what are the drawbacks of compression drivers?

[augerpro]

Because more and more people are using horns with smaller woofers. Unless you do a custom waveguide or SEOS8, this is a problem for some 8" woofers. It becomes even more of a problem using 5" woofers with the small but deep (3-4") horns that are available.

BTW the distinction can also show up in your application when you measure off axis, especially vertically. The lobing will be differnt just above and below the crossover between your case and one where the AC's are aligned.

I really don't understand why I have to keep making this point of general versus special to the smart people here at diya: should I say that just because Newton's equations predict the flight of my arrow just fine, that then Relativity is disproven? Come on.

 

[plasnu]

I personally have never considered phase and delay as different things. They do differ in a mathematical sense since delay is the rate-of-change of phase...

We can hear the difference between time aligned and phase aligned speakers.

It would be ignorable difference for most audiophiles who are so accustomed to ill phased sound due to phantom center of the stereo speakers, and ill designed consumer speakers, but they do sound clearly different especially if you listen to music with one speaker.

I have noticed that audiophiles are far less sensitive to the phase / time error than professional audio people in general.

 

[plasnu]

I'm no saying time aligned speakers are superior to the other speakers, btw. Many people prefer nicely distorted tube amp sound, so why not nicely time / phase distorted sound? Backloaded horn has extremely delayed, but musical bass for some music.

 

[gedlee]

No, this is not correct. For time alignment as is conventionally termed, the phase angles across the crossover summation region just need to overlay each other, no matter their slope.

Nearly flat phase, with little to no group delay, is not a requirement for conventional time alignment woofer to tweeter. It is only a requirement if linear phase is also specified...

You are correct of course - I was assuming something that everyone may not have understood. All constant slopes are equivalent, meaning that flat is the same as straight, it's simply a slightly different definition for what we call the acoustic center. Hence, I take out this constant slope to find the non-minimum phase portion of the phase and this yields a "near flat" phase, as I said. I can make the slope anything that I want by just delaying the impulse response, i.e. redefining where x=0 is, i.e. the "time of flight".

 

[mark100]

You are correct of course - I was assuming something that everyone may not have understood. All constant slopes are equivalent, meaning that flat is the same as straight, it's simply a slightly different definition for what we call the acoustic center. Hence, I take out this constant slope to find the non-minimum phase portion of the phase and this yields a "near flat" phase, as I said. I can make the slope anything that I want by just delaying the impulse response, i.e. redefining where x=0 is, i.e. the "time of flight".

Hi, sorry i couldn't get back to your earlier reply...

Yes, any constant slope phase trace is equivalent when viewed under a linear frequency scale, other than for a delay shift applied to the entire spectrum.
(This is one reason for when I said earlier I like to separate phase from delay 😉

But a constant slope still reflects the special condition of linear phase.

A special condition rarely achieved...
I mean, even benign first order IIR crossovers induce 90 degrees of phase trace rotation and curvature.
If there is any changing group delay, there is a non-constant slope, yes?

Getting these phase trace curvatures to overlay on each other through the entire practical summation region is what I would call good time and phase alignment.
Can be very tricky for higher order crossovers....but if done properly they can be considered time and phase aligned under conventional terminology.

My thoughts respectfully presented 🙂

 

[gedlee]

If there is any changing group delay, there is a non-constant slope, yes?

Yes, that has to be the case.

I can see your point that the phase slopes can be the same (time aligned) but there can still be a constant phase difference, hence the differentiation between phase and delay. If that is your point then I agree. I do design to achieve both while optimizing the polar response. Sometimes a phase difference in the drivers can result in a better polar response. Polarity for example, but there is also some phase changes that result from the crossover details.

Getting these phase trace curvatures to overlay on each other through the entire practical summation region is what I would call good time and phase alignment.
Can be very tricky for higher order crossovers....but if done properly they can be considered time and phase aligned under conventional terminology.

Let's go back to how this all came up. It was stated that horns have a problem because they are not "time-aligned" with the woofer - the driver sits further back than the woofer. I was simply pointing out that this was not an unsolvable problem, to wit my own designs (both passive or active), and that it actually turns out to be quite convenient given the inherent delays in LP filters. I was not "misleading" anyone, but perhaps people may have misunderstood some assumptions that I made.

It is hard to state in a short forum all of one's assumptions. The one that get most misunderstood and I have to clear up all the time, is that I assume we are audiophiles listening in small rooms. Pro sound and large venues are a completely different subject, but for many here, that seems to be their unstated assumption.

 

[mark100]

Thanks. Agreed all around 🙂

 

[whgeiger]

Gradient Index Acoustic Lens

Hi Bill

I did my MS Thesis on acoustic lenses, built several and did a model. There is an AES paper on the same. The real problem with them is that they actually lack effectiveness. A high (enough) index of refraction requires a serious step to the plates and this causes a large reflection from the device itself. Hence it is difficult to meet hopeful expectations without serious trade-offs resulting. An acoustic lense suffers the same problems as an optical one and that is reflection at the interface. In optics we have coatings that work well because the wavelengths are so small. In acoustics no such solution is possible.

There is no way that an acoustic lense could ever replicate the directivity control of a waveguide.

Hi Earl,

Never say never!

Exploration of the subject lens technology holds the promise of yielding a superior dispersive acoustic lens as well as others.

The delayed response is due to my infrequent visits here.

Regards,

Bill