. Offsetting the driver can make the on axis response look better, particularly above the dipole peak. You may see offset midrange drivers on dipole system using overly wide baffles. This is because with an overly wide baffle the driver may still be fairly non directional at frequencies above the dipole peak and the nulls in the response above that peak may be rather severe. Offsetting the driver will tend to smooth the null but all it really does is move the problem off axis. At lower frequencies there are implications with regard to asymmetry of the polar response. This isn't to say the asymmetric baffles can be designed for good performance, but it is a much more complex design problem.

[dlr]

Originally posted by Davey
[B]As is usual with most speaker design topics....SL has already addressed this on his website:

http://www.linkwitzlab.com/diffraction.htm

Also, I think you'd find Linkwitz recommending symmetrical placement of drivers in non-dipole conventional box systems as well. (If he had any interest in those.)

Reading through that link (again) does not make it at all clear that he would recommend symmetric placement for a non-dipole:

While I try to minimize visible diffraction ripples in the frequency response for good measure...

If he truly would not recommend it, he would not be concerned about "good measure" the way I read his comments. He definitely does not dismiss it out-of-hand, rather he is cautious to consider it to some degree.

Originally posted by Davey A typical box design measured on-axis will measure a somewhat "smoothed" response when offsetting drivers slightly. However, any off-axis measurements will also show a smoother or less ripply response even when the drivers are symmetrically placed. I know this practice of offsetting drivers is viewed as advantageous by many speaker builders, but it really isn't necessary.

Very little in speakers is absolutely necessary. Most of it is synergy.

The off-axis smoothing is also not a constantly diminishing phenomenon with increasing angle, for symmetric and non-symmetric. The variation is generally greater with a symmetric placement in the shallower angles.

As to necessary, it also depends on ones goal. If, as I always do, a maximum of diffraction control is desired through the use of a felt application and probably when using roundovers, the non-symmetric placement is an absolute necessity if minimization is the goal. This comes from a multitude of measurements of my own.

As SL states:

It has been my experience that the on-axis and off-axis frequency response of a given driver and baffle combination must be measured to assess overall uniformity of response, and that computer models give pessimistic predictions, because the underlying assumptions do not sufficiently describe the real acoustic behavior of drivers and of cabinet shapes.

My data is not from "pessimistic predictions" (guess he hasn't worked much with the BDS), rather it's from first-hand experience, that is, measurements. I believe it does have an audible impact. Subtle, yes, but audible nevertheless.

In a dipole system the baffle width and shape is important to achieve a nicely shaped polar response (un-corrected) that can be equalized for dipole correction and still maintain an even power response. Trial and error experimentation with baffle shapes is necessary in all cases.

Yes, and he goes on to say:

While the "baffle step" cannot be avoided, the additional higher frequency ripples can be easily reduced to a magnitude that is much smaller than the first arriving direct sound, by simply optimizing the proportions of a rectangular baffle

Still not dismissing the non-step related baffle diffraction.

Even his last statement that is arguable correct is not a dismissal. The same can be said of the absoluteness of many things in audio, transient-perfect designs, importance of on-axis response, active vs. passive, group delay, crossover phase shift and others that are all considered gospel by one group or another. With those and any number of others and leaving out the last word on his page, one could say:

Much is hypothesized, little is proven and much is overrated when it comes to_____

No one would have a clue as to hie reference in reading this.

Dave

[Davey]

Dave,

I'm not sure what your point is. Or are you trying to find some conflicting statements within the vast amount of data on SL's website? You probably will find many items. He's mentioned to us on the OUG that his understanding of various things has evolved over the years and some of what's posted might be no longer pertinent, and that he'll update the website when he gets a chance.

I've talked to SL privately on a number of occasions and he's mentioned to me that he thinks symmetrical placement of drivers is fine even for conventional systems. That's what he's told me.....you can believe it or not.....or nit-pick his statements and website to death if you like.

Cheers,

Davey.