Flex Your PCD Mettle:

[augerpro]

Re: Flex Your PCD Mettle:

Too bad the midrange dip is still there. John is saying it is a measurement artifact but it shows up in the impedance too...Maybe it is a back side chamfering issue? Wonder if those show up in impedance though?

[Pete Schumacher]

Re: Flex Your PCD Mettle:

It's still not listed in the "store" so purchase would be difficult, unless you call them directly.

The "dip" in response may not be that big a deal depending on where it's crossed to the woofer, baffle shape, and overall system sensitivity.

[Zilch]

Re: Flex Your PCD Mettle:

Also tried DE10 on PH715 with its asymmetric dispersion pattern, of which neither PE nor Pyle seems to be aware, or they're not telling anyone, at least, even though its obvious just by looking at it. I'll load them into something else to see if they'll do their image stabilization "trick."

Toggle between inboard and outboard below to see the difference:

[Skywave-Rider]

Re: Flex Your PCD Mettle:

Also tried DE10 on PH715 with its asymmetric dispersion pattern, of which neither PE nor Pyle seems to be aware, or they're not telling anyone, at least, even though its obvious just by looking at it. ...

:p

[Zilch]

Re: Flex Your PCD Mettle:

Installed:




Crossover:




Horizontal Polar Response, 0° - 45° in 5° increments:




Horizontal Polar Waterfall:




Horizontal Polar Map:




It gets where we want to be, but the directivity issue remains.

I'm going after that, now.

[Maybe.... :rolleyes: ]

[Skywave-Rider]

Re: Flex Your PCD Mettle:

Zilch,
Did you build out the waveguide forward to help clear the baffle recess? I see the woofer flange under the wg. And I think I see gasket material under there, but can't tell how far forward that is.

[Zilch]

Re: Flex Your PCD Mettle:

Despite being recessed, the woofer flange sticks forward of the baffle ~1/8", so I had to make up the difference with thick foam gasket tape.

Today, I'm using power tools (unsupervised ;) ), and barring incident(s), we'll have the answer here soon. Wavelets indicate I'm on track; if so, I'll post 'em.

[If not, it is what it is.... :p ]

[Skywave-Rider]

Re: Flex Your PCD Mettle:

What it is is already not too bad, I bet.

[Zilch]

Re: Flex Your PCD Mettle:

What it is is already not too bad, I bet.

Yes to that -- I moved the baffle forward 3/4"; only a 1/4" recess remains:




Horizontal Polar Response:




Horizontal Polar Waterfall:




Horizontal Polar Map:




Conventional Polar Plot:




I've made them toggleable for direct comparison to those in Post #350.

It cleaned up considerably, and some of what remains is minimum phase (@ 2 & 3 kHz), now correctable via the crossover filters.

[The green tag is the price: $7 the pair, marked down from $12. :D ]

[Skywave-Rider]

Re: Flex Your PCD Mettle:

It cleaned up considerably, and some of what remains is minimum phase (@ 2 & 3 kHz), now correctable via the crossover filters.

[The green tag is the price: $7 the pair, marked down from $12. :D ]

Once you do that will you post verticals?

[Zilch]

Re: Flex Your PCD Mettle:

Once you do that will you post verticals?

Yes, that's a routine part of the process, now. Even on this preliminary basis, the forward vertical lobe appears to be quite tall at this comparatively low crossover frequency and close (~9") center-to-center distance between woofer and waveguide. I want to take a closer look at the horizontal polars of both individually for some quantitative guidance as to where we might optimally set the crossover frequency for this combination.

For now, here's my theory regarding these findings: while the experts (Linkwitz, Elliott, Zaph, and others) tell us that, in theory, the controlled directivity of waveguides obviates conventional concerns regarding edge diffraction and overhang reflection effects, this is only true in the response region where they have full pattern control. If we use them at lower frequencies in the transition zone where they are losing control, the pattern widens, and they illuminate the cabinet edges, with concomitant consequences. Even where they DO have control, that is not perfect, i.e., the response is not fully attenuated off-axis, and in detail, measurable effects occur there, as well, as may be seen by toggling* between the response curves posted here.

Conversely, as the woofer's directivity narrows above a frequency related to its piston diameter, if that is occurring below the crossover frequency, or in the crossover region, its interaction with the cabinet edge is unstable, as well. Bottom line: we cannot mitigate design issues related to directivity but by dealing with them directly; the response anomalies they generate are not amenable to correction via the crossover filters alone, alas.... :(

*Toggle: open these two links and click back and forth between them:

Before: http://techtalk.parts-express.com/at...3&d=1275942236

After: http://techtalk.parts-express.com/at...4&d=1275976679

[Pete Schumacher]

Re: Flex Your PCD Mettle:

Bottom line: we cannot mitigate design issues related to directivity but by dealing with them directly; they are not amenable to correction via the crossover filters alone.... :(

I see no reason for a frown. That is excellent information to know, and knowing it, can now be accounted for.

A clean baffle is a good baffle. Roundovers aren't just nice to look at, they remain functional and a useful tool when creating a completed custom product.

[Dennis H]

Re: Flex Your PCD Mettle:

Zilch, with your vintage cabinets, I'd think some absorbers (wool felt, open-cell foam, etc) on the baffle would help with diffraction effects. Your round pipe insulation didn't help much as I recall but that's closed cell so it acts more like a round reflector than an absorber. Dlr has some nice measurements of the beneficial effects of 1/2" felt on a dome tweeter. You'd probably want something at least as thick as the sidewall overhang.

http://www.speakerdesign.net/felt_amelioration/feltssenter.html

[Skywave-Rider]

Re: Flex Your PCD Mettle:

....Bottom line: we cannot mitigate design issues related to directivity but by dealing with them directly; the response anomalies they generate are not amenable to correction via the crossover filters alone, alas.... :(....

If the WG is losing pattern control at 2K in the horizontal (leading to diffraction effects there as well), and you can't make the woofer play higher due to FR and directivity issues (50º at 2k for an 8"cone or thereabouts,) is there anything to be done?

Don't say "3-way...."

[Zilch]

Re: Flex Your PCD Mettle:

Hmmm. I'm checking the charts here, and it looks like an 8" woofer (6" piston diameter) should have a 90° beamwidth (DI of 9, Q = 8) at ~one wavelength, i.e., 2.260 kHz, 10" (8" piston) = 1.695 kHz, 12" (10" piston) = 1.365 kHz, and 15" (13" piston) = 1.043 kHz.

The answer to your larger question is that there is considerable leeway in designing for a smooth transition through the crossover region horizontally. In the vertical, the nulls enter the picture as major players.

I may have the measurement data later today for this particular woofer in this "revised" cab....

Edit: Here is RobH's software generated polar plot for the 12" JBL 2206H:




And the 10" JBL 2123H: